| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics |
2Department of Ecology and Evolutionary Biology, Yale University, Osborn Memorial Laboratories, P.O. Box 208106, New Haven, Connecticut 06520 USA; and 3Department of Systematic Botany, Institute of Biological Sciences, University of Aarhus, Nordlandsvej 68, DK-8240 Risskov, Denmark; 4Department of Botany, University of Wisconsin, 430 Lincoln Drive, Madison, Wisconsin 53706 USA; and 5The Lewis B. and Dorothy Cullman Program for Molecular Systematics Studies, The New York Botanical Garden, Bronx, New York 10458-5126 USA
Received for publication July 11, 2002. Accepted for publication January 16, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Australia • dispersal • Goodeniaceae • Hawaiian Islands • ITS • Pacific biogeography • Scaevola
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), contains approximately 400 species in 11 genera and constitutes a significant element of Australian and Pacific island floras. The family is contained within the Asterales as a possible sister group to the Asteraceae (Gustafsson et al., 1996
) or more likely to Calyceraceae + Asteraceae. (Kårehed et al., 1999
). While most genera of Goodeniaceae are almost entirely confined to Australia, one genus, Scaevola, has dispersed and radiated throughout much of the Pacific. Of approximately 130 species of Scaevola, approximately 40 occur outside Australia. Two of these are widespread strand species with distributions throughout the Pacific and Indian Oceans (S. taccada) or in the tropical Americas and Africa (S. plumieri). The remaining species, in contrast, are narrower endemics often occurring well inland as forest trees or shrubs. Island species occur in the Pacific, northward to the Philippines and China, eastward as far as the Marquesas, and northeast to the Hawaiian Islands. Additionally, one species is known from Cuba and one from Socotra.
Most Australian Scaevola have dry fruits and sprawling, herbaceous to shrubby habits. In contrast, almost all of the extra-Australian species have fleshy fruits and are often tall shrubs or trees. These characteristics, coupled with axillary inflorescences, were used by Carolin (1990)
to define Scaevola section Scaevola, which comprises all but two (S. gracilis and S. oppositifolia) of the extra-Australian species along with five species restricted to Australia. However, given the propensity of island colonists to show convergent evolution of such characteristics as fleshy fruits and arborescence (Carlquist, 1965
, 1974
, 1980
) the monophyly of Scaevola sect. Scaevola should not be assumed without additional evidence.
Carolin et al. (1992)
placed nearly all the rest of the genus (64 Australian species and S. gracilis from Tonga and New Zealand) in section Xerocarpa, defined by dry exocarps and determinate inflorescences. Two remaining species, S. oppositifolia and S. enantophylla, which have fleshy fruits, opposite leaves, a vining habit, and occur from northern Australia to the Philippines, have been segregated as a third section, Scaevola sect. Enantiophyllum (Carolin et al., 1992
).
The most successful dispersers in the genus are S. taccada and S. plumieri, which together are pantropical in distribution, each occurring in both the northern and southern tropics, with overlapping distributions on the coasts of the Indian Ocean. Their wide distribution appears to be due in part to their ability to be transported both in avian guts and by oceanic floating (remaining viable after floating in seawater for several months; Guppy, 1917
; Lesko and Walker, 1969
). Given these efficient dispersal mechanisms, an obvious hypothesis to explain the numerous island endemic species would be that S. taccada and S. plumieri have been the prominent colonizers, giving rise to inland species on each of the high island groups where they have established. This hypothesis predicts that island endemics are closely related to populations of the widespread taxon that occurs on their island. Alternatively, if birds can disperse fleshy fruits long distances, islands could be colonized directly from other islands or Australia independent of oceanic dispersal of these widespread species.
The largest number of species in a single archipelago occurs in the Hawaiian Islands, which is home to S. taccada and nine endemic species, of which eight are diploid and one is tetraploid (Patterson, 1990
). Using morphological data, Patterson (1995)
concluded that the endemic diploid taxa form a monophyletic clade. The sister group to the Hawaiian diploids was S. taccada (Patterson, 1995
), a taxon relationship favoring a scenario in which inland species are derived from widespread progenitors. Scaevola glabra, the single tetraploid species, is quite different from the Hawaiian diploids, being instead similar to the New Caledonian S. coccinea, with which it shares a similar tubular flower (Carlquist, 1969
). Scaevola glabra is the only clear example of a stable tetraploid species known in the genus, although a few Australian species have varying chromosome numbers and many species have not been analyzed (Peacock, 1963
; Carr, 1998
). Patterson's (1995)
analysis supported this conclusion, suggesting that, in this case, long-distance dispersal between islands has occurred. However, because Patterson (1995)
included only a small number of taxa, none of them Australian, these suggestions are in need of confirmation.
In 1996, a new Hawaiian Scaevola species was described from the island of Maui (Wagner, 1996
). This species, Scaevola hobdyi, still known only from a holotype with floral buds, is now presumed to be extinct. Based on its linear, spirally arranged leaves, which are very different from all other Hawaiian species, Wagner (1996)
suggested that S. hobdyi represents an additional dispersal event to the Hawaiian Islands.
