/home/precedings/apps/precedings/shared/storage/document/2007/1272/npre20071272-1

Age and origin of enigmatic megaherbs from the

subantarctic islands

Steven J. Wagstaff

, Ilse Breitwieser

, Christopher Quinn

Motomi Ito

Allan Herbarium, Landcare Research, Lincoln 7640, New Zealand

. Royal Botanic

Gardens, Mrs Macquaries Rd, Sydney NSW 2000, Australia

. Department of Systems

Sciences (Biology), University of Tokyo, Komaba, Meguro-ku, Tokyo 153-8902, Japan

Biogeographic relationships in the southern hemisphere have puzzled biologists for

the last two centuries

1,2

. Once joined to form the supercontinent Gondwana,

Africa, Antarctica, Australia, New Zealand and South America are widely

separated by the Pacific and Indian oceans. Sir Joseph Hooker was the first to

suggest that Antarctica served as a corridor for plant migration not unlike the

land-bridges in the northern hemisphere

7,8,9,10

. While the Antarctic flora was

largely erased by glaciation during the Pleistocene, at least some of these Antarctic

plant communities found refuge on the subantarctic islands. Here we provide

support for the hypothesis that giant herbs persisted in the subantactic islands

prior to the onset of Pleistocene glaciation, then dispersed northward in response

to glacial advance. Our findings provide further evidence that Antarctica has

played a pivotal role in shaping southern hemisphere biogeography.

During the early 1800s voyages of discovery brought many explorers to the southern

oceans. Much of the early botanical work was completed by European taxonomists

whose interpretations were influenced by their familiarity with the Old World flora

1,2

They thought southern hemisphere plants had diversified from northern hemisphere

ancestors, and this has likely been the case in large genera such as Veronica,

Ranunculus and Epilobium

3,4

. However, some plant groups such as Nothofagus

and

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

Agathis

have evolved in the southern hemisphere then dispersed in a northerly

direction.

The floras of North America and Eurasia were united by land bridges that existed

during much of the Tertiary

7,8

. However, the southern hemisphere continents have been

isolated by the southern oceans since the break-up of Gondwana at the end of the

Cretaceous. During much of the Tertiary the now widely separated southern continents

South America, Africa, Australia and New Zealand faced a continuous Antarctic

coastline that was clothed in diverse forest vegetation. The abundance of fossils in

Antarctica suggests that it may have served as a corridor for plant migration until late in

the Tertiary

9,10

. The environmental conditions were very different at this time. The

climate was much warmer, and because of their positions at high latitudes, Antarctic

ecosystems experienced long periods of complete darkness followed by short growing

seasons with almost continuous but low-angle daylight. These environmental conditions

supported diverse open woodland vegetation

. The Antarctic ice cap grew from the

Oligocene (35 Myr) culminating in several episodes of glaciation during the

Pleistocene. During major glaciations the polar ice sheets spread considerably and

temperature, marine and vegetation zones were compressed towards the equator

During the glacial maxima the Antarctic vegetation was almost completely eliminated,

though surely a few remnants of this once lush Antarctica found refuge to the north in

isolated circum-Antarctic island archipelagos.

The massive Antarctic ice sheets that persist to the present day have imposed a

limit on our ability to interpret past environments before their formation and during

their repeated waxing and waning

. Given the increasing interest in the role the

subantarctic islands have played in southern hemisphere biogeography, there is a

growing need for phylogenetic studies of taxa endemic to these islands. Such analyses

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

can partly overcome the problems of interpretation that arise from extinctions associated

with past glaciation events

12,13

The subantarctic islands fall into two groups, older islands with a continental

origin whose biota is shared by nearby land masses and younger volcanic islands, where

the history of the biota remains enigmatic

; the New Zealand subantarctic islands fall

within this latter category. The New Zealand subantarctic islands were formed by

volcanic activity during the Tertiary, but interbedded limestone and basement rocks are

found in restricted areas

. The islands now lie in a region of cool surface waters, and

the weather is dominated by strong westerly winds, cool temperatures, persistent

cloudiness and high rainfall. In the Tertiary the winds and circum-polar current were not

as strong and dispersal by windblown propagules north to the islands was more easily

accomplished

. Nutrient inputs are mostly marine in origin from nesting seabirds,

penguins and seals and this coastal eutrophication has occurred at least for the past

