HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

What's this?

This Article Abstract Full Text (PDF) Supplementary Data Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Tate, J. A. Articles by Simpson, B. B. Search for Related Content PubMed Articles by Tate, J. A. Articles by Simpson, B. B. Agricola Articles by Tate, J. A. Articles by Simpson, B. B. Social Bookmarking                            What's this?

Phylogenetic relationships within the tribe Malveae (Malvaceae, subfamily Malvoideae) as inferred from ITS sequence data¹

²Section of Integrative Biology and Plant Resources Center, The University of Texas at Austin, Austin, Texas 78712 USA; ³Real Jardín Botánico, CSIC, Plaza de Murillo, 2. 28014, Madrid, Spain; ⁴Allan Herbarium, Landcare Research, P.O. Box 69, Lincoln 8152, New Zealand; ⁵Department of Biology, The University of North Dakota, Grand Forks, North Dakota 58202 USA; ⁶Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061 USA

Received for publication June 13, 2004. Accepted for publication December 21, 2004.

ABSTRACT

Phylogenetic relationships among genera of tribe Malveae (Malvaceae, subfamily Malvoideae) were reconstructed using sequences of the internal transcribed spacer (ITS) region of the 18S–26S nuclear ribosomal repeat. Newly generated sequences were combined with those available from previous generic level studies to assess the current circumscription of the tribe, monophyly of some of the larger genera, and character evolution within the tribe. The ITS data do not support monophyly of most generic alliances as presently defined, nor do the data support monophyly of several Malveae genera. Two main well-supported clades were recovered, which correspond primarily to taxa that either possess or lack involucral bracts, respectively. Chromosomal evolution has been dynamic in the tribe with haploid numbers varying from n = 5 to 36. Aneuploid reduction, hybridization, and/or polyploidization have been important evolutionary processes in this group.

Key Words: Bayesian analysis • ITS • Malvaceae • Malveae • Malvoideae • molecular phylogeny • parsimony analysis

In recent years, morphological and molecular evidence have shown that many of the traditional families of the Malvales are not monophyletic (Judd and Manchester, 1997 ; Alverson et al., 1998 , 1999 ; Bayer et al., 1999 ). As a result, an expanded circumscription of the Malvaceae has been created, which is composed of nine subfamilies: Bombacoideae (formerly Bombacaceae, in part), Brownlowioideae, Byttnerioideae, Dombeyoideae, Grewioideae, Helicteroideae, Malvoideae (formerly Malvaceae), Sterculioideae (formerly Sterculiaceae, in part), and Tilioideae (formerly Tiliaceae, in part) (Bayer et al., 1999 ; Bayer and Kubitzki, 2003 ). Subfamily Malvoideae (Eumalvoideae of Baum et al., 2004 ) has consistently emerged as a monophyletic group on the basis of both morphological and molecular data (Judd and Manchester, 1997 ; Alverson et al., 1999 ; Bayer et al., 1999 ). In the most recent treatment of Malvoideae, Bayer and Kutbitzki (2003) divide the subfamily into four tribes: Gossypieae, Hibisceae, Kydieae, and Malveae.

As considered here, tribe Malveae includes approximately 70 genera (1000 species) that encompass the majority of the morphological and taxonomic diversity in the subfamily (Table 1) (Fryxell, 1997 ). Traditionally, members of the Malveae have been characterized by a combination of several morphological characters: schizocarpic fruits (sometimes a capsule), mericarps numbering 3 to over 20 and equal to the number of free styles, antheriferous apex of the staminal column, and the absence of lysigenous cavities ("gossypol glands") (Fryxell, 1988 ; Bayer and Kubitzki, 2003 ). The genera of Malveae exhibit a broad geographic distribution, with representatives in both tropic and temperate areas exploiting a variety of habitats. Around 15 of the 70 Malveae genera have mostly temperate distributions, while some of the largest genera in the tribe (Abutilon, Sida, Nototriche) have primarily tropical distributions (Table 1).

View this table: [in this window] [in a new window]   Table 1. Genera of tribe Malveae (Malvaceae), their alliance associations, reported chromosome numbers and geographic distributions (Fryxell, 1997 ; Bayer and Kubitzki, 2003 , with modification). Chromosome numbers in brackets are those reported in the literature, but are seemingly out of sync with numbers published for the rest of the genus

Most recently, Bayer and Kubitzki (2003) provided a comprehensive treatment for the tribe, as well as for the entire subfamily and family. Fourteen Malveae alliances were maintained, but their generic compositions were altered somewhat (Table 2). The genera previously segregated into the Herrisantia, Robinsonella, and Sida alliances by Fryxell (1988) were subsumed into the Abutilon alliance. Members of the Bakeridesia and Fryxellia alliances were included with the Batesimalva alliance. The Malacothamnus alliance was maintained in name as originally proposed by Bates and Blanchard (1970) , but its generic composition follows Fryxell (1988) . The Napaea alliance was included with the Sphaeralcea alliance. In the present study, we will follow the taxonomy of Bayer and Kubitzki (2003) with slight modification to reflect recent taxonomic changes: the addition of Navaea to the Malva alliance (Fuertes Aguilar et al., 2003 ), Tropidococcus to the Modiola alliance (Fernandez et al., 2003 ; Krapovickas, 2003 ) and Andeimalva to the Sphaeralcea alliance (Tate, 2003 ).

Recent molecular studies of the Malvales and the Malvoideae (as Malvaceae sensu stricto) have provided preliminary evidence for phylogenetic relationships within the subfamily as well as within the tribe Malveae. Tribe Gossypieae was sister to Malveae based on rbcL and atpB (Bayer et al., 1999 ) and ndhF (Alverson et al., 1999 ) sequence data. All five tribes were represented in a recent phylogenetic analysis of tribe Hibisceae, using chloroplastic ndhF and rpl16 intron sequences (Pfeil et al., 2002 ). Although only a few genera of Malveae were included, the resulting trees placed Malveae and Gossypieae at the base of an unresolved clade and sister to most of the Hibisceae. Another cpDNA based study, using restriction site data (La Duke and Doebley, 1995 ), sampled more extensively in the Malveae and placed the tribe in a clade that was sister to the remaining tribes of subfamily Malvoideae. Although La Duke and Doebley's study did not support monophyly of the Malveae alliances, it did identify two major clades: one composed of the Abutilon and Sida alliances and the other composed of the remaining alliances. A recent phylogenetic analysis based on sequence data from the internal transcribed spacer (ITS) regions of the 18–26S nuclear ribosomal repeat (Fuertes Aguilar et al., 2003 ) examined the phylogenetic relationships of the Abutilon and Sida alliances. Although their sampling was not exhaustive, neither alliance was supported as monophyletic (Fuertes Aguilar et al., 2003 ).