To examine the relationships and dispersal patterns of the extra-Australian Scaevola species, we sequenced the internal transcribed spacers and 5.8S gene (collectively, ITS) of nuclear ribosomal DNA for 51 Scaevola species and eight outgroup taxa. The outgroup includes species from Goodenia, Diaspasis, Velleia, and Verreauxia and were chosen based on their positions close to the Scaevola lineage in both rbcL and morphological studies (Gustafsson et al., 1996
, 1997
). The ITS region has the appropriate variation for intra-generic phylogenetic analyses, is flanked by highly conserved genic regions (18S and 26S), and encompasses the 5.8S gene, simplifying primer design and sequence alignment (Baldwin et al., 1995
). These data were analyzed using phylogenetic methods to elucidate the dispersal patterns, biogeography, and evolutionary history of Scaevola.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. The DNA was extracted from fresh material, silica gel samples, or herbarium sheets. The fresh material (which included only the Hawaiian species) was extracted and purified with a modification of the cetyl trimethyl ammonium bromide (CTAB) isolation method (Doyle and Doyle, 1987
). These samples were accessioned into the Hawaiian Plant DNA Library (HPDL; Morden et al., 1996
). Silica gel and herbarium material were extracted with CTAB buffer, glass milk, and NaI solution (Bio 101, Qbiogene, Carlsbad, California, USA) as described in Struwe et al. (1998)
.
DNA sequencing and alignment
The ITS region, including ITS 1, 2, and 5.8S, was amplified using primers 90 (5'-TATGCTTAAAYTCAGCGGGT-3') and 91 (5'-AACAAGGTTTCCGTAGGTGA-3') modified from Baldwin (1992
) or primers 4 (5'-TCCTCCGCTTATTGATATGC-3') and LEU (or ITS-I) (5'-GTCCACTGAACCTTATCATTTAG-3') (White et al., 1990
; Urbatsch et al., 2000
). These primer pairs were also used for sequencing along with two internal primers, 2 (GCTGCGTTCTTCATCGATGC) and 3b (GCATCGATGAAGAACGTAGC) (White et al., 1990
; Baum et al., 1998
). Amplifications utilized the following cycling program: 95°C, 50 s; 60°C, 50 s; 72°C, 1 min 50 s; repeat 30 cycles; hold 4°C. The reactions were performed using Taq DNA Polymerase (Boehringer Mannheim, Indianapolis, Indiana, USA or QIAGEN Valencia, California, USA) in 25 µL, with final concentrations of 2.5 mmol/L MgCl2, 0.5 µmol/L of each primer, 0.8 mmol/L dNTPs, 0.004% BSA (New England Biolabs, Beverly, Massachusetts, USA), and 0.001 mmol/L TMACl (Sigma, St. Louis, Missouri, USA). The samples were cleaned with the Qiagen PCR cleanup kit (Qiagen). The PCR products were sequenced using BigDye or dRhod systems (PE Applied Biosystems, Foster City, California, USA) according to the manufacturer's instructions and viewed with an ABI 377 (Applied Biosystems). Sequences were initially aligned with Clustal W (Thompson et al., 1994
) using a gap open penalty = 15 and gap extension penalty = 6.66. The alignment was modified by eye in Sequencher 3.0 (Gene Codes Corporation, Ann Arbor, Michigan, USA).
Phylogenetic analyses
In initial analyses all characters were coded as unordered with equal weight, and gaps were coded as missing. The matrix was analyzed using the maximum parsimony criterion (Fitch, 1971
) with PAUP* 4.0b8 (Swofford, 2001
). Heuristic searches were performed with 100 random taxon addition replicates using tree bisection-reconnection (TBR) branch-swapping, with steepest descent off. Internal branch support was estimated with a bootstrap analysis (Felsenstein, 1985
) from 1000 pseudoreplicates subject to simple taxon addition heuristic searches as described but with MaxTrees set to 1000.
To evaluate the dependence of the phylogeny on the model of evolution assumed, we explored alternative weighting schemes: (1) transition : transversion (ti : tv) bias = 2:1 and 3:1, (2) gapmode = newstate, and (3) gaps scored as additional characters weighted 1, 2, or 3. Searches were conducted as in the flat-weighted analysis. These analyses are not reported unless they bear on the specific conclusions reached.
We tested a number of a priori hypotheses of monophyly (see Results) using the Wilcoxon signed ranks (WSR; Templeton, 1983
) and Kishino-Hasegawa (KH; Kishino and Hasegawa, 1989
) tests as implemented in PAUP*. The 864 most parsimonious trees were compared to the optimal trees found during constrained parsimony searches (using the same settings as for unconstrained searches). The Kishino-Hasegawa test was performed using a maximum likelihood optimality criterion with the HKY85 model of molecular evolution (Hasegawa et al., 1985
) with empirical base frequencies both with and without allowance for heterogeneity of rates among sites using a discrete approximation to a gamma distribution.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Maximum parsimony analysis produced 864 trees of 767 steps, with a consistency index of 0.6310 and a retention index of 0.8180. Despite the limited sampling of taxa outside of Scaevola, these data indicate that Goodenia is not monophyletic and includes the genera Verreauxia and Velleia (Fig. 1). Additionally, the ITS data strongly support the removal of Scaevola collaris from the rest of the genus and the inclusion of the monotypic genus, Diaspasis, in Scaevola. The genus Scaevola, excluding S. collaris but including Diaspasis, was well supported with 100% bootstrap support (Fig. 1).