10,000 years

. Such nutrients, long summer day-lengths, and adequate water lead to

high primary production in the subantarctic islands. In association with anaerobic

conditions this has created extensive peat deposits. The New Zealand subantarctic

islands were granted World Heritage status by UNESCO in 1998 because they

encapsulate a diverse set of natural heritage values, many of which are unique

These natural heritage values were already recognized by early naturalists. Sir

Joseph Hooker was struck by the unique vegetation particularly the extensive meadows

dominated by lush perennial herbs with large leaves. He wrote in his notes on the Ross

Expedition

: "The most extraordinary of the megaherbs is the Pleurophyllum meadow,

a community dominated by the large-leafed herbaceous composite, producing a floral

display second to none outside the tropics." Megaherbs have adapted not only to the

harsh climate and short growing season, but also to the highly acidic, peaty soil. The

megaherb growth habit has evolved independently in diverse species from a range of

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

different genera and families such as Gunnera, Plantago, Stilbocarpa and the composite

Pleurophyllum that was admired by Hooker (see above).

We tested the hypothesis that the endemic genus Pleurophyllum is a

geographically isolated relict of a once lush Antarctic flora that then dispersed

northward in response to glacial advance and survived the LGM in refugia on the

subantarctic islands of New Zealand. Pleurophyllum is a genus of three species,

restricted to Auckland, Campbell, Antipodes and Macquarie islands. In some areas on

Campbell Island, P. speciosum with its large coriaceous leaves is the dominant life

form. The other two species (P. criniferum and P. hookeri) often grow sympatrically.

Pleurophyllum speciosum forms an enormous rosette, up to 1.2 m across, of huge, broad

pleated leaves. Its spectacular capitula, about 6 cm in diameter (including the long ray-

florets), are arranged in subcorymbose racemes, about 60 cm tall. The flowers (Fig. 1)

show considerable colour variation, probably due to the mineral content of the soil or

the degree of acidity. The colouration seems to vary from bluish pink to pink and even

white. The other two related species, P. hookeri and P. criniferum, with scapes up to 60

cm and 2 m respectively, have silvery lance-shaped leaves and maroon florets. Their

ray-florets are very inconspicuous. The two may be distinguished by the narrow smooth

silvery leaves in P. hookeri being pointed and erect, compared with the broad rounded

heavily veined greenish leaves of P. criniferum that form a rosette close to the ground.

Previous researchers have suggested that Damnamenia and Olearia are closely

related to Pleurophyllum

18,19

. The monotypic genus Damnamenia, a segregate from

Celmisia

, is also endemic to Campbell and Auckland islands. In alpine scree, immense

drifts of D. vernicosa carpet the ground. It grows to 35 cm tall and the capitula are 5 cm

wide. Olearia is Australasia's largest genus of Compositae. Drury

has already pointed

out that one group of Olearia, the macrocephalous olearias [O. chathamica, O. colensoi,

O. lyallii, O. oporina (incl. O. angustifolia), and O. semidentata] are more similar to

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

Pleurophyllum than to other olearias. All members of the group are characterized by

terminal, solitary or racemose capitula, long villose achenes, subduplex subrufescent

pappus, and large, flat woolly haired leaves. Their floret colour, leaf venation and leaf

hair-type also distinguishes them from the other olearias. With the exception of O.

colensoi, all of them occur in the southern part of New Zealand (Fig. 2): O. lyallii is

restricted to the Snares and Auckland islands, O. semidentata and O. chathamica to the

Chatham Islands, and O. oporina to Stewart Island and southernmost South Island.

Olearia colensoi grows in montane to subalpine scrub from lat. 38

southwards (North,

South, Stewart and Solander islands), descending to sea level in the southernmost part

of its range.

An analysis of the phylogeny of the large woody Australasian genus Olearia

based on ITS data

demonstrated the polyphyly of the genus and also provided some

support for each of the above conclusions, in that the sole macrocephalous Olearia

species included in the study, O. chathamica, was placed with Pleurophyllum, and

Damnamenia was also included in the clade.