Previous studies have demonstrated that sufficient variation exists in ITS to resolve phylogenetic relationships within and between genera in the Malvoideae (Seelanan et al., 1997 ) and particularly in the Malveae (Ray, 1995 ; Whittall et al., 2000 ; Andreasen and Baldwin, 2001 ; Fuertes Aguilar et al., 2003 ; Tate and Simpson, 2003 ). Furthermore, several of these studies have also revealed that some genera are not monophyletic as currently circumscribed. Among these are Abutilon and Sida (Fuertes Aguilar et al., 2003 ), Malva and Lavatera (Ray, 1995 ), and Tarasa (Tate and Simpson, 2003 ). We extended these earlier studies with a broader sample representing most of the genera in tribe Malveae. The main objectives of this study were to reconstruct phylogenetic relationships in tribe Malveae, to assess the amount of congruence between the inferred relationships and the existing classification, to identify potential morphological synapomorphies that might support the reconstructed clades, and finally, to examine character evolution within the tribe.

MATERIALS AND METHODS

Taxon sampling
We sampled 68 genera (121 species) in our study representing all of the 14 alliances recognized by Bayer and Kubitzki (2003) . To assess monophyly of the genera, as well as intrageneric variation, two or more species from the same genus were included when possible. The outgroups included Gossypium, Kokia, Lebronnecia, and Thespesia (tribe Gossypieae), and Howittia (incertae sedis fide Bayer and Kubitzki, 2003 ). Members of the Gossypieae were included based on previous molecular phylogenies for subfamily Malvoideae, which indicated that tribe Gossypieae is sister to Malveae (Alverson et al., 1999 ; Bayer et al., 1999 ). Originally, Howittia was included in the Malveae by Bentham and Hooker (1862) , but later workers suggested that it should be placed in tribe Hibisceae (Edlin, 1935 ; Fryxell, 1968 ). Recent molecular analyses based on cpDNA sequence data found Hibisceae to be paraphyletic, with four Hibisceae genera (Camptostemon, Radyera, Howittia, and Lagunaria) placed sister to the remaining members of the Malvoideae (Pfeil et al., 2002 ). Tribes Malveae and Gossypieae (both of which were monophyletic) formed a clade sister to a clade containing tribes Decaschistieae, Malvavisceae, and the remaining Hibisceae genera (Pfeil et al., 2002 ). Although it is clear that Howittia does not belong in either tribe Hibisceae or Malveae, we include the genus here to represent a more distantly related lineage of Malvoideae.

The taxa sampled, voucher information, and GenBank accession numbers are available as a Data Supplement (Appendix 1) accompanying the online version of this article.

DNA extraction and ITS amplification
Total DNA was extracted from fresh material, herbarium specimens or silica—gel-dried material (Chase and Hills, 1991 ) by various modifications of the CTAB protocol (Doyle and Doyle, 1987 ). The internal transcribed spacer (ITS) region of the 18S–26S nuclear ribosomal repeat was amplified by the polymerase chain reaction (PCR) as previously described (Fuertes Aguilar et al., 2003 ; Tate and Simpson, 2003 ). Amplification products were separated on a 1% agarose gel, stained with ethidium bromide, and then visualized with UV on a transilluminator. PCR products were cleaned using QIAquick spin columns (Qiagen, Valencia, California, USA) following the manufacturer's instructions. Cycle sequencing was performed using Big Dye terminator chemistry (Applied Biosystems, Foster City, California, USA). Bidirectional automated sequencing using the forward and reverse amplification primers was conducted on an ABI 3700 or 377 at the DNA Analysis Laboratory at The University of Texas at Austin or an ABI 3100 at The University of North Dakota.

Sequence alignment and phylogenetic analysis
The boundaries of ITS were determined by comparison to a published Gossypium sequence in GenBank (U12719, http://www.ncbi.nlm.nih.gov/). Forward and reverse sequences were assembled into contigs and edited using Sequencher (Gene Codes Corporation, 1995). The sequences were aligned using Clustal X (Thompson et al., 1997 ), with manual adjustments as needed. Conserved regions in ITS1 (Liu and Schardl, 1994 ) and ITS2 (Hershkovitz and Zimmer, 1996 ) were used to identify potential pseudogenes and confirm the alignment at those positions. Sequences that did not have these conserved regions were considered to be pseudogenes and were excluded from the phylogenetic analyses. The highly conserved 5.8S was not available for all sequences, so the region was excluded from the phylogenetic analyses. Because homology assessment for several nucleotide positions across distantly related genera was uncertain, we employed a conservative alignment strategy. By setting the gap penalty low, we favored introducing gaps, which created autapomorphies rather than forcing synapomorphies. We also conducted phylogenetic analyses with and without these uncertain regions as described next.

Both parsimony and Bayesian analyses of the ITS sequence data were conducted. For parsimony, heuristic tree searches were performed using PAUP* version 4.0b10 (Swofford, 2002 ) with 1000 random addition replicates, tree bisection reconnection (TBR) branch swapping, ACCTRAN character-state optimization, and gaps coded as missing. To reduce the amount of time spent swapping on suboptimal trees, only five trees were held at each replicate, to an arbitrary maximum of 10 000 trees saved. The best trees were then swapped to completion. Bootstrap support for the internal nodes was determined by 1000 bootstrap replications (Felsenstein, 1985 ) with uninformative characters excluded and using the maximum likelihood parameters estimated from Modeltest version 3.06 (Posada and Crandall, 1998 ) to conduct a neighbor-joining bootstrap.

Bayesian analyses were conducted using MrBayes 3.0 (Huelsenbeck and Ronquist, 2001 ), with the likelihood parameters estimated using Modeltest, the Markov chain Monte Carlo algorithm (Larget and Simon, 1999 ) with four simultaneous chains (three heated and one cold), and trees saved every 100 generations. Two independent runs of two million generations each (corresponding to at least five times the burn-in period) were performed to ensure the analyses converged on the same "plateau." Trees from the burn-in period were discarded, and a 50% majority rule consensus tree was constructed from the remaining trees (Wilcox et al., 2002 ). Posterior probabilities for the clades reconstructed from each independent run were also compared to ensure proper mixing (Huelsenbeck et al., 2002 ).

