| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Article |
Jepson Herbarium and Department of Integrative Biology, 1001 Valley Life Sciences Building #2465, University of California, Berkeley, California 94720-2465 USA
Received for publication May 25, 2006. Accepted for publication December 6, 2006.
ABSTRACT
Phylogenetic analyses of nuclear rDNA transcribed spacers and cytogenetic studies of interspecific hybrids reported here uphold Carlquist's hypothesis (1965
, Island Biology) that shrubby tarweeds (Deinandra) of Guadalupe Island, Mexico, are products of in situ radiation in the California Islands, where evidence of plant diversification has been equivocal. Based on the rDNA findings, the Guadalupe Island endemics (D. frutescens, D. greeneana subsp. greeneana, and D. palmeri) constitute a clade that arose since the late Pliocene, well after the origin of Guadalupe Island and diversification of annual, mainland Californian lineages of Deinandra. High interfertility and normal meiosis in F1 hybrids between the three endemics contrast with reduced interfertility (to complete intersterility) and meiotic irregularities in F1 hybrids between other, mostly mainland species of Deinandra. Cloned rDNA sequences provided no convincing evidence of introgression among the Guadalupe Island deinandras; morphological, phenological, and/or habitat differences among those taxa indicate ecological barriers to gene flow and a probable role for ecological divergence in diversification. Biosystematic and molecular phylogenetic data for shrubby tarweeds of Guadalupe Island and another secondarily woody, oceanic-island tarweed lineage, the Hawaiian silversword alliance, reveal strikingly similar evolutionary histories. Both groups violate Baker's Rule by stemming from self-incompatible ancestors in western North America, and each has undergone within-island diversification without evolution of strong sterility barriers among lineages. Evolutionary parallels between these Hawaiian and California Island lineages of Madiinae, first suggested by Carlquist, may reflect characteristics of tarweeds that facilitate insular colonization and adaptive radiation.
Key Words: adaptive radiation • Baker's Rule • biogeography • California Islands • Compositae • Hawaii • phylogeny • speciation
Island floras contain extraordinary examples of autochthonous evolution and paleo-endemism (Carlquist, 1965
, 1974
; Stuessy and Ono, 1998
), although the importance of in situ diversification vs. persistence of relict lineages is still imperfectly understood and controversial for some archipelagos or island groups (see Carlquist, 1995
; Baldwin et al., 1998
), especially those close to continents. The flora of the California Islands (i.e., the Channel Islands of southern California and the Pacific islands of Baja California, Mexico; Fig. 1) is well known for examples of woody endemics (e.g., Lyonothamnus A. Gray) that are represented only as fossils on the nearby Californian mainland, where changing climatic conditions since the late Tertiary have been implicated in loss of plant lineages that were able to persist in relatively equable, insular environments (Axelrod, 1958
, 1967
; Raven and Axelrod, 1978
). Evidence for plant diversification on the California Islands is limited and equivocal (see Raven, 1963
; Ray, 1995
; Davis, 1997
), especially in light of various insular species with characteristics commonly associated with island evolution (e.g., arborescence or woody stems) that also have narrow distributions in climatically similar areas along the mainland coast of California or Baja California, Mexico (Raven, 1963
, 1967
; Wallace, 1985
; Moran, 1996
). As noted by Thorne (1969)
, such plants might not have been questioned as examples of insular evolution if mainland populations had been extirpated prior to documentation, in the absence of fossils. Even if isolated endemics can be shown to have evolved on the California Islands, the question remains as to whether such divergence differs in any qualitative way from divergence elsewhere in California, where lineages endemic to the islands often have close relatives with similarly restricted distributions (see Hickman, 1993
). Examples of insular adaptive radiation, i.e., rapid diversification marked by ecological shifts, would provide the best evidence for evolutionary change associated with island conditions (see Schluter, 2000
).
|
suggested that shrubby tarweeds (Compositae–Madiinae) of the California islands now treated in Deinandra Greene (previously in Hemizonia DC.), represent an incipient or limited adaptive radiation that parallels, on a small scale, rapid ecological diversification of another insular group in Madiinae, the Hawaiian silversword alliance (Argyroxiphium DC., Dubautia Gaudich., and Wilkesia A. Gray; see Carlquist et al., 2003
). The well-studied silversword alliance has been regarded as "...one of the most remarkable examples of adaptive radiation in plants" (Raven et al., 2005
, p. 210). A closely related example of island diversification, in Deinandra, could be valuable for understanding whether members of Madiinae share characteristics that result in similar evolutionary propensities under insular situations, even in relatively near-shore island settings compared to Hawaii. Five shrubby or facultatively perennial taxa of Deinandra occur exclusively on the California Islands; the other 19 taxa in the genus are mostly annuals that all occur in mainland California, with three species also found on the California Islands (see Baldwin, 2003a
).
Three of the shrubby taxa of Deinandra (D. frutescens, D. greeneana subsp. greeneana, and D. palmeri) are known only from the highest and most remote of the California Islands, Guadalupe Island (GI; Fig. 1), an oceanic seamount situated 260 km west of the Baja California peninsula and reaching 1295 m above sea level. Compared to the other California Islands, GI has a disharmonic flora of relatively low native diversity (ca. 156 species) for its area (250 km2) and a relatively high degree of single-island endemism—21.8% of species and infraspecific taxa (Moran, 1996
; also see Raven, 1963
, 1967
; Moran et al., 1967
; Thorne, 1969
; Wallace, 1985
).
Considering the great isolation, height, size, age (7 ± 2 million yr; Engel and Engel, 1970
), and level of floristic endemism of GI, the potential for discovering examples of diversification there may be greater than on other California Islands. Deinandra, with more endemic taxa on the island than any other vascular-plant genus (Moran, 1996
), has appeared to be especially promising as a candidate example of in situ radiation (Carlquist, 1965
). Thorne (1969)
, who appreciated the merits of Carlquist's (1965)
hypothesis of adaptive radiation of shrubby tarweeds in the California Islands, raised a concern that the plants may not have evolved in an insular setting, based on Moran's (1969) documentation of potentially relict populations of one of the three GI species, D. greeneana, on the mainland of Baja California (i.e., D. greeneana subsp. peninsularis). Here, I revisit Carlquist's (1965)
hypothesis of adaptive radiation in the California Islands from molecular phylogenetic and cytogenetic perspectives, with special attention to evolutionary origins of the endemic taxa of Deinandra on GI.
MATERIALS AND METHODS
Molecular analyses
Total DNA was extracted from rosette leaves of 1–10 individuals from each of the 21 species in Deinandra and two outgroup species of the most closely related genus, Holocarpha Greene, using a modification of Doyle and Doyle's (1987)
method (see Baldwin and Wessa, 2000
). Two geographically separated populations were sampled for most mainland species of Deinandra and for the only insular species occurring on more than one major island, D. clementina. Both subspecies of D. greeneana were included. Collection and voucher data are presented in the Appendix. Amplifications and sequencing of the 18S–26S rDNA internal transcribed spacer (ITS) region (ITS-1, 5.8S gene, and ITS-2), with the use of primer ITS5 (White et al., 1990
) rather than of ITS-I for sequencing most PCR products, and a segment of the external transcribed spacer (ETS) upstream of the 18S gene followed published methods (Baldwin and Markos, 1998
; Baldwin and Wessa, 2000
). ITS and ETS amplification products for the three GI taxa and other perennials were cloned using the zero blunt TOPO cloning kit (Invitrogen, Carlsbad, California, USA), following the manufacturer's protocol.
ITS and ETS sequences were aligned manually, without uncertainty (see the Appendix for GenBank accession numbers). Only one representative of each set of identical sequences was included in sequence matrices (deposited in TreeBASE, http://www.treebase.org, with trees). Phylogenetic analyses using maximum likelihood (ML) and maximum parsimony (MP) were conducted using PAUP* 4.0 (Swofford, 2002
), beta version 10, with 100 random-addition sequences for each heuristic search. Molecular evolutionary models for ML analyses (TrN+G for all data sets; settings archived in TreeBASE) were determined by hierarchical likelihood-ratio (LR) tests using Modeltest (Posada and Crandall, 1998
), version 3.6. Reliability of clades was assessed in PAUP* using nonparametric bootstrapping, with ML and MP (100 "full heuristic" replicates; 10 random-addition sequences per replicate). Alternative hypotheses of relationship were assessed using the Shimodaira–Hasegawa (SH) test (Shimodaira and Hasegawa, 1999
; see Goldman et al., 2000
), with resampling estimated log-likelihood (RELL) optimization and 1000 bootstrap replicates (one-tailed test), using PAUP*.