Our analysis of the ITS and ETS sequences is that these are largely congruent.

This suggests the two data sets are converging on the same evolutionary tree. The ETS

sequences provide strong support for a clade comprised of Damnamenia, Pleurophyllum

and the macrocephalous Olearia species (bootstrap 92%) whereas this relationship is

not recovered by ITS sequences. The analysis of the combined data provides even

greater support and resolution (bootstrap 96%). Maximum likelihood analysis of the

combined data set (Fig. 3) again recovered the clade comprised of Damnamenia,

Pleurophyllum, and the macrocephalous Olearia species (the macrocephalous clade,

Fig. 2). Damnamenia is sister to the other members of this clade. The three species of

Pleurophyllum form a clade (bootstrap 64%), as do Olearia angustifolia and O. oporina

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

(bootstrap 100%), O. semidentata and O. chathamica (bootstrap 99%), and O. colensoi

and O. lyallii (bootstrap 100%).

Our results reveal significant rate variation across the lineages: the annual species

Senecio vulgaris is evolving at a faster rate than the arborescent Dasyphyllum

dicanthoides. We therefore used a rate-smoothing procedure to transform the maximum

likelihood tree. We calibrated the rate for the combined analyses using a fossil date of

38 million years, which represents the stem age of the Asteroideae in our analysis (Fig.

4). The mean substitution rates estimated for Pleurophyllum are similar to those

calculated for Abrotanella

, Robinsonia

and Dendroseris

despite the different

methods calibration (see Table 1). However they differ substantially from the mean

value for 21 angiosperm families calculated by Kay et al.

. They note that long-lived

woody plants appear to have a slower substitution rate; the fastest rate was recorded in

Gentianella, whereas the slowest rate was in the Winteraceae. Andreasen and Baldwin

noted a similar pattern in which the nucleotide substitution rate was faster in annual

species of Sidalcea than in perennial species, and the substitution rates in ITS and ETS,

though different, were correlated.

Our findings suggest that the stem age of many lineages of Asteraceae predate the

late Tertiary extinction of plants in Antarctica; similar results were reported by Kim et

al.

. The combined maximum likelihood analysis suggests the Damnamenia,

Pleurophyllum, and macrocephalous Olearia clade diverged during the mid-Pliocence

about 4.1 Myr ago. The crown radiation occurred at the boundary between the Pliocene

and Pleistocene. Independent analyses of the ITS and ETS sequences provided similar

divergence estimates. The late Tertiary was a time of intense environmental upheaval in

New Zealand brought about by the uplift of the Southern Alps, glaciation, and

volcanism

. Glaciers nearly completely covered the Auckland Islands and a

considerable portion of Campbell Island during the last glacial maximum, while

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

Macquarie Island remained unglaciated

. The retreat of the glaciers on Auckland and

Campbell islands occurred around 15,000 years ago. Pollen profiles from post-glacial

peat profiles indicate megaherbs such as Pleurophyllum were present in abundance at

the end of the last glaciation maximum (LGM) and survived close to the glacial

terminus.

The ancestors of megaherb clade must have survived the LGM in the New

Zealand subantarctic islands. There has subsequently been at least one instance of

dispersal to the mainland of New Zealand and to the Chatham Islands, about 800 km to

the east of New Zealand. There are a number of equally plausible area reconstructions,

provided our phylogenetic inference is correct. Most of these involve either several

independent dispersal events from the subantarctic islands to the mainland of New

Zealand, Stewart Island and the Chatham Islands or dispersal in a more stepwise fashion

from the New Zealand subantarctic islands to the mainland, and then subsequent

dispersal from the New Zealand mainland to the Chatham Islands and back to the New

Zealand subantarctic islands.

The megaherb growth form is not unique to the subantarctic islands and has

evolved independently in several distinct lineages of flowering plants. The large-leaved

buttercup Ranunculus lyallii is common on the mainland of New Zealand and

Mysotidium hortensia, a large forget-me-not, is endemic to the Chathams Islands.