|
|
The data strongly reject the monophyly of the extra-Australian taxa (WSR test, P = 0.0001; KH test, P = 0.0001) suggesting, instead, at least six dispersal events from Australia into the Pacific basin. Four of these dispersal events each led to single species (Dispersals 1–4, Figs. 2 and 3): one, S. oppositifolia in S. sect. Enantiophyllum, involved its dispersal from Australia to the Philippines; another, in S. sect. Xerocarpa subsect. Parvifoliae, resulted in the tetraploid S. glabra in the Hawaiian Islands; and two events in S. sect. Xerocarpa subsect. Biloculatae ser. Globuliferae, one giving rise to S. beckii in New Caledonia and the other giving rise to S. gracilis in New Zealand and Tonga.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Additionally, all Scaevola have 1–2 (–4) seeded, indehiscent fruit, although this character is also present in Dampiera and a few species of Goodenia sensu lato (including Verreauxia, Carolin et al., 1992
). Finally, the presence of barbulae at the edge of petal wings and in the throat of the corolla is restricted to Scaevola, although not present in all species (Carolin et al., 1992
).
Diaspasis correctly placed in the genus Scaevola
Diaspasis filifolia is a monotypic Australian genus that is very similar to Scaevola in fruit and seed morphology. The molecular data suggest that it is embedded within Scaevola, apparently as sister to representatives of S. sect. Xerocarpa subsect. Pogonanthera (Fig. 1). A similar close relationship was found with rbcL data (Gustafsson et al., 1996
), and the overlapping similarities of the genera were discussed by Carolin (1978)
. These results suggest that the original classification of D. filifolia R. Br. as Scaevola clandestina F. Muell was correct (von Mueller, 1859
).
The corolla of D. filifolia differs from most Scaevola in having the lobes arranged almost radially or in a 2 + 3 arrangement (i.e., with two abaxial and three adaxial petals). However, this condition is approached in S. striata, which is also in subsect. Pogonanthera (Carolin et al., 1992
). Characters that further support a close relationship of D. filifolia with S. ramosissima and S. phlebopetala include flowers in terminal racemes with leaf-like bracts, well-developed peduncle with obsolete pedicel, and a weak-stemmed (often decumbent, prostrate, or ascending) herbaceous habit.
Exclusion of Scaevola collaris from Scaevola
The molecular data suggest that S. collaris is more closely related to elements of the non-monophyletic Goodenia than to the remainder of Scaevola, despite having the fan-shaped corolla. The exclusion of S. collaris from Scaevola is also supported by seed testa anatomy, which is more similar to Goodenia species than to Scaevola species (M. H. G. Gustafsson, unpublished data). Additionally, the fruit of S. collaris is unique in Scaevola because of its beak and sponge-like (although hard), lacunate endocarp (Carolin et al., 1992
). A more thorough sampling of Goodenia and related genera is necessary to clarify the placement of S. collaris within that genus.
Subgeneric classification
Carolin (1990)
used morphological characters to separate the genus Scaevola into three sections—Scaevola, Xerocarpa, and Enantiophyllum. Species in S. sect. Scaevola are shrubs to small trees with fleshy exocarps. Although the section was also defined by axillary inflorescences (Carolin, 1990
), this characterization is problematic because some species (e.g., S. floribunda from Fiji) assigned to the section have terminal inflorescences. Scaevola sect. Xerocarpa included all the remaining Australian species, which range from herbaceous to small shrubs, have terminal inflorescences, and generally have fruit with dry exocarps. Scaevola gracilis was also included in S. sect. Xerocarpa (despite having fleshy fruits) based on its terminal inflorescence. The third section, Enantiophyllum, has been circumscribed to include from one to 10 species (Leenhouts, 1957
), but currently includes just two species (Carolin et al., 1992
), S. oppositifolia and S. enantophylla. This section is characterized by fleshy mesocarps, axillary inflorescences, a vining habit, and opposite leaves. The latter two characteristics are unique in the family.
Scaevola sect. Enantiophyllum is represented in our study by both species. Based on ITS data, this section is monophyletic and sister to the remainder of the genus. Thus, our data support its recognition. In contrast, the molecular data do not support Carolin's sections Scaevola or Xerocarpa, which appear intermingled across the entire tree (Fig. 1).
Despite the non-monophyly of S. sect. Xerocarpa, the subsections and series recognized by Carolin et al. (1992)
are broadly consistent with the molecular phylogenetic analysis. All species sampled from subsect. Biloculatae ser. Globuliferae (except for S. collaris; see earlier) fall in a clade, globulifera, that includes no other subsections or series of S. sect. Xerocarpa (but does include a number of species formerly assigned to S. sect. Scaevola). The five species sampled from the other series of subsect. Biloculatae, ser. Pogogynae form a well-supported clade. The two species from subsect. Parvifoliae form a clade, whereas the three species from subsect. Pogonanthera fall into two clades. Therefore, despite the limited sampling within the Australian clades, our data indicate that many of Carolin's smaller delineations within S. sect. Xerocarpa are monophyletic, despite the non-monophyly of the section itself.
Among the sampled species assigned to S. sect. Scaevola, three (S. glabra, S. tomentosa, and S. spinescens) appear to each be most closely related to separate species from S. sect. Xerocarpa rather than S. sect. Scaevola. The rest of S. sect. Scaevola is divided between two separate clades (northern and southern clades, Fig. 2) composed of multiple extra-Australian species (Fig. 1). The southern clade is sister to the entire globulifera clade, which includes the northern clade.