Rate constancy of molecular evolution across lineages was tested using a tree-wide LR test (Felsenstein, 1988
), as implemented in PAUP*. In the rate-constant ITS tree, a maximum-age calibration of 15 million yrs ago (Ma) for the tree node representing the most recent common ancestor (MRCA) of Deinandra was based on the assumption that diversification of the genus, with all species endemic to the summer-dry California Floristic Province or adjacent deserts, would not have preceded the onset of summer drying in western North America at mid-Miocene (e.g., Axelrod, 1992
; Flower and Kennett, 1994
; see Baldwin and Sanderson, 1998
). Although 15 Ma probably precedes the MRCA of Deinandra, which is nested within one of four major lineages of the primarily Californian, tarweed clade (Baldwin, 2003b
), use of that date allowed for placement of an outer bound on the estimated age of the GI taxa, in lieu of fossils for Madiinae. Based on the external calibration at 15 Ma, maximum nodal ages were estimated from ML branch lengths using r8s, version 1.70 (Sanderson, 2003
), and compared to nodal ages estimated using the minimum and maximum ITS nucleotide substitution rates reported for other angiosperms (Kay et al., 2006
). Standard errors on maximum divergence-time estimates were obtained using a nonparametric bootstrap procedure (see Baldwin and Sanderson, 1998
). Diversification rate was calculated using the simple estimator [ln(N) – ln(N0)]/T, with N0 = initial diversity (2, for crown group), N = standing diversity, and T = inferred crown-group age (Stanley, 1979
; Rosenzweig and Vetault, 1992
).
Evolution of vegetative persistence (annual vs. perennial habit) and ability or inability to undergo self-fertilization was examined by mapping character states onto MP and ML trees in MacClade 4.0 (Maddison and Maddison, 2000
), using parsimony. Data on habit and selfing ability for each taxon were obtained from Tanowitz (1982)
, Baldwin (2003a)
, and B. G. Baldwin and E. A. Friar (Rancho Santa Ana Botanic Garden) (unpublished data).
Cytogenetic analyses
Reciprocal crosses were made in all pairwise combinations between each of the insular shrubby taxa (i.e., D. clementina, Channel Islands; D. frutescens, D. greeneana subsp. greeneana, and D. palmeri, GI) and between each of the insular shrubs and one of the two mainland shrubs (D. minthornii, southern California). The other mainland shrub, D. greeneana subsp. peninsularis (Baja California, Mexico), was crossed with one of the GI species, D. frutescens. Parental plants were grown from field-collected stem cuttings or seeds (embryos excised and germinated on wet filter paper under continuous light) in ca. equal amounts potting mix and pumice or sand in a growth chamber (13–16 h light, 18–21°C; 8–11 h dark, 11°C) or in U.C. greenhouses. Crosses were performed by rubbing flowering heads together; all of the taxa are self-incompatible, as established in part earlier (Keck, 1959
; Tanowitz, 1985
) and confirmed by absence of seed set in unmanipulated heads and absence of non-hybrid progeny from crosses. Conditions for cultivating F1 hybrids were similar to those for the parental individuals. One apparent F1 hybrid between sympatric D. palmeri and D. greeneana subsp. greeneana was grown from a field-collected seedling.
Floral buds of hybrids for meiotic analyses were fixed for 5 d in 6 parts chloroform, 3 parts 95% ethanol, 1 part glacial acetic acid. Chromosomal associations were examined from 40–150+ squashed microsporocytes (at diakinesis and meotic metaphase I) in acetocarmine mixed with Hoyer's solution (Beeks, 1955
) at 630-1000x magnification using phase microscopy. Fertilities of hybrids were estimated from pollen stainability in aniline blue dye in lactophenol (Maneval, 1936
) based on a sample of at least 200 pollen grains per plant. Vouchers of hybrids were deposited in the Jepson Herbarium (JEPS).
Mean pollen stainabilities for hybrids of each crossing combination involving the insular endemics and D. minthornii were analyzed using distance methods to fit trees to estimates of interfertility. Neighbor-joining (Saitou and Nei, 1987
) and Fitch and Margoliash's (1967)
method, as implemented in PHYLIP 3.6 (Felsenstein, 2004
) as NEIGHBOR and FITCH, were used to generate trees without the constraint of evolutionary rate constancy. UPGMA (unweighted pair group method using arithmetic averages) clustering (in NEIGHBOR) and a modified Fitch-Margoliash approach (KITSCH) were also used in PHYLIP to generate topologies under the assumption of rate-constant evolution.
RESULTS
rDNA polymorphism
Direct sequencing of pooled PCR products of the ETS and ITS regions from individual samples of Deinandra and both outgroup taxa yielded little evidence of within-site polymorphism except in sequences of each of the three Guadalupe Island (GI) taxa (D. frutescens, D. greeneana subsp. greeneana, and D. palmeri), which had a moderate frequency of strong double-peaks in electropherograms of both spacer regions. Cloning of ETS and ITS sequences verified that the double peaks reflect divergent repeats in both spacer regions, with 37 sites polymorphic across the three GI taxa, 18 of those sites having polymorphism shared by two or three of the GI taxa, and 12 of the shared polymorphic sites having one unique nucleotide state for that site (i.e., a state not found at that site across the other sequences of Deinandra and the outgroup).
Analyses of cloned sequences
Of 52 cloned ETS and ITS-region sequences obtained for the three GI taxa, 26 were unique and included as operational taxonomic units in phylogenetic analyses. For the ITS region, 3–5 unique sequences (differing by at least one substitution or insertion/deletion) were recovered from the set of (7–9) clones in each GI taxon. The ML and MP trees from analysis of ITS data resolved two moderately well-supported lineages of clones that each included sequences from all three GI taxa (Fig. 2); none of the clones of GI taxa was placed elsewhere in the trees. One of those clone lineages (from "locus 1") provided robust resolution of relationships among the GI endemics. Evidence of recombination among cloned sequences was lacking; based on the topology in Fig. 2 and on accelerated-transformation (ACCTRAN) mapping of character-states, only two of 23 nucleotide substitutions in the two lineages of clones occurred in parallel and conceivably could be explained by recombination. Inability to reject rate-constant evolution for the ITS data set (see Divergence time and diversification rate estimates) and lack of substitutions in the 5.8S subunit outside the relatively variable 3' end provide no evidence that either of the two ITS lineages of GI taxa in Fig. 2 represents an inactive rDNA locus (pseudogene).
|
Subsequent phylogenetic analyses that included 10–18 ITS clones each from samples of the other perennials (D. clementina, D. greeneana subsp. peninsularis, D. martirensis, and D. streetsii) did not alter the results shown in Fig. 2. All clones from each of those four taxa constituted a clade of identical phylogenetic position to the sequence obtained from pooled PCR products of the same taxon (results not shown).
Analyses of ETS sequences (including clones) resolved an internally weakly resolved clade comprising all ETS clones of the three GI taxa; in the ML tree, two ETS lineages, each including clones of all three GI taxa, were (weakly) resolved within the primary GI lineage (results not shown).
Monophyly of D. greeneana (i.e., subsp. greeneana + subsp. peninsularis) was rejected for ETS data based on SH-test results, either with all cloned sequences of D. greeneana subsp. greeneana in a clade with D. greeneana subsp. peninsularis (P = 0.003) or with the sequence of D. greeneana subsp. peninsularis united with one of the two sets of ETS sequences of D. greeneana subsp. greeneana (clone 15) (P = 0.012).
ETS+ITS analyses
Phylogenetic analysis of ETS+ITS-region sequences of Deinandra and outgroup taxa using ML and MP criteria (treating all sites that varied across clones of the same sample as polymorphic) consistently resolved a clade comprising the GI taxa, with 63% ML and MP bootstrap support (results not shown). One of the two MP trees resolved the Channel Islands endemic, D. clementina, as sister to the three GI taxa (D. frutescens, D. greeneana subsp. greeneana, and D. palmeri); the relationship between D. clementina and the GI taxa was unresolved in the other MP tree. Maximum-parsimony and ML bootstrap support for basal tree structure (i.e., deep clades) and for relationships among the southern annuals was higher in general than support for clades of northern annuals and shrubs in Deinandra. Non-identical sequences from different populations of the same species constituted well-supported clades in the ETS+ITS trees except for the two subspecies of D. greeneana, which were placed in different clades.