Elsewhere the Chilean Gunnera, the Hawaian silverswords, Argyroxiphium, and the

Kenyan Dendrosenecio and Lobelia are notable examples. However, the megaherbs of

the subantarctic islands are unique with corrugated leaves, stereome tissue, hairy and

occasionally coriaceous lamina, a rosette growth form, fleshy root system and colourful

flowers

. The megaherb growth form may confer a selective advantage in subantarctic

environments

. Nutrient availability may be limiting as it varies most with moisture-

holding capacity and the acidity of the peat. Large leaves may intercept nutrients from

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

marine aerosols channelling nutrients directly to the stem base and roots. The large

overlapping leaves could act as solar panels focusing radiant energy towards the

growing apex of the plant. Hence the detrimental effects of cold temperatures and

decreases in wind speed would be moderated with the rosette growth form further

reducing reductions in water loss by transpiration. Light is also limiting and the big

leaves of Pleurophyllum suppress competitors with their dense and compact growth

form.

Present biogeographic interpretations of the Asteraceae neglect to consider the

importance of Antarctica as a corridor for migration during the early diversification of

the family

. The northern hemisphere circumboreal flora was once linked by land

bridges during the Tertiary. We would like to suggest a similar scenario of a

circumantarctic flora that flourished until the late Tertiary. Pulses of dispersal occurred

to the north during glacial advance with retreat and intermingling during interglacial

periods. Unlike the northern hemisphere, the circum-Antarctic vegetation was nearly

completely eliminated during the last ice age. The distinctive flora of the subantarctic

islands may harbour some of the last remnants of this once diverse flora, and plants such

as Pleurophyllum may be the key to resolving this puzzle.

METHODS

We included sequences from the three species of Pleurophyllum, Damnamenia

vernicosa and representatives of all of the macrocephalous species of Olearia. More

distant outgroups were five representatives from most of the other tribes of Asteroideae

recognized by Bremer

with greater sampling within tribe Astereae. We rooted the

analysis on the long branch leading to Dasyphyllum, which is placed in subfamily

Barnadesioideae.

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

Voucher information along with GenBank (http://www.ncbi.nlm.nih.gov)

accession numbers is detailed in the Appendix. The complete data sets are available on

request from the first author and were deposited in TreeBASE

(http://www.treebase.org/treebase) study accession number = ????? and matrix

accession numbers = ??????.

DNA extraction, amplification and sequencing. Total DNA was extracted from

leaves following a modification of the CTAB method of Doyle and Doyle

. The

amplification and sequencing procedure for the ITS region has been described

previously

. We also amplified the ETS region following the procedure of Baldwin and

Marcos

. DNA samples were labelled with fluorescent dyes (Big Dye Chemistry) and

then sequenced by the Allan Wilson Centre (Massey University) DNA sequencing

facility. In all instances both the forward and reverse DNA strands were sequenced.

Data analysis. The sequences were initially aligned using ClustalX

and gaps were

inserted in the data matrix. The resulting alignments were then visually inspected and

minor changes were made manually to ensure positional homology prior to the

phylogenetic analyses. The phylogenetic analyses were accomplished using PAUP*

version 4.0b10

with both parsimony and maximum likelihood selected as optimality

criteria. The parsimony analysis was conducted with the PAUP* settings TBR branch-

swapping, MULPARS, RANDOM ADDITION with 1,000 replicates. Duplicate trees

were eliminated using the condense trees option collapsing branches with a maximum

length of zero. The characters were unordered and equally weighted and gaps were

treated as missing data. The most appropriate maximum likelihood model and parameter

estimates were determined by the Akaike Information Criterion test (AIC) implemented

in Modeltest vers. 3.06

. Congruence of the data matrices was assessed using the ILD

test of Farris et al.

37,38

with 100 data partition replicates with uninformative positions

deleted/excluded

39,40,41

. Support for clades was estimated by bootstrap analyses

with

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

1,000 replications excluding uninformative sites; starting trees were obtained by

RANDOM ADDITION with one replication for each bootstrap replication, TBR

branch-swapping, MULPARS in effect and a MAXTREE limit of 1,000.