Some morphological characters are consistent with the northern and southern clades not being closely related. For instance, species in the southern clade have inflorescences in dense, many flowered, terminal racemes, whereas species in the northern clade have lax, axillary, cymose inflorescences with no more than eight flowers per cyme (one-flowered in S. coriacea and S. gaudichaudii). Additionally, the species in the southern clade have conspicuous barbulae in the throat of the corolla, a characteristic that is common in the family (Carolin et al., 1992
), but completely absent in the Hawaiian endemic diploid lineage (although found in other members of the northern lineage).
Biogeography and dispersal
Scaevola dispersal ability
Our phylogenetic analyses suggest that Scaevola, despite being nested within a family otherwise confined to the Australian continent, has considerable dispersal ability. The phylogeny implies even more dispersal events than a pure count of species and geographic distributions would suggest. For example, it indicates that the most isolated island archipelago in the world, the Hawaiian Islands, was colonized three separate times.
While the traditional classification of Scaevola (Carolin, 1990
) implies few origins of fleshy fruit, our molecular data suggest that fleshy fruits have evolved multiple times (Fig. 2). In fact, fruit fleshiness may have been a key innovation permitting transoceanic dispersal, given the fact that every species found outside of Australia has fleshy (drupaceous) fruits. Fleshy fruits likely facilitate dispersal by migrating sea birds, which can nest in montane habitats (Fosberg, 1947
), and these birds occasionally add small fruits to their diets (Ziegler, 2002
). In each case of extra-Australian dispersal (except in S. sect. Enantiophyllum), fleshy fruits appear to have evolved concomitantly (Fig. 2). Not all cases are as clear, however, because the Australian sister groups of S. gracilis (Dispersal 3) and of the northern clade (Dispersal 6) have partially fleshy fruits (Carolin et al., 1992
; Sykes, 1998
).
Single-species dispersal events
Scaevola oppositifolia, distributed in New Guinea, Indonesia, and the Philippines, is currently considered a separate species from S. enantophylla (Carolin et al., 1992
), which occurs in northern Australia. Scaevola oppositifolia represents a case of dispersal northward from Australia, without subsequent speciation, although considerable morphological variability has resulted in descriptions of up to 10 separate species (Leenhouts, 1957
).
Additionally, the ITS data suggest that the endemic Hawaiian tetraploid Scaevola glabra is sister to the two species in S. sect. Xerocarpa subsect. Parvifoliae, S. angulata and S. depauperata (Carolin et al., 1992
). These two taxa are, like S. glabra, tall shrubs (
1.5 m) and are often glabrous. The two Australian species differ from S. glabra in many morphological features, e.g., nontubular corollas and terminal inflorescences. Some similarity is seen, however, in the calyxes of the three species, which are basally fused with well-developed, free lobes. It is noteworthy that S. angulata has dispersed to islands off the coast of northern Australia, suggesting some significant dispersal ability.
Furthermore, the New Caledonian species S. beckii is not related to other New Caledonian species but rather to the Australian clade that includes S. globosa, S. humifusa, and S. repens. Phylogenetic reconstruction of the ITS data forcing S. beckii to associate with other island clades resulted in a significant increase in tree length (11 steps, Table 1). Additionally, the three Australian species are morphologically similar to each other and are united with S. beckii by having tiny flowers, bearded corollas, and one-seeded fruits. Thus, the evidence at hand supports S. beckii being a separate dispersal from Australia that has not resulted in subsequent speciation.
Similarly, Scaevola gracilis, occurring in New Zealand, the Kermadec Islands, and Tonga, is nested within an Australian lineage in our analysis (Fig. 2). Although, it falls in the general vicinity of the northern island clade, forcing it into that clade resulted in significantly longer trees (seven steps, Table 1).
The exact affinities of S. gracilis are unclear based on ITS data (Fig. 1). Scaevola gracilis is a broader endemic, being unusual among island species in its presence on multiple islands. This distribution may reflect the fact that this species has fruit that, unlike other montane species (Guppy, 1906
), have retained the ability to float in seawater for up to 10 d (Sykes, 1998
). Additionally, birds have been witnessed eating its fruits (Oliver, 1910
) and may contribute to inter-island dispersal. Carolin's (1990)
morphological analysis resolved S. gracilis and S. calendulacea as sister species. Other authors have also noted significant similarities between these two species as well as the other members of this clade (S. crassifolia, S. lanceolata, and S. virgata) in habit, type of indumentum, and presence of terminal spikes of sessile flowers (Carolin et al., 1992
; Sykes, 1998
). Interestingly, Scaevola calendulacea has white, partially fleshy fruits and occurs on the southeastern coast of Australia (Sykes, 1998
). The relationships among these taxa require further investigation.
Southern clade
The southern clade comprises three well-supported lineages: (1) the widespread, strand species S. taccada, (2) the New Caledonian species (excluding S. beckii), and (3) the South Pacific taxa (excluding S. gracilis). The relationship among these three clades is unresolved.