RESULTS

ITS sequence characteristics and phylogeny reconstruction
The aligned region, including ITS1 and ITS2, contained 644 characters: 166 characters were constant, 98 were parsimony uninformative, and 380 were parsimony informative. ITS1 contributed 158 informative characters while ITS2 had 222. The ITS1 spacer varied in length from 253–297 base pairs (bp), while ITS2 varied from 207–231 bp. The GC content of ITS1 was 46.7–60.1% (mean 53.3%) and ITS2 was 50–67.8% (mean 56.9%).

From parsimony analyses, 10 000 most parsimonious (MP) trees of 2980 steps with a CI = 0.28 (excluding uninformative characters), RI = 0.67, and RC = 0.21 were saved. In the ITS tree, both Gossypieae and Malveae are monophyletic, with Malveae comprised of two main clades. One of the main clades (hereafter referred to as clade A) consists of genera placed in the Abutilon, Anoda, Batesimalva, Gaya, Malacothamnus (in part), Plagianthus, and Sphaeralcea (in part) alliances (Fig. 1). The second large clade (clade B) contains genera from the Anisodontea, Kearnemalvastrum, Malacothamnus, Malope, Malva, Malvastrum, Modiola, Sidalcea, and Sphaeralcea alliances (Fig. 2). Most of the infratribal alliances are not monophyletic and many of the genera are also not monophyletic, including Abutilon, Iliamna, Sida, Tarasa, Tetrasida, and Wissadula. Taxa from two of the alliances are found in both main clades, apart from the remainder of their alliance as currently circumscribed. Sidasodes is placed in clade A, while the remaining Sphaeralcea alliance genera are in clade B, and Neobrittonia is in clade A, whereas the rest of the Malacothamnus alliance is in clade B.

View larger version (58K): [in this window] [in a new window]   Figs. 1–2. Majority rule (50%) consensus of 10 000 MP trees based on ITS sequence data of tribe Malveae. Frequency of reconstructed clades is indicated above the branches and bootstrap support (above 70%) for 1000 replicates is shown below the nodes. Alliance associations are shown at right and follow Bayer and Kubitski (2003) with slight modification (see Table 1 ). Clade A

View larger version (55K): [in this window] [in a new window]   Figs. 3–4. Majority rule (50%) consensus of the post-burn in trees resulting from Bayesian analysis of ITS sequence data. Frequency of clades is shown above the branches and represent Bayesian posterior probabilities. Taxa that change position as compared to the parsimony analysis (Figs. 1 –2 ) are outlined in grey. Clade A

View larger version (52K): [in this window] [in a new window]   Fig. 4. Clade B.

Utility of ITS in tribe Malveae
In this study, we present the first comprehensive phylogeny for tribe Malveae. Studies in other angiosperm families have employed the ITS region for phylogenetic reconstructions at the tribal level, including one other Malvoideae tribe, Gossypieae (Seelanan et al., 1997 ). However, the utility of this region for phylogenetic reconstruction at higher taxonomic levels certainly will depend on the level of divergence for the genera under consideration. Across tribe Malveae, the use of the ITS region for phylogeny reconstruction is likely at its limit, given the alignment difficulties we experienced. For this same reason, the inclusion of genera from other tribes of Malvoideae, most notably the Hibisceae, was not feasible. Similarly, the high homoplasy levels [CI = 0.28 (excluding uninformative characters), RC = 0.21] indicate that this marker may be beyond the limit for a tribal level phylogeny. However, homoplasy levels have been shown to increase when a large number of taxa are analyzed (Sanderson and Donoghue, 1989 ). Moreover, when log transformed values for CI and number of taxa from our study are compared to the regression analyses conducted by Givnish and Sytsma (1997) , our data fall within the range expected for DNA sequence data. The exclusion of troublesome areas in the Malveae alignment from the phylogenetic analyses did not produce conflicting relationships among the taxa, but did result in a lack of resolution for several areas of the tree. As demonstrated by previous studies in the Malveae, however, the ITS region does provide sufficient resolution at lower taxonomic levels (Ray, 1995 ; Andreasen and Baldwin, 2001 , 2003 ; Fuertes Aguilar et al., 2003 ; Tate and Simpson, 2003 ).

The challenges of using ITS for phylogeny reconstruction in groups known or suspected to have experienced hybridization or polyploidization are widely appreciated (Baldwin et al., 1995 ; Wendel et al., 1995 ; Alvarez and Wendel, 2003 ; Fuertes Aguilar and Nieto Feliner, 2003 ). The ITS region, as part of the nuclear ribosomal repeat, is expected to undergo concerted evolution (Zimmer et al., 1980 ), and therefore, the repeats within a given taxon are often assumed to be homogeneous. In some cases, concerted evolution may fail to homogenize the repeats in hybrids or allopolyploids, particularly if these are recently formed entities (and sufficient time has not passed for homogenization of the repeats), if the repeats are located on different chromosomal segments (and interlocus concerted evolution does not occur) or if the hybrid or polyploid reproduces asexually (Baldwin, 1992 ; Alvarez and Wendel, 2003 ). Polyploidy has been well documented in tribe Malveae, not only within genera, but also within species (see Fryxell, 1997 ). Pseudogene formation, biased PCR amplification, and interlocus recombination are just a few of the processes that can potentially confound the use of ITS for phylogeny reconstruction (reviewed in Alvarez and Wendel, 2003 ). Despite these potential shortcomings, the recovered generic relationships based on ITS sequence data are also supported by geography, chromosome number, and morphology (discussed later, Figs. 5, 6). Nonetheless, the findings presented here are preliminary and require further corroboration from one or more independent data sets, either from the chloroplast or from a low- or single-copy nuclear gene. As a conservative measure, we do not propose taxonomic changes for the tribe at present.

View larger version (37K): [in this window] [in a new window]   Figs. 5–6. Summary of generic relationships in tribe Malveae based on ITS sequence data, showing the presence or absence of an epicalyx, geographic distribution, and reported chromosome numbers. See also Table 1 . Clade A

View larger version (30K): [in this window] [in a new window]   Fig. 6. Clade B. Figure Abbreviations: Am, America; Argen, Argentina; Aus, Australia; Calif, California; Carib, Caribbean; Col, Colombia; CR, Costa Rica; Medit, Mediterranean; Mex, Mexico; Tex, Texas; US, United States of America