The likelihood ratio test (LR = -2 log LR; where LR is the difference between the

ln likelihood of the tree, with and without enforcing a molecular clock and the 2

distribution, with n2 degrees of freedom, where n is the number of taxa) was used to

determine whether the data satisfied the assumptions of a molecular clock

. In the

absence of a molecular clock, we used the semiparametric rate smoothing by penalized

likelihood approach to accommodate rate heterogeneity across lineages

44,45

. This

procedure is implemented in the program r8s

and uses a likelihood model combined

with a smoothing parameter estimated by cross-validation to estimate divergence times.

The checkGradient command was implemented to provide an additional assessment of

the divergence estimates. We calculated substitution rates for each gene and for the

combined data matrix using a calibration date of 38 Myr BP determined from the fossil

record

47,48

. In our analysis this is the split between subfamily Barnadesioideae and the

remaining members of the Asteraceae and represents the stem age of subfamily

Asteroideae. Confidence intervals were calculated by applying this procedure to 100

bootstrap trees. The influence of different tree topologies on the divergence estimates

was tested using the topological constraint option in PAUP*.

Present distributions were mapped on the inferred phylogeny and the most

parsimonious reconstructions of ancestral areas were determined using MacClade vers.

4.07

Godley, E. Botany of the Southern Zone Exploration to 1843. Tuatara 13,140

181 (1965).

Godley, E. Botany of the Southern Zone Exploration, 18431847. Tuatara 18,

4993 (1970).

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

Albach, D. & Chase, M. W. Paraphyly of Veronica (Veroniceae:

Scrophulariaceae): evidence from internal transcribed spacer (ITS) sequences of nuclear

ribosomal DNA. J. Plant Res. 114, 918 (2001).

Lockhart, P. J., McLenachan, P. A., Havell, D., Glenny, D., Huson, D. & Jensen

U. Phylogeny, radiation, and transoceanic dispersal of New Zealand alpine buttercups:

molecular evidence under split decomposition. Ann. Missouri Bot. Gard. 88, 458477

(2001).

5. Knapp,

M. et al. Relaxed molecular clock provides evidence for long-distance

dispersal of Nothofagus (Southern Beech). PLoS Biology 3, (2005).

Knapp, M., Mardulian, R., Havell, D., Wagstaff, S. J. & Lockhart P. J. The

drowning of New Zealand and the problem of Agathis. Syst. Biol. (2007).

Milne, R. I. Northern hemisphere plant disjunctions: a window on Tertiary land

bridges and climate change. Ann. Bot. 98, 465472 (2006).

Hewitt, G. The genetic legacy of the Quaternary ice ages. Nature 405, 907913

(2000).

Ashworth, A. C. & Cantrill, D. J. Neogene vegetation of the Meyer Desert

Formation (Sirius Group) Transantarctic Mountains, Antarctica. Palaeogeog.,

Palaeoclimatol., Palaeoecol. 213, 6582 (2004).

10.

Hill, R. S., Harwood, D. M., Webb, P-N. Nothofagus beardmorensis

(Nothofagaceae), a new species based on leaves from the Pliocene Sirius Group,

Transantarctic Mountains, Antarctica. Rev. Palaeobot. Palynol. 94, 1124 (1996).

11.

Webb, P-N. The Cenozoic history of Antarctica and its global impact. Ant. Sci.

2, 321 (1990).

12.

Bergstrom, D. M & Chown, S. L. Life at the front: history ecology and change

on southern ocean islands. Trends Ecol. Evol. 12, 472477 (1999).

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

13.

Convey P. & Stevens, M. I. Antarctic biodiversity. Science 317, 18771878

(2007).

14.

McGlone, M. S. The late Quaternary peat, vegetation and climatic history of the

Southern Oceanic Islands of New Zealand. Quat. Sci. Rev. 21, 683707 (2002).

15.

McGlone, M. S. in Biological Invasions in New Zealand (eds Allen, R. B. &

Lee, W. G.) 186: 1732 (Berlin Heidelberg, SpringerVerlag, 2006).

16.

Chown, S. L., Rodrigues, A. S. L., Gremmen, N. J. M. & Gaston, K. J. World

heritage status and conservation of southern ocean islands. Conserv. Biol. 15, 550557

(2001).

17.