Scaevola taccada occurs on tropical Pacific and Indian Ocean coasts including most islands that house endemic montane Scaevola species (Fig. 3C). The fruit morphology of S. taccada is distinct from the island endemics in its white fleshy exocarp and an additional corky layer underneath, which aids in oceanic flotation. The potential for S. taccada to disperse between islands appears greater than that for the montane endemics (Guppy, 1917
; Lesko and Walker, 1969
), supporting the possibility that on each island strand S. taccada have spawned montane endemic radiations. However, this hypothesis is contradicted by the fact that all accessions of S. taccada, from diverse geographic locations, form a well-supported monophyletic clade (Fig. 1).
An alternative scenario to explain the distribution of island endemics is via stepping-stone dispersal eastward from New Caledonia across the South Pacific into Fiji, Tahiti, Samoa, and the Marquesas Islands (Fig. 3C). This scenario is compatible with an apparent trend in inflorescence evolution. Island endemics in the southern clade have characteristically dense thyrsoid inflorescences. This feature becomes successively more pronounced from west to east, with the inflorescence in the Marquesan Island species superficially resembling a capitulum.
Northern clade
The northern clade contains four island lineages: (1) the widespread Scaevola plumieri, (2) the endemic Hawaiian diploids, (3) S. socotraensis (Socotra), and (4) S. wrightii (Cuba). Also included in the clade is S. nitida (Australia). Monophyly of this clade is not well supported based on ITS data nor is the relationship among the five component lineages resolved. Additionally, due to a lack of variation, the ITS data does not rule out the possibility that the Bornean taxa, S. chanii and S. micrantha, may be part of the same dispersal from Australia.
Scaevola plumieri is a widespread strand species that occurs on coasts of the tropical Americas and Africa, Madagascar, Ceylon, southern India, the Mascarenes, and the Galapagos Islands (van Balgooy, 1975
, Fig. 3D). This species occurs in coastal habitats and has seeds that can float and remain viable for 4–5 mo (Guppy, 1917
; Carlquist, 1974
), although it lacks the corky layer of S. taccada. The finding that S. socotraensis and S. wrightii are closely related to S. plumieri is not surprising, given the biogeography of these species and their morphological similarity (M. H. G. Gustafsson, personal observation). Biogeographically, the relationship of the Hawaiian endemic diploids to these lineages is more anomalous because S. plumieri does not occur in the Hawaiian Islands, although the coastal species S. coriacea (from the Hawaiian Islands) shares more morphological similarity with S. plumieri than with S. taccada (observed by Carlquist, 1969
). Additionally, St. John (1962)
described S. coriacea as being the most similar to S. socotraensis.
Due to a lack of parsimony informative variation, the northern clade is a polytomy in the strict consensus tree. A single point mutation supports a clade of S. plumieri and S. socotraensis, and a different point mutation supports a clade of S. plumieri and the Hawaiian diploid clade. No base changes support the clade of S. plumieri and S. wrightii with the exclusion of the other two taxon groups. As with S. taccada, the monophyly of S. plumieri accessions (from Baja California to Senegal) argues against S. plumieri itself being the progenitor of multiple insular endemics.
The placement of S. nitida within this clade is not well supported by molecular data nor are there morphological characters that support this result. Therefore, S. nitida may be misplaced by these data (due to insufficient taxon sampling of Australian species), and a single, widely dispersed fleshy-fruited ancestor may have given rise to the northern clade (and perhaps also the Bornean species).
Colonization of the Hawaiian Islands
Fosberg (1947)
hypothesized a single origin for all Hawaiian Scaevola species. The morphological analysis by Patterson (1995)
with limited taxon sampling, on the other hand, supported at least two dispersal events with S. glabra clearly separate from the rest. Patterson (1995)
argued for a third introduction; however, this was not strongly supported by his data, which resolved S. taccada as sister to the endemic Hawaiian diploid taxa. Most workers (Carlquist, 1967
; Patterson, 1990
, 1995
; Sakai et al., 1995
) have proposed three separate introductions, which concurs with our molecular phylogeny. Wagner (1996)
suggested that S. hobdyi represented a fourth, separate introduction. In fact, because the single known specimen of S. hobdyi has anomalous vegetative features (narrowly linear, whorled leaves) and only a single floral bud, its relationships to Scaevola were initially overlooked. However, the ITS sequence from DNA extracted from the type specimen of S. hobdyi is identical to S. kilaueae, thus firmly placing it within the Hawaiian diploid clade and rejecting the four introductions hypothesis.
Hawaiian Scaevola represents immigrations from Australia (S. glabra), Polynesia (S. taccada), and possibly the Americas (the diploid endemic clade from Pacific coast S. plumieri). Molecular phylogenetic studies of other Hawaiian plant groups have found evidence of affinities with various geographic areas: Africa/Madagascar (Kokia [Seelanan et al., 1997
], Hesperomannia [Kim et al., 1998
]), the Americas (mints [Lindqvist and Albert, 2002
], silversword alliance [Baldwin et al., 1991
], Geranium [Pax et al., 1997
], Gossypium [Dejoode and Wendel, 1992
], Bidens [Ganders et al., 2000
], Rubus [Howarth et al., 1997
]), the Arctic (Viola [Ballard and Sytsma, 2000
]), Polynesia (Bidens [Ganders et al., 2000
], Metrosideros [Wright et al., 2000
], Pittosporum [Gemmill et al., 2001
], Tetraplasandra [Costello and Motley, 2001
], Tetramolopium [Lowrey, 1995
]), and Asia (Lobelioideae [Knox et al., 1993
]). It is perhaps not surprising, therefore, that Scaevola lineages appear to have dispersed into the Hawaiian Islands from three divergent localities.