The non-monophyly of the alliances in the ITS phylogeny is generally consistent with a previous phylogeny based on cpDNA restriction site data for Malveae (La Duke and Doebley, 1995 ). In that study, two main clades were recovered: one containing Abutilon and Sida (Abutilon alliance) and a second containing Alcea, Lavatera, and Malva (Malva alliance), Iliamna and Malacothamnus (Malacothamnus alliance), Modiola and Modiolastrum (Modiola alliance), Sphaeralcea, Tarasa, and Urocarpidium (Sphaeralcea alliance), and Callirhoë; (Sidalcea alliance). The taxon sampling here is more extensive than in the cpDNA study, but essentially the same topology is recovered: one clade (clade A) contains the Abutilon, Anoda, Batesimalva, Gaya, and Plagianthus alliances, with the aforementioned outliers from the Malacothamnus and Sphaeralcea alliances, and the second large clade (clade B) consists of genera from the Anisodontea, Kearnemalvastrum, Malacothamnus, Malope, Malva, Malvastrum, Modiola, Sidalcea, and Sphaeralcea alliances. These two clades correspond primarily to the absence (clade A) or presence (clade B) of involucral bracts subtending individual flowers (epicalyx) (Figs. 5, 6). However, this character is variable in species of Malvella (clade A, Fig. 5) and Callirhoë; (clade B, Fig. 6) and is completely absent in species of Nototriche (clade B, Fig. 6). The loss of an epicalyx in Nototriche clearly represents an independent event, because this genus is firmly placed within clade B. The lability of the presence or absence of an epicalyx in Malvella and Callirhoë; is an interesting question that merits further investigation, particularly from a developmental perspective. Within each of the main clades, other morphological characters (particularly those of the carpel previously emphasized for classification) appear to be quite labile, such that general trends presently cannot be well defined. Similarly, the lack of strong support, both bootstrap values and Bayesian posterior probabilities, at the base of clades A and B, makes rigorous character reconstructions tentative at this time. Early classifications of tribe Malveae (Bentham and Hooker, 1862 , through Hutchinson, 1967 ) emphasized carpel morphology, specifically the number and position of ovules in each carpel. Bates (1968) proposed that the separation of uniovulate and pluriovulate genera into separate tribes was likely artificial and suggested that relationships between uniovulate and pluriovulate lineages should not be disregarded. Our findings based on ITS data, support Bates' astute observation that this character has been over-emphasized. Although paraphyly may be an expected outcome of phylogenetic analyses (e.g., Brummitt, 2002 ), we find support for many of the recovered generic relationships based on chromosome number and geographic distribution (Figs. 5, 6), two criteria used by Bates to delimit the alliances. For the remainder of the discussion, we will focus on the overall pattern of alliance and generic relatedness within clades A and B.

Alliances and genera of clade A
As mentioned, clade A contains those genera that lack involucral bracts and belong to the Abutilon, Anoda, Batesimalva, Gaya, Malacothamnus, Plagianthus, and Sphaeralcea alliances (Figs. 1, 3, 5). The clade as a whole is geographically and chromosomally diverse, with taxa distributed in the Americas and the South Pacific, and most with base chromosome numbers of x = 6, 7, and 8 (few with x = 5, 13, 15). Support for most of the basal nodes is relatively weak (no BS, but 100% BPP for clade A), with only a few receiving >70% BS (Figs. 1, 3). Only one alliance (Anoda) is monophyletic based on the ITS data. Anoda and Periptera were suggested to be closely related (Bates, 1987 ; Fryxell, 1997 ); both genera possess ephemeral mericarp walls (also found in Cristaria, to which they are not closely related) and are primarily distributed in Mexico. Periptera (only one species counted) has a haploid chromosome number of n = 13, while Anoda is more chromosomally diverse with n = 13, 14, 15, 18, 30, or 45 (Bates, 1987 ; Fryxell, 1997 ). Interestingly, Bates (1987) noted that the only n = 13 species of Anoda (A. thurberi) forms a very robust hybrid (in greenhouse crosses) with Periptera punicea (n = 13) and that these may represent a lineage derived within Anoda.

The Batesimalva alliance, composed of six genera (36 species), is dispersed throughout clade A. Bakeridesia and Horsfordia (both n = 15) form a clade sister to the Anoda alliance with good support (83% BS, 100% BPP) (Figs. 1, 3, 5). Both genera possess capitate stigmas and conspicuously ornamented mericarps (Fryxell, 1997 ), but in Horsfordia the wings are apical, and in Bakeridesia the wings are dorsal. No relationship between the two genera was previously suggested. Two other genera of the Batesimalva alliance, Briquetia (n = 7) and Dirhamphis (n = 7, 15), are included in a clade with Hochreutinera (n = 7, placed in the Abutilon alliance), plus Gaya (n = 6, 12, Gaya alliance) and Billieturnera (n = 8, Abutilon alliance). Krapovickas (1970) suggested a close relationship among Dirhamphis, Briquetia, and Hochreutinera, which is supported by the ITS data (100% BS, 100% BPP). Fryxell (1988) placed Dirhamphis, Horsfordia, Batesimalva, and Briquetia in the Batesimalva alliance, primarily on the basis of fruit morphology. Later, Fryxell and Stelly (1993) advised that this alliance might need modification, because new chromosome counts cast doubt on their association with one another. Further, they suggested that the two Dirhamphis species (one n = 7, the other n = 15) may not be congeneric.

In the ITS phylogeny, Fryxellia (n = 8) is at the base of a clade containing many Sida species (n = 6, 7, 8, 14, 16, 28) plus Dendrosida (n = 21, see later), although this clade receives no support (Figs. 1, 3, 5). Fryxell and Valdés (1991) speculated that Fryxellia could be related to Batesimalva or Anoda because it shares some morphological features with each genus. The placement of Neobrittonia (Malacothamnus alliance) as sister to Batesimalva (Batesimalva alliance) and not with the remaining Malacothamnus alliance in clade B (compare Figs. 1, 3 to 2, 4) is supported by several characters including a shared chromosome number of n = 16 (although one species of Batesimalva is n = 12), the absence of an epicalyx (involucral bracts) (Fig. 5), the presence of basal spines on the dehiscent mericarps, rough or warty seeds, and a pubescent staminal column, all of which are lacking in the Malacothamnus alliance. Fryxell (1988 , 1997 ) did not indicate why he thought Neobrittonia should be included in the Malacothamnus alliance, but Bates (1968) originally placed Neobrittonia amongst the other pluriovulate genera of the Abutilon alliance (e.g., Bakeridesia, Herissantia, Pseudabutilon, and Wissadula).