Hooker, J. D. The Botany of the Antarctic Voyage of J. M. S. Discovery Ship

Erebus and Terror, In the Years 18391843, Vol. 1, Flora Antarctica, Part 2. (London:

Reeve, 1847).

18.

Given, D. R. Damnamenia gen. nov. A new subantarctic genus allied to

Celmisia Cass. (Astereae Compositae). New Zealand J. Bot. 11, 785796 (1973).

19.

Dury, D. G. A clarification of the generic limits of Olearia and Pleurophyllum

(Astereae Compositae). New Zealand J. Bot. 6, 459466 (1968).

20.

Cross, E. W., Quinn, C. J., & Wagstaff, S. J. Molecular evidence for the

polyphyly of Olearia (Astereae: Asteraceae). Plant Syst. Evol. 235, 99120 (2002).

21.

Wagstaff, S. J., Breitwieser, I. & Swenson, U. Origin and relationships of the

austral genus Abrotanella (Asteraceae) inferred from DNA sequences. Taxon 55, 95

106 (2006).

22.

Sang, T., Crawford, D. J. Stuessy, T. F. & Silva, O. M. ITS sequences and

phylogeny of the genus Robinsonia (Asteraceae). Syst. Bot. 4, 5564 (1995).

23.

Sang, T., Crawford, D. J., Kim, S-C. & Stuessy, T. F. Radiation of the endemic

genus Dendroseris (Asteraceae) on the Juan Fernandez Island: Evidence from

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

sequences of the ITS regions of nuclear ribosomal DNA. Amer. J. Bot. 81, 14941501

(1994).

24.

Kay, K. M., Whittall, J. B. & Hodges, S. A. A survey of nuclear ribosomal

internal transcribed spacer substitution rates across angiosperms: an approximate

molecular clock with life history effects. BMC Evol. Biol. 6, 36 doi:10.1186/1471-2148-

6-36 (2006)

25.

Andreasen, K. & Baldwin, B. Unequal evolutionary rates between annual and

perennial lineages of checker mallows (Sidalceae, Malvaceae): evidence from 18S26S

rDNA internal and external transcribed spacers. Mol. Biol. Evol. 18, 936944 (2001).

26.

Kim, KJ., Choi, KS. & Jansen, R. K. Two chloroplast DNA inversions

originated simultaneously during the early evolution of the sunflower family

(Asteraceae). Mol. Biol. Evol. 22, 17831792 (2005).

27. Fleming,

The Geological History of New Zealand and its life. (Auckland

University Press, Auckland, reprinted 1989).

28.

Nicholls, V. J. & Rapson, G. L. Biomass allocation in Subantarctic Island

megaherbs, Pleurophyllum speciosum (Asteraceae) and Anisotome latifolia (Apiaceae).

New Zealand. J. Ecol. 23, 8793.

29.

Funk, V. A. et al. Everywhere but Antarctica: Using a supertree to understand

the diversity and distribution of the Compositae. Biol. Skr. 55, 343374 (2005).

30. Bremer,

Asteraceae -- Cladistics and Classification. (Oregon: Timber Press,

1994).

31.

Doyle, J. J. & Doyle, J. L. A rapid DNA isolation procedure for small quantities

of fresh leaf tissue. Phytochem. Bull. 19, 1115 (1987).

32.

Wagstaff, S. J. & Breitwieser, I. Phylogenetic relationships of New Zealand

Asteraceae inferred from ITS sequences. Plant Syst. and Evol. 231, 203224 (2002).

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

33.

Baldwin, B. G. & Marcos, S. Phylogenetic utility of the external transcribed

spacer (ETS) of 18S26S rDNA: congruence of ETS and ITS trees of Calycadenia

(Compositae). Proc. Nat. Acad. Sci. USA 95, 449463 (1998).

34.

Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F. & Higgins D. G.

The ClustalX windows interface: flexible strategies for multiple sequence alignment

aided by quality analysis tools. Nucl. Acids Res. 24, 48764882 (1997).

35.

Swofford, D. L. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other

Methods), version 4.0b10. (Sinauer, Sunderland, MA, 2002).

36.

Posada, D. & Crandall, K. A. Modeltest: testing the model of DNA substitution.

Bioinformatics 14, 817818 (1998).

37.