Divergence time and diversification rate estimates
Rate constancy of molecular evolution in the ITS region could not be rejected for Deinandra. A maximal calibration of the base of the ITS tree at 15 Ma (see Materials and Methods) yielded two maximum-age estimates for the most recent common ancestor of the GI taxa: 1.9 ± 0.6 Ma, from "locus 1," and 0.6 ± 0.3 Ma, from "locus 2" (Fig. 2). The r8s-estimated rate of ITS1+ITS2 evolution is 3.37 x 10–9 substitutions/site/year (sub/site/yr) for Deinandra under the 15 Ma calibration; the range of divergence times for the GI clade obtained by applying ITS evolutionary rates from other angiosperms (0.38–8.34 x 10–9 sub/site/yr; Kay et al., 2006
) is 22.1–0.12 Ma. Much narrower ranges of divergence times for the GI clade were obtained using Kay et al.'s (2006) estimated rates from lineages of herbs (4.9–0.12 Ma), the ancestral habit in Deinandra (see next section), or from other Compositae (3.37– 0.13 Ma). Rate-constancy of ETS evolution across lineages was rejected for the ETS trees (not shown), which were not used to estimate divergence times in Deinandra. The diversification rate for the GI clade was estimated at 0.21 species per million years, using the calibration-based age for ITS "locus 1" of 1.9 Ma (Fig. 2).
Character evolution
Annual habit was reconstructed as ancestral in Deinandra based on mapping of persistence on ITS, ETS, and ITS+ETS trees under the parsimony criterion. One to three origins of perenniality in Deinandra were resolved on the same trees, with perenniality ancestral in the GI clade. As expected, self-incompatibility (SI) was resolved as ancestral in Deinandra, with separate origins or (in some ETS trees) possibly one origin of self-compatibility (SC) accounting for SC of D. arida and D. mohavensis. No secondary origins of SI were resolved except as an equivocal outcome (in D. paniculata) in some ETS trees.
Cytogenetic findings
Vigorous F1 hybrids were produced in all crossing combinations attempted (Table 1). Pollen stainabilities (
fertilities) of hybrids were <50% except for all hybrids between the GI taxa (
= 74.0–99.8%) and hybrids between D. clementina and D. greeneana subsp. greeneana ( = 65.3%). For all hybrids, the modal chromosomal association at diakinesis or metaphase I was 12 bivalents. Hybrids with reduced chromosome pairing all involved one of the two mainland shrubs, i.e., D. greeneana subsp. peninsularis or D. minthornii (see Table 1).
|
DISCUSSION
Recent diversification on Guadalupe Island
Carlquist's (1965)
hypothesis that shrubby tarweeds provide an example of plant diversification in the California Islands is upheld by resolution of a young, monophyletic group comprising the three endemic, Guadalupe Island (GI) taxa of Deinandra in all rDNA trees from ITS, ETS, and combined ETS+ITS analyses. Results from analyses of ITS sequences (including clones) yielded two lines of support for monophyly of the GI group (Fig. 2), in accord with the ETS trees (not shown); evidently, two major rDNA loci of the GI taxa have been diverging throughout the history of the GI lineage. (Such rDNA variation in the GI taxa could easily have been misinterpreted as evidence for hybridization if direct sequences of pooled PCR products rather than clones had been examined).
Calibration-based estimates for onset of diversification of the GI taxa since the late Pliocene or Pleistocene (Fig. 2)—well after the late-Miocene (7 ± 2 Ma) origin of GI (Engel and Engel, 1970
)—are consistent with estimates of divergence time based on published rates of ITS evolution for other angiosperms. The r8s-estimated ITS1+ITS2 evolutionary rate of 3.37 x 10–9 sub/site/yr for Deinandra under the 15 Ma basal calibration falls well within the ranges of rates of ITS nucleotide substitution reported for other flowering plants, for herbaceous angiosperm lineages, or for Compositae alone (Kay et al., 2006
). The range of lineage-divergence times for Deinandra obtained by applying ITS evolutionary rates from herbaceous angiosperm lineages in general (4.9–0.12 Ma) or from other Compositae (3.37–0.13 Ma) closely flanks the calibration-based estimates. The comparison to herbaceous angiosperm lineages is appropriate for Deinandra because the basal condition in the genus is unequivocally herbaceous and no shifts in ITS rates accompanied evolution of woodiness, just as in the "Madia" lineage (Baldwin, 1996
; Baldwin and Sanderson, 1998
; see Kay et al., 2006
).
Use of a 15-Ma basal tree-calibration for the "Madia" lineage (including the Hawaiian silversword alliance) yielded an ITS evolutionary rate (3.00 x 10–9 sub/site/yr) nearly identical to the rate obtained with the same basal calibration of Deinandra (Baldwin and Sanderson, 1998
; Kay et al., 2006
). That calibration of the "Madia" lineage yielded divergence-time estimates for single-island endemic Hawaiian-silversword-alliance lineages that were consistent with (i.e., about equal to or younger than) island ages (Baldwin and Sanderson, 1998
). Use of a 15-Ma basal tree-calibration for another (annual) group in Madiinae, the genus Layia (Baldwin, 2005
), yielded a higher ITS substitution rate (7.67 x 10–9 sub/site/yr, not reported previously) that also falls within the ranges of rates for other herbaceous angiosperm lineages or for Compositae (Kay et al., 2006
); when applied to the rate-constant ITS tree for Layia, those ranges of rates from other plants yielded nodal divergence times within Layia that include the calibration-based estimates, as in Deinandra, e.g., ca. 0.6 Ma for L. discoidea, not younger dates, as reported earlier (Baldwin, 2005
). More recent divergence times, and higher rates of ITS evolution, for Deinandra, the "Madia" lineage, and Layia cannot be ruled out, based on the use of a calibration date that defensibly could be applied to a deeper node in subtribe Madiinae or tribe Madieae (see Baldwin et al., 2002
; Baldwin, 2003b
).
Interfertility and phylogeny
Judging from results of molecular phylogenetic and pollen-stainability analyses, time since divergence from a common ancestor explains levels of species interfertility estimated here from hybrid pollen, i.e., high interfertility of the recently diverged GI taxa vs. reduced fertility (and in some instances reduced meiotic chromosomal pairing) of hybrids between GI taxa and more distantly related perennials from elsewhere (Table 1). The common tree found from neighbor-joining and Fitch–Margoliash analyses of pollen-stainability data for the woody deinandras is completely congruent with placement of those taxa in the rDNA trees (Fig. 2). The other optimal Fitch–Margoliash tree and the topologically congruent UPGMA and ultrametric Fitch-Margolish (KITSCH) results differed in pattern from the rDNA trees only within the GI clade, wherein branch lengths from pollen data are extremely short. Generally lower levels of interfertility or lack of crossability between annual species or between annuals and perennials of Deinandra (Clausen, 1951
; Tanowitz, 1982
) may reflect more ancient divergence of some annual taxa (Fig. 2) and/or more rapid evolution of sterility barriers in annuals compared to perennials (Archibald et al., 2005
). The ability to predict phylogeny from interfertility data is not expected (given that interfertility is symplesiomorphic) and has shown better correlation with estimates of genetic similarity than with phylogeny in Hawaiian Schiedea (Weller et al., 2001
).
Deinandra greeneana: relict or neoendemic?
Occurrence of D. greeneana on GI (D. greeneana subsp. greeneana) and on the Baja California peninsula and the near-shore Todo Santos Islands (D. greeneana subsp. peninsularis) now appears to reflect a minor taxonomic problem rather than evidence of relict status for any shrubby tarweeds on GI, as justifiably concerned Thorne (1969)
and Carlquist (1974)
. In the rDNA trees, sequences of D. greeneana subsp. peninsularis ("peninsularis" in Fig. 2) were placed outside the GI lineage, without any conflicting signal that could conceivably unite D. greeneana subsp. peninsularis with any of the GI taxa, including D. greeneana subsp. greeneana ("greeneana" in Fig. 2). Monophyly of D. greeneana also was rejected based on results of the conservative SH test.
Reduced pollen fertility and (in a minority of cells) reduced chromosomal pairing in a hybrid between D. greeneana subsp. peninsularis and the GI species D. frutescens contrasts with normal pollen fertility and chromosomal pairing in hybrids between GI taxa (Table 1) and is consistent with molecular evidence against an especially close relationship of D. greeneana subsp. peninsularis to any single insular taxon, including D. greeneana subsp. greeneana. Moran (1969)
noted differences between D. greeneana subsp. greeneana and D. greeneana subsp. peninsularis in growth form, climatic setting, and characteristics of stems, leaves, flowers, and fruits; based on results presented here, the two taxa should be treated as distinct species (B. Baldwin, in press
). The possibility that D. greeneana subsp. peninsularis is the closest living relative of the Guadalupe Island clade remains tenable and could not be rejected based on SH test results.