Two of the three genera of the Gaya alliance (x = 6), Lecanophora and Cristaria, group together, while the third genus, Gaya, is well removed from these. Cristaria and Lecanophora have long been allied because they share the unique character of a carpocrater, a cup-shaped structure formed by expanded bases of the carpels, which are fused to the receptacle base (Bates, 1968 ; Fryxell, 1997 ). Although Gaya shares a common chromosome number with these two genera (Fig. 5), based on morphological characters, the genus is relatively isolated from other genera. Bates (1968) and Bates and Blanchard (1970) suggested that the genera of the Gaya alliance actually represent two distinct lineages, one composed of Gaya and the other of Cristaria + Lecanophora, which is supported here. In the ITS phylogeny, Gaya is sister to a clade composed of Dirhamphis, Briquetia, and Hochreutinera, although there is no bootstrap support and only low BPP (75%) for this relationship (Figs. 1, 3). The more inclusive clade of the taxa (with Billieturnera) also had no bootstrap support (Fig. 1), but the BPP was much higher (99%) (Fig. 3).

The Abutilon alliance with 23 genera (400 species) is the largest in the tribe (Bayer and Kubitzki, 2003 ), and its members are also scattered throughout clade A (Figs. 1, 3). Several genera of this alliance are apparently not monophyletic, e.g., Abutilon, Sida, and Tetrasida. The Abutilon-Sida complex was the subject of a recent phylogenetic investigation using ITS (Fuertes Aguilar et al., 2003 ), which also revealed that the two genera were not monophyletic. Sida (100 spp.) has long been recognized as a heterogeneous assemblage (Fryxell, 1985 ). Attempts to create a more natural group have resulted in several segregate genera: Allosidastrum, Bastardiopsis, Billieturnera, Dendrosida, Krapovickasia, Malvella, Meximalva, Rhynchosida, Sidastrum, and Tetrasida (see Fryxell, 1997 ). In the ITS phylogeny, the remaining named Sida species still do not form a monophyletic group (Figs. 1, 3), which is consistent with the treatment of Fuertes Aguilar et al. (2003) and suggests that further taxonomic adjustments are needed. A "core" Sida clade (Fuertes Aguilar et al., 2003 ) was reconstructed with Fryxellia (n = 8) as its sister and Dendrosida (n = 21) derived within it (Fig. 5). These core Sida species have base chromosome numbers of both x = 7 and x = 8, and belong to different sections as outlined by Fryxell (1985) : S. cordifolia (section Cordifoliae), S. turneroides (section Ellipticifoliae), S. aggregata (section Muticae), S. glutinosa and S. urens (section Nelavagae), S. rhombifolia (section Sidae), S. abutifolia (section Spinosae), and S. linifolia (section Stenindae). The remaining species of Sida are distributed throughout clade A (Fig. 1), including S. fibulifera and S. platycalyx (incertae sedis, fide Fuertes Aguilar et al., 2003 ), which are sister to the Sidastrum + Meximalva clade. Fryxell (1997) suggested that Meximalva and Dendrosida were potentially close relatives to Sida, and, in fact, both genera are closely related to species of Sida, but they occur in separate clades. Similarly, S. oligandra (section Oligandrae) is removed from the core Sida species and is either sister to Robinsonella, based on parsimony analyses (Fig. 1), or is unresolved at the base of the larger Abutilon alliance clade in the Bayesian analysis (Fig. 3). Two other Sida species, S. hookeriana (section Hookeriana) and S. hermaphrodita (section Pseudo-Napaea), are included with members of the Plagianthus alliance (100% BS, 100% BPP), plus Sidasodes colombiana of the Sphaeralcea alliance (82% BS) (Figs. 1, 3). Sida hookeriana is found in Australia, so its inclusion with the Plagianthus group is more tenable, although morphologically the two are disparate. The reconstruction of S. hermaphrodita, which is found in the northeastern United States, with the primarily South Pacific taxa of the Plagianthus alliance is somewhat perplexing. Fryxell (1997) suggested that this species might be better segregated into a distinct genus. Fryxell and Fuertes Aguilar (1992) noted the similarity of Sidasodes (from the Andes of Colombia and Peru) to Sida hermaphrodita on the basis of fruit morphology; however, these taxa were not thought to share other features. In the ITS study by Fuertes Aguilar et al. (2003) , S. hermaphrodita, S. hookeriana, and Sidasodes also formed a clade sister to the other members of the Abutilon and Sida alliances, a finding that is corroborated here. Chloroplast sequence data also support the sister relationship of S. hermaphrodita and S. hookeriana to genera of the Abutilon and Sida alliances (J. Beck, R. Small, University of Tennessee, personal communication). Further evaluation of these three species is needed to determine their systematic position.

One of the other two genera from the Abutilon alliance that is not monophyletic is Abutilon (160 spp.), one of the largest Malveae genera. Like Sida, several species were removed and new genera created, including Bakeridesia, Bastardia, Corynabutilon, Herissantia, Hochreutinera, Pseudabutilon, and Tetrasida (Fryxell, 1997 ), most of which were included in the present study. Only three species of Abutilon were sampled here; these do not form a monophyletic group, but are paraphyletic to Bastardia and Bastardiopsis, two of the segregate genera (Figs. 1, 3). These last two genera are the only ones in the tribe that possess capsular fruits; all other genera are schizocarpous (Fryxell, 1997 ). Expanded sampling within Abutilon certainly will be needed to determine if other species should be removed and elevated to generic status.

The third genus of the Abutilon alliance resolved as non-monophyletic in the ITS phylogeny is Tetrasida. The genus is chromosomally unknown and currently contains five species (Fryxell and Fuertes Aguilar, 1992 ; Fryxell, 2002 ) found in Peru and Ecuador. Two species were included here to represent the genus: T. chachapoyensis clusters with species of Wissadula (n = 7), while T. weberbaueri is placed sister to Allowissadula holosericea (n = 8) (Figs. 1, 3, 5). Krapovickas (1969) included the species now considered as Tetrasida in Abutilon section Tetrasida because he believed the condition of a four-merous corolla (for which the genus was named) in the species was not sufficiently consistent to merit generic recognition. However, Fryxell and Fuertes Aguilar (1992) resurrected the genus, including two species, and later described three new species (Fryxell, 2002 ).