Despite the fact that multiple introductions into the Hawaiian Islands have been hypothesized for other plant groups (e.g., Fosberg, 1947
; Sakai et al., 1995
), this case is the first in which three separate introductions in a single angiosperm genus have been supported with molecular data. Application of molecular data, while supporting some hypotheses, has refuted many others. The most notable cases where multiple introductions were hypothesized were the three genera comprising the Hawaiian silversword alliance (Argyroxiphium, Dubautia, and Wilkesia) and the six genera of the Hawaiian lobeliods (Brighamia, Cyanea, Clermontia, Delissia, Lobelia, and Trematolobelia) that are each now thought to be the result of single introduction events (Baldwin et al., 1991
; Givnish et al., 1996
). Furthermore, the three genera of Araliaceae, Tetraplasandra, Munroidendron, and Reynoldsia, are now presumed to be the result of one introduction event (Costello and Motley, 2001
), while alternatively, the two morphologically similar species of Hawaiian Rubus are now considered the result of two introductions, rather than one (Howarth et al., 1997
).
It is also interesting that neither of the Hawaiian endemic Scaevola lineages (S. glabra and the Hawaiian diploids) share affinities with any South Pacific or Polynesian species. In other Hawaiian genera with relatives in Polynesia (Metrosideros [Wright et al., 2000
], Pittosporum [Gemmill et al., 2001
], Tetraplasandra [Costello and Motley, 2001
], Tetramolopium [Lowrey, 1995
], and Bidens [Ganders et al., 2000
]), the Hawaiian groups appear to be closely related to their Polynesian cousins. Based on the present Scaevola distribution, it is startling that Hawaiian endemic Scaevola species are related to species in northern Australia and an American/African strand species, and not to the Polynesian species.
The Hawaiian Islands appear to have been colonized by both the southern clade (S. taccada) and the northern clade (the eight endemic diploids). Additionally, it appears that S. glabra originated from a third long-distance dispersal event. Therefore, despite their extreme isolation, the Hawaiian Islands have been colonized by three distinct lineages. Given that areas closer to Australia contain fewer lineages, the multiple colonizations of the Hawaiian Islands is surprising. One possible explanation is that, in Scaevola, niche availability is a greater barrier to ecological establishment than fruit dispersal. If so, it should not be surprising that more colonization could occur on the isolated but ecologically open Hawaiian Islands than in more proximate but more heavily occupied land masses. This hypothesis is supported by the fact that no new species of Scaevola have established themselves on continental areas outside Australia, despite the colonization by S. taccada and S. plumieri.
|
|
FOOTNOTES |
|---|
6 Author for reprint requests (dianella.howarth{at}yale.edu
)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Baldwin B. G. D. W. Kyhos J. Dvorak G. D. Carr 1991 Chloroplast DNA evidence for a North American origin of the Hawaiian silversword alliance (Asteraceae). Proceedings of the National Academy of Sciences, USA 88: 1840-1843
Baldwin B. G. M. J. Sanderson J. M. Porter M. F. Wojciechowski C. S. Campbell M. J. Donoghue 1995 The ITS region of nuclear ribosomal DNA: a valuable source of evidence on angiosperm phylogeny. Annals of the Missouri Botanical Garden 82: 247-277[CrossRef][ISI]
Ballard J. H. E. K. J. Sytsma 2000 Evolution and biogeography of the woody Hawaiian violets (Viola, Violaceae): arctic origins, herbaceous ancestry and bird dispersal. Evolution 54: 1521-1532[ISI][Medline]
Baum D. A. R. L. Small J. F. Wendel 1998 Biogeography and floral evolution of Baobabs (Adansonia, Bombacaceae) as inferred from multiple data sets. Systematic Biology 47: 181-207[CrossRef][ISI][Medline]
Carlquist S. 1965 Island life: a natural history of the islands of the world. The Natural History Press, Garden City, New York, USA
Carlquist S. 1967 The biota of long distance dispersal. V. Plant dispersal to Pacific Islands. Bulletin of the Torrey Botanical Club 94: 129-162[CrossRef][ISI]
Carlquist S. 1969 Wood anatomy of Goodeniaceae and the problem of insular woodiness. Annals of the Missouri Botanical Garden 56: 358-390[CrossRef][ISI]
Carlquist S. 1974 Island biology. Columbia University Press, New York, New York, USA
Carlquist S. 1980 Hawaii, a natural history. Pacific Tropical Botanical Garden, Lawai, Hawai‘i, USA
Carolin R. C. 1966 Seeds and fruit of the Goodeniaceae. Proceedings of the Linnean Society of New South Wales 91: 58-83
Carolin R. C. 1978 The systematic relationships of Brunonia. Brunonia 1: 9-29