Farris, J. S., Källersjö, M., Kuge, A. G. & Bult, C. Testing significance of

incongruence. Cladistics 10, 315319 (1994).

38.

Farris, J. S., Källersjö, M., Kuge, A. G. & Bult, C. Constructing a significance

test for incongruence. Syst. Biol. 44, 570572 (1995).

39.

Ramírez, MJ. Further problems with the incongruence length difference test:

"hypercongruence" effect and multiple comparisions. Cladistics 22, 289295 (2006).

40.

Cunningham, C. Can three incongruence tests predict when data should be

combined? Mol. Biol. Evol. 14, 733740 (1997).

41.

Quicke, D. L. J., Jones, O. R., & Epstein, D. R. Correcting the problem of false

incongruence due to noise imbalance in the incongruence length test (ILD) test. Syst.

Biol. 56, 496503 (2007).

42.

Felsenstein, J. Confidence limits on phylogenies: an approach using the

bootstrap. Evolution 39, 783791 (1985).

43.

Felsenstein, J. Phylogenies from molecular sequences: inference and reliability.

Ann. Rev. Genet. 22, 521565 (1988).

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

44.

Sanderson, M. J. A nonparametric approach to estimating divergence times in

the absence of rate constancy. Mol. Biol. Evol. 14, 12181231 (1997).

45.

Sanderson, M. J. Estimating absolute rates of molecular evolution and

divergence times: a penalized likelihood approach. Mol. Biol. Evol. 19, 101109 (2002).

46.

Sanderson, M. J. r8s 1.50. Computer program and documentation available from

http://phylo.ucdavis.edu/r8s/r8s.html. (2002).

47.

DeVore, M. L. & Stuessy, T. F. in Advances in Compositae Systematics (eds

Hind, D. J. N., Jeffrey, C. & Pope, G. V.) 2340 (Royal Botanic Gardens, Kew, 1995).

48.

Graham, A. in: Compositae Systematics. Proceedings of the International

Compositae Conference, Kew, 1994, Vol. 1. (eds Hind, B. J. N. & Beentje, H. J.) 123

140 (Royal Botanic Gardens, Kew, 1996).

49.

Maddison, W. P. & Maddison D. R. MacClade version 4.07. Analysis of

Phylogeny and Character Evolution. (Sinauer, Sunderland, MA, 2005).

Supplementary Information is linked to the online version of the paper at www.nature.com/nature

Acknowledgements We thank the following for their assistance: Colin Meurk, Phil Novis, Rich Leschen,

Peter de Lange, the New Zealand Plant Radiation Network and the New Zealand Plant Conservation

Network. This research was funded by the New Zealand Foundation for Research Science and

Technology.

Author information Reprints and permissions information is available at

npg.nature.com/reprintsandpermissions. Correspondence and requests for materials should be addressed

to S.J.W. (wagstaffs@landcareresearch.co.nz).

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

Table 1 Comparison of nucleotide substitution rates

Taxon Genomic

region

Nucleotide

substitution rate

Calibration Reference

Pleurophyllum

ETS

9.59 × 10

-9

per

site per year

First appearance of

modern Asteraceae in

the fossil record

(DeVore & Stuessy

1995; Graham 1996)

This paper

Pleurophyllum

ITS

4.98 × 10

-9

per

site per year

First appearance of

modern Asteraceae in

the fossil record

(DeVore & Stuessy

1995; Graham 1996)

This paper

Abrotanella

ITS

8.19 × 10

-9

per

site per year

First appearance of

modern Asteraceae in

the fossil record

(DeVore & Stuessy

1995; Graham 1996)

Wagstaff et

al. (2006)

Dendroseris

ITS

6.06 × 10

-9

per

site per year

Age of Masatierra in

Juan Fernandez Islands

approximately 4 Myr;

age of Masafurera

approximately 12 Myr

Sang et al.

(1994)

Schmidt &

Schilling

(2000)

Robinsonia

ITS

7.83 × 10

-9

per

site per year

Age of Masatierra in

Juan Fernandez Islands

approximately 4 Myr;

age of Masafurera

approximately 12 Myr

Sang et al.