The Plagianthus alliance contains two genera from Australia (Gynatrix, Lawrencia), two from New Zealand (Hoheria, Plagianthus), and one from Tasmania (Asterotrichion) for a total of 23 species. Only Hoheria and Plagianthus have chromosome counts available and both are n = 21 (Bates and Blanchard, 1970 ). As discussed earlier, in the ITS phylogeny (Figs. 1, 3), this alliance forms a moderately supported clade (82% BS; 68% BPP) with Sidasodes colombiana (from the Andes of Colombia and Peru), Sida hermaphrodita (section Pseudo-Napaea, from the eastern United States), and S. hookeriana (section Hookerianae, from southwestern Australia). The genera of the Plagianthus alliance are morphologically diverse, ranging from annual herbs to prostrate subshrubs (Lawrencia) and large trees (Plagianthus and Hoheria) that differ considerably in flower and fruit structure (Melville, 1966 ; Lander, 1984 ). In this group, there is a tendency towards dioecy and a reduction in the number of locules in the ovary. Plagianthus is unilocular with a single (rarely two) pendulous ovule in each flower. The styles also show a gradation from the long linear stigmas of Lawrencia and Gynatrix to clavate forms in Plagianthus, Asterotrichion, and Hoheria.

Alliances and genera of clade B
Clade B was resolved as a well-supported group (81% BS, 100% BPP) and is composed of genera from the Anisodontea, Kearnemalvastrum, Malacothamnus, Malope, Malva, Malvastrum, Modiola, Sidalcea, and Sphaeralcea alliances (Figs. 2, 4). As mentioned, all members of this clade retain the symplesiomorphic character of having an epicalyx (with the exception of Nototriche, which lacks involucral bracts, but clearly belongs in this clade) (Fig. 6). Clade B contains primarily American taxa, but also includes European, Asian, and South African genera. The predominant base chromosome number is x = 5 (Sphaeralcea alliance, Sidalcea, Modiolastrum), although some clades are complex chromosomally (e.g., the clade that includes Anisodontea through Callirhoë;, with n = 12, 13, 14, 21, 22, etc.) (Fig. 6). As with clade A, many of the basal nodes in clade B lack robust support (Figs. 2, 4). Relationships with the greatest support are those between congeneric taxa, although there is strong support for the Malope + Navaea + Malva + Lavatera clade (91% BS; 100% BPP) and the Sphaeralcea + Tarasa + Nototriche clade (80% BS; 100% BPP) (Figs. 2, 4). Three alliances in clade B are monophyletic: the Modiola alliance (78% BS, 76% BPP) composed of Modiola and Modiolastrum, and the monogeneric Kearnemalvastrum and Malvastrum alliances, although the sister groups to these latter alliances are not well supported.

The Sphaeralcea alliance, the largest of clade B with 12 genera (230 species) (excluding Sidasodes, which is better aligned with genera of clade A), is not monophyletic (Figs. 2, 4). A clade composed of Andeimalva (n = 6), along with Nototriche, Sphaeralcea, and Tarasa (all x = 5) is sister to the rest of clade B. Other genera of the Sphaeralcea alliance occur in a grade (Palaua, Urocarpidium, Fuertesimalva, Acaulimalva), with the remaining genera (Eremalche, Calyculogygas, Monteiroa, and Napaea) scattered amongst genera from other alliances. In the case of Eremalche, the sister relationship to Sidalcea (Sidalcea alliance) is supported by geographic distribution (both are found primarily in California and northern Mexico), a morphological similarity and a shared basic chromosome number of x = 5 (Fig. 6). Two other genera of the Sphaeralcea alliance, Calyculogygas (n = 5) and Monteiroa (n = 10), are in a clade with Modiola (n = 9) and Modiolastrum (n = 5, 15, 50). The placement of these two genera with Modiolastrum is plausible given that they are all found in eastern South America (Argentina, Brazil, and Uruguay) and have a common base chromosome number (Fig. 6). Modiola is a monotypic genus that is widespread throughout pantropical America, extending into temperate areas, and is thought to represent an aneuploid lineage closely related to the x = 5 genera of the Sphaeralcea alliance (Bates, 1968 ). Krapovickas (1945) first suggested the close relationship of Modiola and Modiolastrum based on gross morphological features, and later noted that even their chromosomes were similar in size and satellite morphology (Krapovickas, 1949 ). Both genera were originally included in the Sphaeralcea alliance (Bates, 1968 ; Bates and Blanchard, 1970 ), but were later separated into their own generic alliance by Fryxell (1988) . One other genus in the Sphaeralcea alliance that is relatively isolated is Napaea (n = [14], 15), a monotypic dioecious genus from the central United States. Although originally allied to Sidalcea (Iltis and Kawano, 1964 ; Bates, 1968 ), Napaea was later segregated into its own alliance by Bates and Blanchard (1970) . Krebs (1993) noted that Napaea dioica shared pollen and fruit characters with Sphaeralcea and suggested that it was aligned better with genera of the Sphaeralcea alliance. In the ITS phylogeny, Napaea is sister to the cytologically complex genus Callirhoë; (Sidalcea alliance, see Table 1), which is found in the central United States to northeastern Mexico (Dorr, 1990 ) (Fig. 6). Interestingly, gynodioecy, a rather rare phenomenon in the Malveae, occurs in several species of Callirhoë; (Dorr, 1990 ). Although Napaea and Callirhoë; were placed in separate alliances, a close relationship between them was suggested (Fryxell, 1997 ). As is found throughout the Malveae, the ITS phylogeny indicates that the phylogenetic relationships of many genera of the Sphaeralcea alliance are supported more by geography and shared chromosome numbers than some previously emphasized morphological characters.

Within the Sphaeralcea alliance, only Tarasa (n = 5, 10) is not monophyletic in the ITS phylogeny (Figs. 2, 4), a finding consistent with a previous study based on ITS and chloroplast sequence data (Tate and Simpson, 2003 ; Tate, 2003 ). Sphaeralcea (n = 5, 10, 15, 25) is sister (80% BS, 100% BPP) to a clade containing Tarasa (n = 5, 10) and Nototriche (n = 5, 10, 15, 20) (no support for the next node, but the clade including T. thyrsoidea and T. operculata has 100% BS, 100% BPP) (Figs. 2, 4, 6). A close relationship among these three genera was proposed by Krapovickas (1960 , 1971 ), but they were retained as separate genera because they were considered to be distinct from one another. Species of Sphaeralcea (40 spp.) are herbs or shrubs found in temperate mid-elevation habitats of North and South America (Chile and Argentina) with one- to three-seeded mericarps that have a dehiscent, smooth upper portion and an indehiscent, laterally reticulate lower portion (Fryxell, 1997 ). Tarasa species (27 spp.) are either annuals or perennial shrubs found at mid (800 m) to high (up to 4000 m) elevations in the Andes (Peru to Chile and Argentina) and have one-seeded mericarps that are completely dehiscent (Krapovickas 1954 , 1960 ). Nototriche (100 spp.) contains primarily acaulescent cushion plants (although a handful of annual species have been described) found above 4000 m in the Andes from Ecuador to southern Chile and Argentina (Fryxell, 1997 ) and has one-seeded, dehiscent mericarps. Interestingly, the lower elevation perennial species of Tarasa are more similar morphologically to Sphaeralcea, while the high elevation annuals are more similar to Nototriche (Tate and Simpson, 2003 ). Additional data will be needed to define the boundaries of Tarasa and Nototriche or to determine if Nototriche should be considered a section of Tarasa.