Ecological diversity
In addition to evidence for rapid diversification, the other major criterion for adaptive radiation—ecological change associated with evolutionary divergence (see Schluter, 2000
)—also appears to be fulfilled by the shrubby tarweeds of GI based on morphological, phenological, habitat, and biogeographic considerations. Deinandra frutescens is known only from the northern end of the island at mid to high elevations, close to where GI pine [Pinus radiata D. Don var. binata (Engelm.) Lemmon] and GI cypress [Callitropsis guadalupensis (S. Watson) D. P. Little subsp. guadalupensis] groves occur and where climatic conditions have been inferred to be much wetter than at the desertic, south end of the island and on offshore islets, where D. greeneana subsp. greeneana and D. palmeri are endemic and grow intermixed as major elements of scrub vegetation (Moran, 1996
). The three taxa are conspicuously different in life form (Fig. 1), both in nature and in cultivation under common conditions: D. frutescens is an openly branching, ascending shrub, D. greeneana subsp. greeneana has a Medusa-head-like appearance, with a mounded center and radiating, prostrate side branches, and D. palmeri is a low-growing shrub with a flattened aspect. Based on the ITS data (Fig. 2), sympatry between D. greeneana subsp. greeneana and D. palmeri, at the south tip of the island, is likely secondary; the montane D. frutescens and desertic D. greeneana subsp. greeneana constitute a robustly supported clade (based on "locus 1") that may well have undergone initial divergence on GI in allopatry with the lineage represented by D. palmeri.
Reproductive isolation
Ecological differences between the completely interfertile and sympatric Deinandra greeneana subsp. greeneana and D. palmeri in such major attributes as seasonal activity, flowering time, and vegetative morphology have been implicated in reproductive isolation of the two taxa (Carlquist, 1965
). Climatic conditions at the low, south tip of GI and on the southern islets are extremely dry, with a recorded mean annual rainfall of 133 mm (range 14–693 mm), <10% of which occurs in summer months, when, in contrast with other vascular plants of the vicinity, D. greeneana subsp. greeneana is most actively flowering (Moran, 1996
). Deinandra palmeri flowers in spring months and is "...completely dormant, looking almost dead in summer" (Moran, 1996
, p. 87), but the overlap in flowering time with D. greeneana subsp. greeneana is sufficient to allow for at least occasional hybridization, with putative hybrids documented from both the main island (Moran, 1996
) and nearby Inner Islet (Rebman et al., 2002
).
Post-dispersal selection against hybrids at the juvenile stage, implicated in ecological isolation of various woody Compositae (e.g., Kyhos et al., 1981
), including Hawaiian Madiinae (Carr, 1995
), may be operating to limit gene flow between D. greeneana subsp. greeneana and D. palmeri. Natural hybridization between the two taxa is probably more extensive than is evident from mature plants; one of six seedlings growing in proximity to the two taxa that I removed from the field and grew under growth-chamber conditions was indistinguishable at maturity from an artificially synthesized F1 hybrid. Although the two taxa share the same habitat (notwithstanding possible micro-site differences), the leaves of each have highly contrasting morphological characteristics that may function similarly in imparting reflectivity and/or enhancing water retention in a common, desert-like, exposed setting, as suggested by Carlquist (1965)
. Leaves of D. palmeri have a dense covering of silky, white hairs that give the plants a shiny appearance, reminiscent of the highly reflective leaves of Hawaiian silverswords (Robichaux et al., 1990
; Melcher et al., 1994
); leaves of D. greeneana subsp. greeneana have sparse or no non-glandular hairs and are generally covered by a shiny exudate, produced from short-stipitate or sessile glands (Fig. 1). The possibility that the somewhat intermediate leaf characteristics of F1 hybrids are not optimal for the environmental conditions faced by the parental taxa warrants further study.
Evolutionary parallels with the Hawaiian silversword alliance
Carlquist's (1965)
suggestion that diversification of the shrubby tarweeds of the California Islands has paralleled, on a limited scale, adaptive radiation of the closely related Hawaiian silversword alliance (Argyroxiphium, Dubautia, and Wilkesia) is borne out by comparisons of evolutionary and biological attributes of the two groups. Based on diverse lines of data, the 31 woody or semi-woody species constituting the silversword alliance (Carr, 1985
, 1999
; Baldwin and Carr, 2005
) descended from herbaceous, western North American tarweed ancestors in the "Madia" lineage (Baldwin et al., 1991
; Baldwin, 1996
; Barrier et al., 1999
) and underwent rapid ecological diversification (Baldwin and Robichaux, 1995
; Baldwin and Sanderson, 1998
) in the absence of strong sterility barriers (Carr and Kyhos, 1986
), with limited natural hybridization documented in 41 different species combinations (Carr, 2003
). Most diversification in the silversword alliance evidently occurred on single islands or volcanoes, with dispersal among islands often not associated with diversification (Baldwin and Robichaux, 1995
; Baldwin, 1997
).
Based on the rDNA trees (Fig. 2), shrubby tarweeds on GI and other shrubby taxa of Deinandra are nested within a well-supported grade of annual, mainland species and represent at least one example of woodiness derived from an ancestrally herbaceous condition, as resolved for the silversword alliance and as commonly associated with plant evolution on oceanic islands (Carlquist, 1974
, 1995
; Baldwin et al., 1998
; Givnish, 1998
; Panero et al., 1999
). Lack of sterility barriers associated with diversification and occurrence of natural hybridization apparently limited by ecological differences between parental taxa are other features that the shrubby tarweeds of GI share with the silversword alliance (Carr and Kyhos, 1986
; Friar et al., 2006
) and various other examples of plant insular diversification (e.g., Ganders and Nagata, 1984
; Lowrey and Crawford, 1985
; Kim and Carr, 1990
; Mayer, 1991
; Smith et al., 1996
; Motley and Carr, 1998
), and with recently diverged woody plants in general (Grant, 1981
; Ellstrand et al., 1996
). In both the GI and Hawaiian-silversword-alliance lineages, at least partial interfertility between even the most morphologically and ecologically disparate species contrasts with intersterility between closely related, herbaceous species on the North American mainland (Clausen, 1951
; Kyhos et al., 1990
). As noted above, that contrast in patterns of interfertility may reflect relatively recent divergence of the insular species on Hawaii and GI compared to the mainland lineages and/or slower loss of interfertility between lineages of woody plants compared to annuals (Archibald et al., 2005
; Baldwin, 2006
).
Strong self-incompatibility (SI) in representatives of the three taxa of Deinandra from GI is unusual for plants of oceanic-island floras and parallels evidence for widespread SI in the closely related Hawaiian silversword alliance (Carr et al., 1986
). Carr et al. (1986)
noted that occurrence of sporophytic SI in the silversword alliance appeared to represent a major exception to Baker's Rule, which predicts that success of long-distance dispersal in plants will tend to be associated with capacity for self-fertilization (Baker, 1955
, 1967
). Carr et al.'s (1986)
hypothesis of ancestral SI in the silversword alliance was upheld by subsequent molecular phylogenetic results that place Hawaiian Madiinae among North American lineages of mostly self-incompatible taxa (Baldwin et al., 1991
; Baldwin, 1996
). Additional exceptions and possible exceptions to Baker's Rule have become evident from an increasing number of reports of SI or partial SI in oceanic-island floras worldwide (e.g., Anderson et al., 2001
; Nielsen et al., 2003
; Newstrom and Robertson, 2005
).
Mapping of evolutionary shifts between SI and SC on the rDNA trees for Deinandra indicated that SI was ancestral for all of the perennials, including the GI taxa (only the annuals D. arida and D. mohavensis are self-compatible [Tanowitz, 1982
; B. G. Baldwin, E. A. Friar, Rancho Santa Ana Botanic Garden, unpublished data]). Although frequency of SI has not been assessed for the flora of GI or the California Islands in general, selection for self-pollination in a recently introduced angiosperm of the Californian Channel Islands was found to be associated with establishment (Schueller, 2004
), in keeping with expectations of Baker's Rule and in insular settings that are not as remote as GI (see Fig. 1).
As noted by Carlquist (1965)
, the major disparity in diversity between the silversword alliance and the relatively depauperate assemblage of island taxa in Deinandra is consistent with greater opportunities for adaptive radiation in the Hawaiian archipelago than in the mostly smaller, less environmentally heterogeneous, and far less isolated California Islands. Differences in timing of diversification also may account for some of the difference in diversity between the silversword alliance and island deinandras. Although GI may be older than Kaua‘i (the oldest high island of the Hawaiian chain), diversification of Deinandra on GI evidently began later than diversification of the silversword alliance (
5.2 ± 0.8 Ma; Baldwin and Sanderson, 1998
). The estimated diversification rate of GI Deinandra (0.21 species per million years, for "locus 1") is approximately one-third to one-half the rate of diversification of the Hawaiian silversword alliance (0.56 ± 0.17 species per million years; Baldwin and Sanderson, 1998
), based on basally calibrated, ultrametric ITS trees of each group.