(1995)

Tarweeds/Hawaiian

silverswords

ITS

3.00 × 10

-9

per

site per year

Aridification

accompanying uplift of

Baldwin &

Sanderson

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

Figure 1 Subantarctic island herbfields. A. Panoramic view of Campbell

Island. B. Lyall Ridge herbfield community dominated by Pleurophyllum

speciosum. C. Inflorescence of P. speciosum. Photos by Colin Meurk.

Figure 2 Distribution of the macrocephalous clade. Damnamenia vernicosa:

monotypic genus endemic to Campbell and Auckland islands. Pleurophyllum:

genus of three species, P speciosum, P. hookeri, and P. criniferum, endemic to

Auckland, Campbell, Antipodes and Macquarie islands. Olearia lyallii: endemic

to the Snares and Auckland islands. O. colensoi: North, South, Stewart and

Solander islands. O. angustifolia: confined to Stewart Island, the South Island

shores of Foveaux Strait and surrounding smaller islands. O. oporina: South

Island, sounds of Fiordland from Martins Bay southward. O. chathamica:

endemic to the Chatham Islands, known from the southern tablelands, Pitt,

Mangere and South-East islands. O. semidentata: endemic to the Chatham

Islands.

Figure 3 Maximum likelihood analysis of the combined data set. The

macrocephalous clade was again recovered, with Damnamenia sister to the

other members of this clade. The three species of Pleurophyllum form a clade

as do Olearia angustifolia and O. oporina, O. semidentata and O. chathamica,

and O. colensoi and O. lyallii.

Figure 4 Calibrated rate for the combined analyses using a fossil date of

38 million years. Our results suggest the macrocephalous clade diverged

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

Olearia chrysophylla

Olearia oppositifolia

Olearia rosmarinifolia

Olearia myrsinoides

Olearia megalophylla

Olearia ledifolia

Olearia cheesemanii

Olearia rani

Olearia albida
Olearia paniculata

Olearia solandri

Olearia traversii

Pachystegia insignis

Celmisia asteliifolia

Celmisia tomentella

Celmisia mackaui

Olearia chathamica

Olearia semidentata

Olearia colensoi

Olearia lyallii

Olearia angustifolia

Olearia oporina

Pleurophyllum criniferum

Pleurophyllum hookeri

Pleurophyllum speciosum

Damnamenia vernicosa

Chiliotrichium rosmarinifolium

Lagenifera pumila
Vittadinia australis

Brachyscome humilis

Inula orientalis

Argyrotegium fordianum

Cotula coronopifolia

Senecio vulgaris

Abrotanella muscosa

Calendula officinalis

Dasyphyllum diacanthoides

0.01 substitutions/site

Maximum Likelihood
GTR + G + I model
-Ln = 9092.75295

New Zealand

Subantarctic islands
Subantarctic islands

Stewart Island

Subantarctic islands

Chatham Islands

New Zealand

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007

Oligocene

23.8

5.3 1.8

Million Years

Miocene

Pleistocene

Pliocene

Divergence Time Estimate
Method = Penalized Likelihood
Algorithm = Truncated Newton
Fixage = 38 Million Years

Olearia chrysophylla
Olearia rosmarinifolia
Olearia oppositifolia
Olearia myrsinoides
Olearia megalophylla
Olearia ledifolia
Olearia cheesemanii
Olearia rani
Olearia albida
Olearia paniculata
Olearia solandri
Olearia traversii
Pachystegia insignis
Celmisia asteliifolia
Celmisia tomentella

Celmisia mackaui

Olearia chathamica
Olearia semidentata

Olearia colensoi
Olearia lyallii

Olearia angustifolia
Olearia oporina
Pleurophyllum criniferum
Pleurophyllum speciosum
Pleurophyllum hookeri
Damnamenia vernicosa

Chiliotrichium rosmarinifolium
Lagenifera pumila
Vittadinia australis
Brachyscome humilis
Inula orientalis
Argyrotegium fordianum
Cotula coronopifolia
Senecio vulgaris
Abrotanella muscosa
Calendula officinalis
Dasyphyllum diacanthoides

Eocene

33.7

Nature Precedings : hdl:10101/npre.2007.1272.1 : Posted 28 Oct 2007