Another finding in the ITS phylogeny related to Tarasa is the placement of Urocarpidium albiflorum with Fuertesimalva (Figs. 2, 4). This species, the type of the genus Urocarpidium, was suggested to be synonymous with Tarasa operculata due to its apically plumose awns on the mericarps (Fryxell, 1996 ). The genus Fuertesimalva was created to accommodate the remaining species of Urocarpidium (Fryxell, 1996 ) that do not possess this character. The results of the ITS phylogeny (and also chloroplast data, Tate and Simpson, 2003 ), do not support the separation of U. albiflorum from the remaining species of Fuertesimalva, nor its inclusion in Tarasa and argue for the original generic composition and name. Morphological characters that support the placement of U. albiflorum with Fuertesimalva rather than Tarasa include mericarps that are indehiscent, glabrous, and laterally "ridged" (vs. dehiscent, with stellate pubescence on the dorsal and apical surfaces, and the lateral walls that are smooth or faintly reticulate in Tarasa), and calyx trichomes that are simple and hirsute (vs. stellate stipitate in Tarasa). Thus, the occurrence of an apical awn on the mericarps of U. albiflorum and Tarasa species appears to be a convergent character, and we recognize the former as separate from the latter.

Most included members of the Malacothamnus alliance (Iliamna, Malacothamnus, and Phymosia, excluding Neobrittonia, which is a member of clade A), form a clade in the ITS phylogeny (Figs. 2, 4). However, Iliamna was also reconstructed as paraphyletic. Two species of Iliamna (n = 33) (I. bakeri and I. latibracteata) that are endemic to northern California/southern Oregon are more closely related to Sidalcea (n = 10, 20, 30) and Eremalche (n = 10, 20), which are distributed along the western coast of North America, than to the remaining members of Iliamna (Fig. 6). The other two species included here (I. rivularis, found in the Rocky Mountains of the United States, and I. remota, found in Illinois, Indiana, and Virginia), cluster with Phymosia umbellata (Mexico, Guatemala, and Caribbean) (n = 17) and Malacothamnus fasciculatus (California) (n = 17). Morphologically, Iliamna is distinct from Sidalcea and Eremalche. Characters distinguishing Iliamna from both genera include a perennial habit (annual Eremalche, annual or perennial in Sidalcea), deciduous stipules (persistent in both Sidalcea and Eremalche), carpels with multiple seeds (single in Sidalcea and Eremalche), and dehiscent mericarps (indehiscent in Eremalche). Chloroplast (rpl16 intron and trnL-F spacer) sequence data place Iliamna bakeri and I. latibracteata in a clade with other western Iliamna species, while I. remota and I. corei are outside this "core" Iliamna clade with Phymosia (T. Bodo Slotta, unpublished data).

The Malvastrum alliance contains a single genus, Malvastrum, which, like Abutilon and Sida, was at one time a repository for many taxa that were difficult to place. Over the years, however, several species were removed from the heterogeneous Malvastrum and placed in other genera, including Acaulimalva, Anisodontea, Malacothamnus, Monteiroa, Nototriche, Sphaeralcea, Tarasa, and Urocarpidium (Fuertesimalva) (Hill, 1982 ; Fryxell, 1997 ). Since Hill's (1982) revision of Malvastrum, the genus is a cohesive American taxon of 15 species that share a base chromosome number of x = 6 (Fig. 6). Within the Malveae, Malvastrum appears to be rather isolated, but based on the ITS phylogeny, it is closely related to other North (Eremalche, Sidalcea) or South (Calyculogygas, Modiola, Modiolastrum, Monteiroa) American genera (Figs. 2, 4).

Like many of the other alliances in clade B, the Malope alliance, comprised of Malope and Kitaibela, is not monophyletic. Previously, these two genera, along with Palaua (Sphaeralcea alliance), were placed in a separate tribe Malopeae, because they shared the unique feature of multiverticillar carpels (Table 1). All three are members of clade B, but are not closely related to one another (Figs. 2, 4) , which suggests that this unusual morphological feature has evolved on three separate occasions. Bates (1968) placed Palaua in the Sphaeralcea alliance with the other x = 5 genera, while retaining Kitaibela and Malope as the sole members of the Malope alliance. He also proposed that the evolution of the carpels into superposed verticils in Palaua occurred independently from that of Malope and Kitaibela, a hypothesis supported here. In the ITS phylogeny, Malope (n = 22, 25) is included in a clade (91% BS, 100% BPP) with some members of the Malva alliance (Lavatera, Malva, and Navaea), which share a geographic distribution in the Mediterranean region (Fig. 6). Kitaibela (n = 21, 22) is sister to Alcea (n = 13, 21) from the Malva alliance (96% BS, 100% BPP); both of these genera are found in the Mediterranean, East European, and West Asian regions. Bates' (1968) discussion of the Malva, Malope, and Anisodontea alliances was included in the same section, as he suspected their close relationship.

The Malva alliance (five genera, 100 species), as alluded to in the previous paragraph, is not monophyletic (Figs. 2, 4). The genera of this alliance are found predominantly in the Mediterranean/European region (although some Lavatera species occur in North America and Australia) and have various multiples of an x = 7 base chromosome number, which is shared by most of the other closely related genera based on the ITS data (Fig. 6). Lavatera, Malva, and Navaea phoenicea form a strongly supported clade (94% BS, 100% BPP) with Malope as its sister (Figs. 2, 4). Two species of Alcea cluster together (100% BS, 100% BPP) and are sister to Napaea dioica. In a previous ITS phylogenetic study, Ray (1995) found that the individual genera Malva and Lavatera were not monophyletic, but that the North American species of Lavatera were more closely related to Malva than to the Old World Lavatera species. Subsequently, the New World species of Lavatera were transferred to Malva (Ray, 1998 ). The sister position of Navaea with Lavatera and Malva is also supported by cpDNA sequence data (Fuertes Aguilar et al., 2002 ). Morphological groups (Lavateroid and Malvoid groups) were outlined by Ray (1995) , but clearly these two genera will require more extensive sampling of morphological and molecular data to sort out their boundaries.