Other examples of diversification in the California Island flora?
The question remains as to whether or not radiation of Deinandra on GI was an exceptional event in the history of Deinandra or the California Island flora in general. The hypothesis that other island endemics in Deinandra represent products of a wider insular radiation including the GI lineage cannot be rejected based on SH-test results for the rDNA data. At a finer evolutionary scale, Carlquist (1965
, 1974
) noted that D. clementina is somewhat morphologically distinct on the different Channel Islands where it occurs and may be undergoing incipient, allopatric divergence; more rapidly evolving molecular regions must be used to assess that hypothesis.
Thorne's (1969)
suggestions of other angiosperm lineages showing possible evidence of evolution in the California Islands include shrubby mallows in Malva L. (previously in Lavatera L.), which also appear on the basis of molecular phylogenetic data to be potential examples of diversification across the California Islands (Ray, 1995
, 1998
), pending results from a more extensive sample of taxa from that highly diverse, widespread genus. Davis' (1997)
suggestion that Malacothrix DC. (Compositae) may be an example of adaptive radiation in the Channel Islands is a subject of current investigation (J. Lee [Korea Research Institute of Bioscience and Biotechnology], B. Baldwin [U. C. Berkeley], and W. S. Davis [Univ. of Louisville], unpublished data). Even Lyonothamnus, a relict genus of Rosaceae in the Channel Islands with a rich mainland fossil record, has been reinterpreted recently, with treatment of the two insular taxa as distinct from all fossil representatives (Erwin and Schorn, 2000
); minimal divergence of the Channel Island endemics in Lyonothamnus based on preliminary molecular data (Bushakra et al., 1999
) conceivably reflects recent diversification in an insular context, as suggested earlier by Thorne (1969)
.
Additional evolutionary studies, including population-level analyses (e.g., Helenurm and Hall, 2005
), offer excellent prospects for refining understanding of origins, relationships, and genetic diversity of the California Islands flora. Such studies are urgent for conserving remaining biodiversity and evolutionary processes in the wake of extreme decimation of plant and animal life by feral mammals, especially on GI (Moran, 1996
; de la Luz et al., 2003
), where recent efforts at goat removal (see Keitt et al., 2005
; Junak et al., 2005
) raise prospects for restoration of a "North Pacific Galápagos" (Niiler, 2000
).
Conclusions
Parallel attributes and histories of the GI deinandras and the Hawaiian silversword alliance may indicate that members of the tarweed–silversword lineage share characteristics that facilitate adaptive radiation under insular conditions. Evolutionary "flexibility" in key traits, such as bill morphology in the highly diverse Hawaiian honeycreepers (Lovette et al., 2002
), has been invoked to explain the capacity for adaptive radiation of some island lineages, as has the capability for external bird dispersal (epizoochory) in plants (Price and Wagner, 2004
), such as tarweeds and silverswords (Carlquist, 1974
). Major habitat shifts in the silversword alliance have been associated in part with repeated changes in leaf and wood anatomy with demonstrated or expected effects on capacity to resist drought stress (see Robichaux et al., 1990
; Baldwin and Robichaux, 1995
; Carlquist, 2003a
, b
) and involve characters (e.g., extracellular, mucilaginous polysaccharide) shared with California tarweeds (Carlquist, 1959
, 2003a
; Morse, 1990
), including Deinandra. Detailed, comparative studies of the GI deinandras and Hawaiian silversword alliance offer exciting prospects for achieving refined understanding of the evolutionary significance of traits associated with ecological diversification in these fascinating island plants.
APPENDIX
Collections of Deinandra Greene and outgroup (Holocarpha Greene) taxa examined in this study, with voucher and GenBank sequence accession numbers. All collections are from California, USA unless otherwise noted. Abbreviations: BGB, B. G. Baldwin; Co., County; E., east(ern); N., north(ern); S., south(ern); W., west(ern).
Taxon—Collection locality, Collector, collection number, (Herbarium); GenBank accession numbers for ITS, ETS.
Deinandra arida (D. D. Keck) B. G. Baldwin—Kern Co., Red Rock Canyon, State Highway 14 at S. Abbot Road junction, N35°21.9' W117°58.8', BGB 731 (JEPS); EF059611, EF059548; Tarweed Creek, N35°21.5' W117°58.6', BGB 732 (DAV, JEPS); EF059610, EF059547. D. bacigalupii B. G. Baldwin—Alameda Co., Livermore Valley (type locality), N37°44' W121°44', BGB 1053 (JEPS); EF059701, EF059601. D. clementina (Brandegee) B. G. Baldwin—Ventura Co., San Nicolas Island (N.E. of airfield, Beach Road grade), N33°15' W119°27', S. Junak s.n. (JEPS); EF059614–EF059623, EF059551; Middle Anacapa Island, N34°00–01' W119°23–24', S. Junak s.n. (JEPS); EF059624, EF059552. D. conjugens (D. D. Keck) B. G. Baldwin—San Diego Co., N. of Poggi Canyon (Chula Vista Community Hospital vicinity), N32°37' W117°01', F. T. Sproul s.n. (JEPS); EF059606, EF059543; S. of Otay Valley (Palm Avenue, E. of Interstate 805), N32°34.9' W117°01.4', E. T. Bauder & D. Truesdale s.n. (JEPS, SD); EF059607, EF059544. D. corymbosa (DC.) B. G. Baldwin—Contra Costa Co., Richmond Field Station (University of California), N37°55' W122°20', BGB 1211 (JEPS); EF059690, EF059590; San Luis Obispo Co., 4.2 km S. of San Carpoforo Creek (along State Highway 1), N35°44.0' W121°18.8', BGB 798 (JEPS); EF059691, EF059591. D. fasciculata (DC.) Greene—Orange Co., University of California, Irvine (Ecological Preserve), N33°38.5' W117°50', BGB s.n. (JEPS); EF059605, EF059542; San Luis Obispo Co., N.W. edge of Laguna Lake village (along Los Osos Valley Road), N35°16' W120°42', BGB 533 (DAV, JEPS); EF059604, EF059541. D. floribunda (A. Gray) Davidson & Moxley—San Diego Co., Bankhead Springs, N32°39' W116°15', F. T. Sproul s.n. (DAV 110998), EF059608, EF059545; S. of Bankhead Springs, N32°37' W116°15', J. P. Rebman 6031 (SD, UC, UCR); EF059609, EF059546. D. frutescens (A. Gray) B. G. Baldwin—Mexico, Baja California, Guadalupe Island, N. end (W. of stone hut ruins), N29°10', W118°18', BGB 688 (DAV); EF059660–EF059667, EF059576–EF059582. D. greeneana (Rose) B. G. Baldwin subsp. greeneana—Mexico, Baja California, Guadalupe Island, S. end (0.4 km N. of village), N28°53' W118°17', BGB 690 (DAV); EF059644–EF059650, EF059555–EF059563. D. greeneana (Rose) B. G. Baldwin subsp. peninsularis (Moran) B. G. Baldwin—Mexico, Baja California, Punta Banda (N.W. of La Bufadora), N31°44' W116°43', J. P. Rebman 6037 (SD); EF059668–EF059683, EF059583. D. halliana (D. D. Keck) B. G. Baldwin—Fresno Co., Cantua Creek drainage, N36°24–25' W120°33–34', BGB 796 (JEPS); EF059697, EF059597; San Luis Obispo Co., Temblor Range (0.8 km below Pinole Spring), N35°30' W120°04–05', S. J. Bainbridge s.n. (JEPS); EF059696, EF059596. D. increscens (H. M. Hall ex D. D. Keck) B. G. Baldwin subsp. increscens—San Luis Obispo Co., 3.2 km N. of Arroyo de la Cruz (along State Highway 1), N35°44.0' W121°18.8', BGB 1231 (JEPS); EF059688, EF059588; W. of Chorro Reservoir (serpentine outcrop), N35°20.3' W120°41.3', Wetherwax & Painter SLO-287 (JEPS, SBBG); EF059689, EF059589. D. kelloggii (Greene) Greene—Contra Costa Co., Oakley, N38°00' W121°43', S. J. Bainbridge