Carolin R. C. 1990 Nomenclatural notes, new taxa and the systematic arrangement in the genus Scaevola (Goodeniaceae) including synonyms. Telopea 3: 477-516
Carolin R. C. D. A. Morrison T. Rajput 1992 Flora of Australia volume 35, Brunoniaceae, Goodeniaceae. Australian Government Publishing Service, Canberra, Australian Capital Territory Australia
Carr G. D. 1998 Chromosome evolution and speciation in Hawaiian flowering plants. In T. F. Stuessy and M. Ono [eds.], Evolution and speciation of island plants, 5–47. Cambridge University Press, Cambridge, UK
Costello A. T. J. Motley 2001 Molecular systematics of Tetraplasandra, Munroidendron and Reynoldsia sandwicensis (Araliaceae) and the evolution of superior ovaries in Tetraplasandra. Edinburgh Journal of Botany 58: 229-242[CrossRef]
Dejoode D. R. J. F. Wendel 1992 Genetic diversity and origin of the Hawaiian Islands cotton, Gossypium tomentosum. American Journal of Botany 79: 1311-1319[CrossRef][ISI]
Doyle J. J. J. L. Doyle 1987 A rapid DNA isolation procedure for small quantities of fresh leaf material. Phytochemical Bulletin 19: 11-15
Felsenstein J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791[CrossRef][ISI]
Fitch W. M. 1971 Toward defining course of evolution: minimum change for a specific tree topology. Systematic Zoology 20: 406-416[CrossRef][ISI]
Fosberg F. R. 1947 Derivation of the flora of the Hawaiian Islands. In E. C. Zimmerman [ed.], Insects of Hawaii, vol. 1, 107–119. University of Hawai‘i Press, Honolulu, Hawai‘i, USA
Ganders F. R. M. Berbee M. Pirseyedi 2000 ITS base sequence phylogeny in Bidens (Asteraceae): evidence for the continental relatives of Hawaiian and Marquesan Bidens. Systematic Botany 25: 122-133[CrossRef][ISI]
Gemmill C. E. C. G. J. Allan W. L. Wagner E. A. Zimmer 2001 Evolution of insular Pacific Pittosporum (Pittosporaceae): origin of the Hawaiian radiation. Molecular Phylogenetics and Evolution 22: 31-42[CrossRef][ISI]
Givnish T. J. E. Knox T. B. Patterson J. R. Hapeman J. D. Palmer K. J. Sytsma 1996 The Hawaiian lobelioides are monophyletic and underwent a rapid initial radiation roughly 15 million years ago. American Journal of Botany 83: 159 (Abstract)
Guppy H. B. 1906 Observations of a naturalist in the Pacific between 1896 and 1899, Plant dispersal. Macmillan, London, UK
Guppy H. B. 1917 Plants, seeds and currents in the West Indies and Azores. Williams and Norgate, London, UK
Gustafsson M. H. G. A. Backlund B. Bremer 1996 Phylogeny of the Asterales sensu lato based on rbcL sequences with particular reference to the Goodeniaceae. Plant Systematics and Evolution 199: 217-242[CrossRef][ISI]
Gustafsson M. H. G. E. Grafstrom S. Nilsson 1997 Pollen morphology of the Goodeniaceae and comparisons with related families. Grana 36: 185-207[ISI]
Hasegawa M. H. Kishino T. Yano 1985 Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution 22: 160-174[CrossRef][ISI][Medline]
Howarth D. G. D. E. Gardner C. W. Morden 1997 Phylogeny of Rubus subgenus Idaeobatus (Rosaceae) and its implication toward colonization of the Hawaiian Islands. Systematic Botany 22: 433-441[CrossRef][ISI]
Kårehed J. J. Lundberg B. Bremer K. Bremer 1999 Evolution of the Australasian families Alseuosmiaceae, Argophyllaceae and Phellinaceae. Systematic Botany 24: 660-682[CrossRef][ISI]
Kim H.-G. S. C. Keeley P. S. Vroom R. K. Jansen 1998 Molecular evidence for an African origin of the Hawaiian endemic Hesperomannia (Asteraceae). Proceedings of the National Academy of Sciences, USA 95: 15440-15445
Kishino H. M. Hasegawa 1989 Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in Hominoidea. Journal of Molecular Evolution 29: 170-179[CrossRef][ISI][Medline]
Knox E. B. S. R. Downie J. D. Palmer 1993 Chloroplast genome rearrangements and the evolution of giant lobelias from herbaceous ancestors. Molecular Biology and Evolution 10: 414-430[ISI]
Leenhouts P. W. 1957 Goodeniaceae. Flora Malesiana ser. 1 5: 335-344
Lesko G. L. R. B. Walker 1969 Effect of seawater on seed germination of two Pacific atoll beach species. Ecology 50: 730-734[CrossRef][ISI]
Lindqvist C. V. A. Albert 2002 Origin of the Hawaiian endemic mints within North American Stachys (Lamiaceae). American Journal of Botany 89: 1709-1724
Lowrey T. K. 1995 Phylogeny, adaptive radiation, and biogeography of Hawaiian Tetramolopium (Asteraceae, Astereae). In W. L. Wagner and V. A. Funk [eds.], Hawaiian biogeography, evolution on a hot spot archipelago, 195–220. Smithsonian Institution Press, Washington, D.C. USA
Morden C. W. V. Caraway T. J. Motley 1996 Development of a DNA library for native Hawaiian plants. Pacific Science 50: 324-335