s.n. (JEPS); EF059692, EF059592; Riverside Co., San Timoteo Canyon Road (2.9 km E. of Redlands Boulevard), N33°58.9' W117°7.7', BGB 1331 (JEPS); EF059693, EF059593. D. lobbii (Greene) Greene—Monterey Co., S. of Lockwood (Jolon Road, 0.8 km N.W. of US 101), N35°52.2' W120°50.5', BGB 718 (DAV, JEPS); EF059700, EF059600. D. martirensis (D. D. Keck) B. G. Baldwin—Mexico, Baja California, Sierra de San Pedro Mártir, Valladares, N30°51.5' W115°42', BGB 771 (UC); EF059643, EF059554. D. minthornii (Jeps.) B. G. Baldwin—Los Angeles Co., Santa Monica Mountains (between Old Topanga Canyon Road and Stunt Road), N34°06' W118°38–39', M. S. Witter 86–64, 86–87 (DAV); EF059612, EF059549; E. of Santa Susana Pass (Santa Puela Place, 0.2 km W. of Topanga Canyon Road), N34°16.7' W118°36.3', BGB 1333 (JEPS); EF059613, EF059550; Ventura Co., Santa Susana Mountains, Rocky Peak, N34°18' W118°38', D. Koutnik s.n. (DAV); in cytogenetic analyses only. D. mohavensis (D. D. Keck) B. G. Baldwin—Kern Co., Sierra Nevada (E. slope), Short Canyon, N35°42.6' W117°55.1', BGB 1052 (JEPS); EF059685, EF059585; Riverside Co., San Jacinto Mountains (N. slope), Azalea Creek, N33°52' W116°48', A. C. Sanders 15774 (RSA, SD, UCR); EF059684, EF059584. D. pallida (D. D. Keck) B. G. Baldwin—Kern Co., W. San Joaquin Valley (State Highway 33, 0.5 km S. of S. junction with Lost Hills Road), N35°25.8' W119°41.0', BGB 696 (DAV, JEPS); EF059694, EF059594; foothills of Tehachapi Range, (State Highway 223, 6.4 km S.W. of State Highway 58 junction), N35°13.4' W118°42.5', BGB 1277 (JEPS); EF059695, EF059595. D. palmeri (Rose) B. G. Baldwin—Mexico, Baja California, Guadalupe Island, S. end (0.4 km N. of village), N28°53' W118°17', BGB 690 (DAV); EF059651–EF059659, EF059564–EF059575. D. paniculata (A. Gray) Davidson & Moxley—Orange Co., San Juan Creek Road (3.2 km E. of Interstate 5), N33°31' W117°38', BGB s.n. (JEPS); EF059686, EF059586; San Diego Co., San Onofre Creek (Basilone Road, <1.6 km E. of State Highway 1), N33°23.2' W117°34.2', E. T. Bauder & D. Truesdale s.n. (JEPS, SD); EF059687, EF059587. D. pentactis (D. D. Keck) B. G. Baldwin—San Luis Obispo Co., Shell Creek Road at State Highway 58 junction, N35°27.5' W120°20.0', BGB 538 (DAV); EF059698, EF059598; Creston (Little Farm Road, 0.5 km S. of State Highway 41 junction), N35°31.9' W120°30.7', BGB 1145 (JEPS); EF059699, EF059599. D. streetsii (A. Gray) B. G. Baldwin—Mexico, Baja California, West San Benito Island (just S. of lighthouse), N28°18' W115°35', S. Junak 5332 (SBBG); EF059625–EF059642, EF059553. Holocarpha heermannii (Greene) D. D. Keck—Tulare Co., N.E. of Lemoncove (State Highway 198, 1 km E. of State Highway 216 junction), N36°24.1' W119°00.8', BGB 686 (DAV); EF059603, EF059540. H. virgata (A. Gray) D. D. Keck—Lake Co., Lower Lake (State Highway 29, 0.4 km W. of State Highway 53 junction), N38°54.6' W122°37.0', BGB 1354 (JEPS); EF059602, EF059539.
FOOTNOTES
1 This work was supported in part by NSF (DEB-9458237), the Lawrence R. Heckard Endowment Fund (Jepson Herbarium), and Roderic B. Park and other Friends of the Jepson Herbarium. The author thanks S. J. Bainbridge, E. T. Bauder, R. M. Beauchamp, S. Junak, D. L. Koutnik, S. N. Martens, E. L. Painter, J. P. Rebman, A. C. Sanders, F. T. Sproul, D. Truesdale, S. G. Weller, M. Wetherwax, and M. S. Witter for help with collecting, K. Klitz for preparing Fig. 1, D. W. Kyhos for cytogenetic advice and assistance, R. L. Moe for technical help, J. L. Strother and two anonymous reviewers for helpful comments on an earlier version of the manuscript, and B. L. Wessa for extensive laboratory assistance. Special thanks to R. Moran, M. Stinson, and the crew of the Pacific Queen for making the author's visit to Guadalupe Island possible and to R. Moran for sharing his unsurpassed botanical knowledge of Guadalupe Island. 
2 Author for correspondence (bbaldwin{at}berkeley.edu
)
LITERATURE CITED
Anderson G. J. Bernardello G. Stuessy T. F. Crawford D. J.. 2001. Breeding system and pollination of selected plants endemic to the Juan Fernandez Islands. American Journal of Botany 88: 220-233.
Archibald J. K. Mort M. E. Crawford D. J. Kelly J. K.. 2005. Life history affects the evolution of reproductive isolation among species of Coreopsis (Asteraceae). Evolution 59: 2362-2369.[Web of Science][Medline]
Axelrod D. I.. 1958. Evolution of the Madro-Tertiary geoflora. Botanical Review 24: 433-509.
Axelrod D. I.. 1967. Geologic history of the Californian insular flora. In R. N. Philbrick [ed.] Proceedings of the symposium on the biology of the California Islands, 1965, Santa Barbara, California 267-315 Santa Barbara Botanic Garden, Santa Barbara, California, USA.
Axelrod D. I.. 1992. Miocene floristic change at 15 Ma, Nevada to Washington, U.S.A. Palaeobotanist 41: 234-239.
Baker H. G.. 1955. Self-compatibility and establishment after "long-distance" dispersal. Evolution 9: 347-349.[CrossRef][Web of Science]
Baker H. G.. 1967. Support for Baker's Law—as a rule. Evolution 21: 853-856.[CrossRef][Web of Science]
Baldwin B. G.. 1996. Phylogenetics of the California tarweeds and the Hawaiian silversword alliance (Madiinae; Heliantheae sensu lato). In D. J. N. Hind and H. Beentje [eds.] Compositae: systematics. Proceedings of the International Compositae Conference Kew, UK,1994, vol. 1, 377-391 Royal Botanic Gardens, Kew, UK.
Baldwin B. G.. 1997. Adaptive radiation of the Hawaiian silversword alliance: congruence and conflict of phylogenetic evidence from molecular and non-molecular investigations. In T. J. Givnish and K. J. Sytsma [eds.] Molecular evolution and adaptive radiation 103-128 Cambridge University Press, Cambridge, UK.
Baldwin B. G.. 2003a. Characteristics and diversity of Madiinae. In S. Carlquist, B. G. Baldwin, and G. D. Carr [eds.] Tarweeds & silverswords: evolution of the Madiinae (Asteraceae) 17-52 Missouri Botanical Garden Press, St. Louis, Missouri, USA.
Baldwin B. G.. 2003b. A phylogenetic perspective on the origin and evolution of Madiinae. In S. Carlquist, B. G. Baldwin, and G. D. Carr [eds.] Tarweeds & silverswords: evolution of the Madiinae (Asteraceae) 193-228 Missouri Botanical Garden Press, St. Louis, Missouri, USA.
Baldwin B. G.. 2005. Origin of the serpentine-endemic herb Layia discoidea from the widespread L. glandulosa (Compositae). Evolution 59: 2473-2479.[Web of Science][Medline]
Baldwin B. G.. 2006. Contrasting patterns and processes of evolutionary change in the tarweed–silversword lineage: revisiting Clausen, Keck, and Hiesey's findings. Annals of the Missouri Botanical Garden 93: 66-96.
Baldwin B. G. In press A new combination and new chromosome counts in the tarweed tribe (Compositae–Madiinae). Madroño..
Baldwin B. G. Carr G. D.. 2005. Dubautia kalalauensis, a new species of the Hawaiian silversword alliance (Compositae, Madiinae) from northwestern Kaua'i, USA. Novon 15: 259-263.
Baldwin B. G. Crawford D. J. Francisco-Ortega J. Kim S.-C. Sang T. Stuessy T. F.. 1998. Molecular phylogenetic insights on the origin and evolution of oceanic island plants. In P. S. Soltis, D. E. Soltis, and J. J. Doyle [eds.] Molecular systematics of plants II: DNA sequencing 410-441 Kluwer, Boston, Massachusetts, USA.
Baldwin B. G. Kyhos D. W. Dvorak J. Carr G. D.. 1991. Chloroplast DNA evidence for a North American origin of the Hawaiian silversword alliance (Asteraceae: Madiinae). Proceedings of the National Academy of Sciences, USA 88: 1840-1843.
Baldwin B. G. Markos S.. 1998. Phylogenetic utility of the external transcribed spacer (ETS) of 18S-26S rDNA: congruence of ETS and ITS trees of Calycadenia (Compositae). Molecular Phylogenetics and Evolution 10: 449-463.[CrossRef][Web of Science][Medline]
Baldwin B. G. Robichaux R. H.. 1995. Historical biogeography and ecology of the Hawaiian silversword alliance (Asteraceae): new molecular phylogenetic perspectives. In W. L. Wagner and V. A. Funk [eds.] Hawaiian biogeography: evolution on a hot spot archipelago 259-287 Smithsonian Institution Press, Washington, D.C., USA.
Baldwin B. G. Sanderson M. J.. 1998. Age and rate of diversification of the Hawaiian silversword alliance (Compositae). Proceedings of the National Academy of Sciences, USA 95: 9402-9406.
Baldwin B. G. Wessa B. L.. 2000. Origin and relationships of the tarweed–silversword lineage (Compositae–Madiinae). American Journal of Botany 87: 1890-1908.
Baldwin B. G. Wessa B. L. Panero J. L.. 2002. Nuclear rDNA evidence for major lineages of helenioid Heliantheae (Compositae). Systematic Botany 27: 161-198.
Barrier M. Baldwin B. G. Robichaux R. H. Purugganan M. D.. 1999. Interspecific hybrid ancestry of a plant adaptive radiation: allopolyploidy of the Hawaiian silversword alliance (Asteraceae) inferred from floral homeotic gene duplications. Molecular Biology and Evolution 16: 1105-1113.[Abstract]
Beeks R. M.. 1955. Improvements in the squash technique for plant chromosomes. Aliso 3: 131-134.[Medline]
Bushakra J. M. Hodges S. A. Cooper J. B. Kaska D. D.. 1999. The extent of clonality and genetic diversity in the Santa Cruz Island ironwood, Lyonothamnus floribundus. Molecular Ecology 8: 471-475.[CrossRef][Web of Science]
Carlquist S.. 1959. Studies on Madinae: anatomy, cytology, and evolutionary relationships. Aliso 4: 171-236.[Medline]
Carlquist S.. 1965. Island life, a natural history of the islands of the world Natural History Press, Garden City, New York, USA.
Carlquist S.. 1974. Island biology Columbia University Press, New York, New York, USA.
Carlquist S.. 1995. Introduction. In W. L. Wagner and V. A. Funk [eds.] Hawaiian biogeography: evolution on a hot spot archipelago 1-13 Smithsonian Institution Press, Washington, D.C., USA.
Carlquist S.. 2003a. Leaf anatomy of Madiinae and adaptive radiation. In S. Carlquist, B. G. Baldwin, and G. D. Carr [eds.] Tarweeds & silverswords: evolution of the Madiinae (Asteraceae) 115-127 Missouri Botanical Garden Press, St. Louis, Missouri, USA.
Carlquist S.. 2003b. Wood anatomy of Madiinae in relation to ecological diversification. In S. Carlquist, B. G. Baldwin, and G. D. Carr [eds.] Tarweeds & silverswords: evolution of the Madiinae (Asteraceae) 129-144 Missouri Botanical Garden Press, St. Louis, Missouri, USA.
Carlquist S. Baldwin B. G. Carr G. D.. 2003. Tarweeds & silverswords: evolution of the Madiinae (Asteraceae) Missouri Botanical Garden Press, St. Louis, Missouri, USA.
Carr G. D.. 1985. Monograph of the Hawaiian Madiinae (Asteraceae): Argyroxiphium, Dubautia, and Wilkesia. Allertonia 4: 1-123.
Carr G. D.. 1995. A fully fertile intergeneric hybrid derivative from Argyroxiphium sandwicense ssp. macrocephalum x Dubautia menziesii (Asteraceae) and its relevance to plant evolution in the Hawaiian Islands. American Journal of Botany 82: 1574-1581.[CrossRef][Web of Science]
Carr G. D.. 1999. A reassessment of Dubautia (Asteraceae: Heliantheae–Madiinae) on Kaua‘i. Pacific Science 53: 144-158.
Carr G. D.. 2003. Hybridization in Madiinae. In S. Carlquist, B. G. Baldwin, and G. D. Carr [eds.] Tarweeds & silverswords: evolution of the Madiinae (Asteraceae) Missouri Botanical Garden Press, St. Louis, Missouri, USA.
Carr G. D. Kyhos D. W.. 1986. Adaptive radiation in the Hawaiian silversword alliance (Compositae–Madiinae). II. Cytogenetics of artificial and natural hybrids. Evolution 40: 959-976.[CrossRef][Web of Science]
Carr G. D. Powell E. A. Kyhos D. W.. 1986. Self-incompatibility in the Hawaiian Madiinae (Compositae): an exception to Baker's Rule. Evolution 40: 430-434.[CrossRef][Web of Science]
Clausen J.. 1951. Stages in the evolution of plant species Cornell University Press, Ithaca, New York, USA.
Davis W. S.. 1997. The systematics of annual species of Malacothrix (Asteraceae: Lactuceae) endemic to the California Islands. Madroño 44: 223-244.
de la Luz J. L. L. Rebman J. P. Oberbauer T.. 2003. On the urgency of conservation on Guadalupe Island, Mexico: is it a lost paradise?. Biodiversity and Conservation 12: 1073-1082.[CrossRef][Web of Science]
Doyle J. J. Doyle J. L.. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15.
Ellstrand N. C. Whitkus R. Riesberg L. H.. 1996. Distribution of spontaneous plant hybrids. Proceedings of the National Academy of Sciences, USA 93: 5090-5093.
Engel A. E. J. Eengel C. G.. 1970. Mafic and ultramafic rocks. In A. E. Maxwell [ed.] The sea, ideas and observations on progress in the study of the seas, vol. 4, part 1, New concepts of sea floor evolution 465-519 Wiley Interscience, New York, New York, USA.
Erwin D. M. Schorn H. E.. 2000. Revision of Lyonothamnus A. Gray (Rosaceae) from the Neogene of western North America. International Journal of Plant Sciences 161: 179-193.[CrossRef][Web of Science][Medline]
Felsenstein J.. 1988. Phylogenies from molecular sequences: inference and reliability. Annual Review of Genetics 22: 521-566.[CrossRef][Web of Science][Medline]
Felsenstein J.. 2004. PHYLIP: phylogeny inference package University of Washington, Seattle, Washington, USA.
Fitch W. M. Margoliash E.. 1967. Construction of phylogenetic trees. Science 155: 279-284.
Flower B. P. Kennett J. P.. 1994. The middle Miocene climatic transition: east Antarctic ice sheet development, deep ocean circulation and global carbon cycling. Palaeogeography, Palaeoclimatology, and Palaeoecology 108: 537-555.[CrossRef]
Friar E. A. Prince L. M. Roalson E. H. McGlaughlin M. E. Cruse-Sanders J. M. de Groot S. J. Porter J. M.. 2006. Ecological speciation in the East Maui–endemic Dubautia (Asteraceae) species. Evolution 60: 1777-1792.[Web of Science][Medline]
Ganders F. R. Nagata K. M.. 1984. The role of hybridization in the evolution of Bidens in the Hawaiian Islands. In W. F. Grant [ed.] Plant biosystematics 179-194 Academic Press, Toronto, Canada.
Givnish T. J.. 1998. Adaptive plant evolution on islands: classical patterns, molecular data, new insights. In P. R. Grant [ed.] Evolution on islands 281-304 Oxford University Press, Oxford, UK.
Goldman N. Anderson J. P. Rodrigo A. G.. 2000. Likelihood-based tests of topologies in phylogenetics. Systematic Biology 49: 652-670.
Grant V.. 1981. Plant speciation, 2nd ed Columbia University Press, New York, New York, USA.
Helenurm K. Hall S. S.. 2005. Dissimilar patterns of genetic variation in two insular endemic plants sharing species characteristics, distribution, habitat, and ecological history. Conservation Genetics 6: 341-353.[CrossRef][Web of Science]
Hickman J. C.. 1993. The Jepson manual: higher plants of California University of California Press, Berkeley, California, USA.
Junak S. Keitt B. Tershy B. Croll D. Luna-Mendoza L. M. Aguirre-Munoz A.. 2005. Esfuezos recientes de conservación y apuntes sobre el estado actual de la flora de Isla Guadalupe. In Santos del Prado G. K. Peters E. Isla Guadalupe: restauración y conservación Instituto Nacional de Ecología, Mexico City, D. F., Mexico.