Oliver W. R. B. 1910 The vegetation of the Kermadec Islands. Transactions and Proceedings of the New Zealand Institute 42: 118-175
Patterson R. 1990 Goodeniaceae. In W. L. Wagner, D. R. Herbst, and S. H. Sohmer [eds.], Manual of the flowering plants of Hawai‘i. University of Hawai‘i Press and Bernice P. Bishop Museum Press, Honolulu, Hawai‘i, USA
Patterson R. 1995 Phylogenetic analysis of Hawaiian and other Pacific species of Scaevola (Goodeniaceae). In W. L. Wagner and V. A. Funk [eds.], Hawaiian biogeography: evolution on a hot spot archipelago, 363–378. Smithsonian Institution Press, Washington, D.C., USA
Pax D. L. R. A. Price H. J. Michaels 1997 Phylogenetic position of the Hawaiian geraniums based on rbcL sequences. American Journal of Botany 84: 72-78[Abstract]
Peacock W. J. 1963 Chromosome numbers and cytoevolution in the Goodeniaceae. Proceedings of the Linnean Socienty of New South Wales 88: 8-27
Sakai A. K. W. L. Wagner D. M. Ferguson D. R. Herbst 1995 Origins of dioecy in the Hawaiian flora. Ecology 76: 2517-2529[CrossRef][ISI]
Seelanan T. A. Schnabel J. F. Wendel 1997 Congruence and consensus in the cotton tribe (Malvaceae). Systematic Botany 22: 259-290[CrossRef][ISI]
St. John H. 1962 A new Scaevola (Goodeniaceae) from Socotra Island. Webbia 17: 45-48
Struwe L. M. Thiv J. W. Kadereit A. S. R. Pepper T. J. Motley P. White J. H. E. Rova K. Potgieter V. A. Albert 1998 Saccifolium (Saccifoliaceae), an endemic of Sierra de la Neblina on the Brazilian-Venezuelan border, is related to a temperate alpine lineage of Gentianaceae. Harvard Papers in Botany 3: 199-214
Swofford D. L. 2001 PAUP*: phylogenetic analysis using parsimony (*and other methods). Sinauer, Sunderland, Massachusetts, USA
Sykes W. R. 1998 Scaevola gracilis (Goodeniaceae) in the Kermadec Islands and Tonga. New Zealand Journal of Botany 36: 671-674[ISI]
Templeton A. R. 1983 Phylogenetic inference from restriction endonuclease cleavage site maps with particular reference to the evolution of humans and the apes. Evolution 37: 221-244[CrossRef][ISI]
Thompson J. D. D. G. Higgins T. J. Gibson 1994 Clustal W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Research 22: 4673-4680
Urbatsch L. E. B. G. Baldwin M. J. Donoghue 2000 Phylogeny of the coneflowers and relatives (Heliantheae: Asteraceae) based on nuclear rDNA internal transcribed spacer (ITS) sequences and chloroplast DNA restriction site data. Systematic Botany 25: 539-565[CrossRef][ISI]
van Balgooy M. M. J. 1975 Pacific plant areas 3, Rijksherbarium, Leiden, Netherlands
von Mueller F. 1859 Goodeniaceae volume 9. Fragmenta Phytographiae Australiae 1: 203-207
Wagner W. L. 1996 Scaevola hobdyi (Goodeniaceae), an enigmatic new species from West Maui, Hawaiian Islands. Novon 6: 225-227[CrossRef]
White T. J. T. Birns S. Lee J. Taylor 1990 Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In M. Innis, D. Gelfand, J. Sninsky, and T. White [eds.], PCR protocols: a guide to methods and applications, 315–322. Academic Press, San Diego, California, USA
Wright S. D. C. G. Yong J. W. Dawson D. J. Whittaker R. C. Gardner 2000 Riding the ice age El Nino? Pacific biogeography and evolution of Metrosideros subg. Metrosideros (Myrtaceae) inferred from nuclear ribosomal DNA. Proceedings of the National Academy of Sciences, USA 97: 4118-4123
Ziegler A. C. 2002 Hawaiian natural history, ecology, and evolution. University of Hawai‘i Press, Honolulu, Hawai‘i, USA
This article has been cited by other articles:
|
|
 |
|
  H. E. Driscoll and D. S. Barrington Origin of Hawaiian Polystichum (Dryopteridaceae) in the context of a world phylogeny Am. J. Botany, August 1, 2007; 94(8): 1413 - 1424. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. T. Harbaugh and B. G. Baldwin Phylogeny and biogeography of the sandalwoods (Santalum, Santalaceae): repeated dispersals throughout the Pacific Am. J. Botany, June 1, 2007; 94(6): 1028 - 1040. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Eggens, M. Popp, M. Nepokroeff, W. L. Wagner, and B. Oxelman The origin and number of introductions of the Hawaiian endemic Silene species (Caryophyllaceae) Am. J. Botany, February 1, 2007; 94(2): 210 - 218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. C. B. Cronk, M. Kiehn, W. L. Wagner, and J. F. Smith Evolution of Cyrtandra (Gesneriaceae) in the Pacific Ocean: the origin of a supertramp clade Am. J. Botany, June 1, 2005; 92(6): 1017 - 1024. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. L. Clement, M. C. Tebbitt, L. L. Forrest, J. E. Blair, L. Brouillet, T. Eriksson, and S. M. Swensen Phylogenetic position and biogeography of Hillebrandia sandwicensis (Begoniaceae): a rare Hawaiian relict Am. J. Botany, June 1, 2004; 91(6): 905 - 917. [Abstract] |