Kay K. M. Whittall J. B. Hodges S. A.. 2006. A survey of nuclear ribosomal internal transcribed spacer substitution rates across angiosperms: an approximate molecular clock with life history effects. BMC Evolutionary Biology 6: article no. 36.
Keck D. D.. 1959. Hemizonia. In P. A. Munz [ed.] A California flora 1117-1124 University of California Press, Berkeley, California, USA.
Keitt B. Junak S. Luna-Endoza L. Aguirre A.. 2005. The restoration of Guadalupe Island. Fremontia 33: 20-25.
Kim I. Carr G. D.. 1990. Cytogenetics and hybridization of Portulaca in Hawaii. Systematic Botany 15: 370-377.[CrossRef][Web of Science]
Kyhos D. W. Carr G. D. Baldwin B. G.. 1990. Biodiversity and cytogenetics of the tarweeds (Asteraceae: Heliantheae–Madiinae). Annals of the Missouri Botanical Garden 77: 84-95.[CrossRef][Web of Science]
Kyhos D. W. Clark C. Thompson W. C.. 1981. The hybrid nature of Encelia laciniata (Compositae: Heliantheae) and control of population composition by post-dispersal selection. Systematic Botany 6: 399-411.[CrossRef][Web of Science]
Lovette I. J. Bermingham E. Ricklefs R. E.. 2002. Clade-specific morphological diversification and adaptive radiation in Hawaiian songbirds. Proceedings of the Royal Society of London, B, Biological Sciences 269: 37-42.
Lowrey T. K. Crawford D. J.. 1985. Allozyme divergence and evolution in Tetramolopium (Compositae: Astereae) on the Hawaiian Islands. Systematic Botany 10: 64-72.[CrossRef][Web of Science]
Maddison D. R. Maddison W. P.. 2000. MacClade 4: analysis of phylogeny and character evolution. Version 4.0 Sinauer, Sunderland, Massachusetts, USA.
Maneval W. E.. 1936. Lacto-phenol preparations. Stain Technology 11: 9-11.
Mayer S. S.. 1991. Artificial hybridization in Hawaiian Wikstroemia (Thymelaeaceae). American Journal of Botany 78: 122-130.[CrossRef][Web of Science]
Melcher P. J. Goldstein G. Meinzer F. C. Minyard B. Giambelluca T. W. Loope L. L.. 1994. Determinants of thermal balance in the Hawaiian giant rosette plant, Argyroxiphium sandwicense. Oecologia 98: 412-418.[CrossRef][Web of Science]
Moran R.. 1969. Twelve new dicots from Baja California, Mexico. Transactions of the San Diego Society of Natural History 15: 265-295.
Moran R.. 1996. The flora of Guadalupe Island, Mexico. Memoirs of the California Academy of Sciences 19: 1-190.[Medline]
Moran R. Carlquist S. Howell T. R.. 1967. Discussion of the flora of Guadalupe Island. In R. N. Philbrick [ed.] Proceedings of the symposium on the biology of the California Islands 1965,Santa Barbara, California,69-71 Santa Barbara Botanic Garden, Santa Barbara, California, USA.
Morse S.. 1990. Water balance in Hemizonia luzulifolia: the role of extracellular polysaccharide. Plant, Cell and Environment 13: 39-48.[Medline]
Motley T. J. Carr G. D.. 1998. Artificial hybridization in the Hawaiian endemic genus Labordia (Loganiaceae). American Journal of Botany 85: 654-660.[Abstract]
Newstrom L. Robertson A.. 2005. Progress in understanding pollination systems in New Zealand. New Zealand Journal of Botany 43: 1-59.
Nielsen L. R. Siegismund H. R. Phillipp M.. 2003. Partial self-incompatibility in the polyploid endemic species Scalesia affinis (Asteraceae) from the Galápagos: remnants of a self-incompatibility system?. Botanical Journal of the Linnean Society 142: 93-101.[CrossRef]
Niiler E.. 2000. Island survivors: on what was once a North American Galápagos, researchers try to save devastated wildlife. Scientific American 283: 18-20.[Web of Science][Medline]
Panero J. L. Francisco-Ortega J. Jansen R. K. Santos-Guerra A.. 1998. Molecular evidence for multiple origins of woodiness and a New World biogeographic connection of the Macaronesian Island endemic Pericallis (Asteraceae: Senecioneae). Proceedings of the National Academy of Sciences, USA 96: 13886-13891.[CrossRef][Web of Science]
Posada D. Crandall K. A.. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817-818.
Price J. Wagner W. L.. 2004. Speciation in Hawaiian angiosperm lineages: cause, consequence, and mode. Evolution 58: 2185-2200.[Web of Science][Medline]
Raven P. H.. 1963. A flora of San Clemente Island, California. Aliso 5: 289-347.
Raven P. H.. 1967. The floristics of the California Islands. In R. N. Philbrick [ed.] Proceedings of the symposium on the biology of the California Islands 1965,Santa Barbara, California,57-67 Santa Barbara Botanic Garden, Santa Barbara, California, USA.
Raven P. H. Axelrod D. I.. 1978. Origin and relationships of the California flora. University of California Publications in Botany 72: 1-134.
Raven P. H. Evert R. F. Eichhorn S. E.. 2005. Biology of plants, 7th ed W. H. Freeman, New York, New York, USA.
Ray M. F.. 1995. Systematics of Lavatera and Malva (Malvaceae, Malveae)—a new perspective. Plant Systematics and Evolution 198: 29-53.[CrossRef][Web of Science]
Ray M. F.. 1998. New combinations in Malva (Malvaceae: Malveae). Novon 8: 288-295.[CrossRef][Web of Science]
Rebman J. P. Oberbauer T. A. de la Luz J. L. L.. 2002. The flora of Toro Islet and notes on Guadalupe Island, Baja California, Mexico. Madroño 49: 145-149.
Robichaux R. H. Carr G. D. Liebman M. Pearcy R. W.. 1990. Adaptive radiation of the Hawaiian silversword alliance (Compositae–Madiinae): ecological, morphological, and physiological diversity. Annals of the Missouri Botanical Garden 77: 64-72.[CrossRef][Web of Science]
Rosenzweig M. L. Vetault S.. 1992. Calculating speciation and extinction rates in fossil clades. Evolutionary Ecology 6: 90-93.[CrossRef][Web of Science]
Saitou N. Nei M.. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4: 406-425.[Abstract]
Sanderson M. J.. 2003. r8s: inferring absolute rates of molecular evolution and divergence times in the absence of a molecular clock. Bioinformatics 19: 301-302.
Schluter D.. 2000. The ecology of adaptive radiation Oxford University Press, New York, New York, USA.
Schueller S. K.. 2004. Self-pollination in island and mainland populations of the introduced hummingbird-pollinated plant, Nicotiana glauca (Solanaceae). American Journal of Botany 91: 672-681.
Shimodaira H. Hasegawa M.. 1999. Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Molecular Biology and Evolution 16: 1114-1116.[Web of Science]
Smith J. F. Burke C. C. Wagner W. L.. 1996. Interspecific hybridization in natural populations of Cyrtandra (Gesneriaceae) on the Hawaiian Islands: evidence from RAPD markers. Plant Systematics and Evolution 200: 61-77.[CrossRef][Web of Science]
Stanley S. M.. 1979. Macroevolution—pattern and process W. H. Freeman, San Francisco, California, USA.
Stuessy T. F. Ono M.. 1998. Evolution and speciation of island plants Cambridge University Press, Cambridge, UK.
Swofford D. L.. 2002. PAUP*: phylogenetic analysis using parsimony (*and other methods), version 4.0 beta Sinauer, Sunderland, Massachusetts, USA.
Tanowitz B. D.. 1982. Taxonomy of Hemizonia sect. Madiomeris (Asteraceae: Madiinae). Systematic Botany 7: 314-339.[CrossRef][Web of Science]
Tanowitz B. D.. 1985. Systematic studies in Hemizonia (Asteraceae: Madiinae): hybridization of H. fasciculata with H. clementina and H. minthornii. Systematic Botany 10: 110-118.[CrossRef][Web of Science]
Thorne R. F.. 1969. The California Islands. Annals of the Missouri Botanical Garden 56: 391-408.[CrossRef][Web of Science]
Wallace G. D.. 1985. Vascular plants of the Channel Islands of southern California and Guadalupe Island, Baja California, Mexico. Contributions in Science (Los Angeles County Museum) 365: 1-136.
Weller S. G. Sakai A. K. Wagner W. L.. 2001. Artificial and natural hybridization in Schiedea and Alsinodendron (Caryophyllaceae: Alsinoideae): the importance of phylogeny, genetic divergence, breeding system, and population size. Systematic Botany 26: 571-584.[Web of Science]
White T. J. Bruns T. Lee S. Taylor J.. 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.
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