General conclusions
Tribe Malveae is a geographically, chromosomally, and morphologically diverse clade. The ITS phylogeny presented here shows that the current circumscription of the tribe into 14 generic alliances is artificial. Instead, two clades can be defined by the presence or absence of involucral bracts, and, perhaps, only these two clades should be named formally. Given the lack of support for many nodes at the base of the tree, additional data (chloroplast and/or other nuclear markers) are needed to corroborate these relationships. Likewise, because many genera (and clades) contain polyploid or aneuploid lineages, more data will likely give insight into chromosome evolution, which already appears to have been rather complex within the tribe. Moreover, determination of the early-branching lineages of the Malveae should shed light on the biogeographic origin of the tribe, which centers in the Americas, but also contains South Pacific and European taxa, and the base chromosome number for the tribe, which has been postulated as x = 8 or 9 (Bates, 1968 ).

View larger version (54K): [in this window] [in a new window]   Fig. 2. Clade B.

¹ The authors thank the curators and staff of the following herbaria for use of herbarium specimens for this study: CHR, LL, MA, MO, NY, and TEX; and R. Small and two anonymous reviewers for comments on the manuscript. This research was supported by a National Science Foundation (NSF) Doctoral Dissertation Improvement Grant to JAT and BBS (DEB-9902230); the Foundation for Research, Science, and Technology to SJW; the Virginia Academy of Science to TABS; an NSF grant to JCL, Christopher Austin, Siegfried Detke, and Kevin Young (DBI-0115985); an NSF grant to JCL and Paul Fryxell (DEB-9420233), and a research grant of MCYT (REN2002–00339) to JFA. The authors each individually express their gratitude to Paul Fryxell for his assistance, knowledge, and generosity throughout many years of studying the Malvaceae.

⁷ Author for reprint requests (e-mail: jtate{at}ufl.edu ). Current address: Florida Museum of Natural History, University of Florida, P.O. Box 117800, Gainesville, Florida 32611 USA

LITERATURE CITED

Alvarez I. J. F. Wendel 2003 Ribosomal ITS sequences and plant phylogenetic inference. Molecular Phylogenetics and Evolution 29: 417-434[CrossRef][Web of Science][Medline]

Alverson W. S. B. A. Whitlock R. Nyffeler C. Bayer D. A. Baum 1999 Phylogeny of the core Malvales: evidence from ndhF sequence data. American Journal of Botany 86: 1474-1486[Abstract/Free Full Text]

Alverson W. S. K. G. Karol D. A. Baum M. W. Chase S. M. Swensen R. McCourt K. J. Sytsma 1998 Circumscription of the Malvales and relationships to other Rosidae: evidence from rbcL sequence data. American Journal of Botany 85: 876-887[Abstract]

Andreasen K. B. G. Baldwin 2001 Unequal evolutionary rates between annual and perennial lineages of checker mallows (Sidalcea, Malvaceae): evidence from 18S–26S rDNA internal and external transcribed spacers. Molecular Biology and Evolution 18: 936-944[Abstract/Free Full Text]

Andreasen K. B. G. Baldwin 2003 Reexamination of relationships, habital evolution, and phylogeography of checker mallows (Sidalcea; Malvaceae) based on molecular phylogenetic data. American Journal of Botany 90: 436-444[Abstract/Free Full Text]

Baldwin B. G. 1992 Phylogenetic utility of the internal transcribed spacers of nuclear ribosomal DNA in plants: an example from the Compositae. Molecular Phylogenetics and Evolution 1: 3-16[CrossRef][Medline]

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][Web of Science]

Bates D. M. 1968 Generic relationships in the Malvaceae, tribe Malveae. Gentes Herbarum 10: 117-135

Bates D. M. 1987 Chromosome numbers and evolution in Anoda and Periptera (Malvaceae). Aliso 11: 523-531

Bates D. M. O. J. Blanchard Jr 1970 Chromosome numbers in the Malvales. II. New or otherwise noteworthy counts relevant to classification in the Malvaceae, tribe Malveae. American Journal of Botany 57: 927-934[CrossRef][Web of Science]

Baum D. A. S. DeWitt Smith A. Yen W. S. Alverson R. Nyffeler B. A. Whitlock R. L. Oldham 2004 Phylogenetic relationships of Malvatheca (Bombacoideae and Malvoideae; Malvaceae sensu lato) as inferred from plastid DNA sequences. American Journal of Botany 91: 1863-1871[Abstract/Free Full Text]

Bayer C. K. Kubitzki 2003 Malvaceae. In K. Kubitzki and C. Bayer [eds.], Flowering plants, dicotyledons: Malvales, Capparales, and non-betalain Caryophyllales, 225–311. Springer-Verlag, Berlin, Germany

Bayer C. M. F. Fay A. Y. D. Bruijn V. Savolainen C. M. Morton K. Kubitzki W. S. Alverson M. W. Chase 1999 Support for an expanded family concept of Malvaceae within a recircumscribed order Malvales: a combined analysis of plastid atpB and rbcL DNA sequences. Botanical Journal of the Linnean Society 129: 267-303[CrossRef]

Bentham G. J. D. Hooker 1862 Malvaceae. In G. Bentham and J. D. Hooker [eds.], Genera plantarum, 195–213. London, England

Brizicky G. K. 1968 Herissantia, Bogenhardia, and Gayoides (Malvaceae). Journal of the Arnold Arboretum 49: 278-279[Web of Science]

Brummitt R. K. 2002 How to chop up a tree. Taxon 51: 31-41[CrossRef][Web of Science]

Chase M. W. H. H. Hills 1991 Silica gel: an ideal material for field preservation of leaf samples for DNA studies. Taxon 40: 215-220[CrossRef][Web of Science]

Dorr L. J. 1990 A revision of the North American genus Callirhoë; (Malvaceae). Memoirs of the New York Botanical Garden 56: 1-76

Doyle J. J. J. L. Doyle 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15

Edlin H. L. 1935 A critical revision of certain taxonomic groups of the Malvales. The New Phytologist 34: 1-20, 122–143 [CrossRef]

Felsenstein J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791[CrossRef][Web of Science]

Fernandez A. A. Krapovickas G. Lavia G. Seijo 2003 Cromosomas de Malváceas. Bonplandia 12: 141-145

Fryxell P. A. 1968 A redefinition of the tribe Gossypieae. Botanical Gazette 129: 296-308[CrossRef]

Fryxell P. A. 1985 Sidus sidarum. V. The North and Central American species of Sida. Sida 11: 62-91

Fryxell P. A. 1988 Malvaceae of Mexico. Systematic Botany Monographs 25: 1-255

Fryxell P. A. 1996 Fuertesimalva, a new genus of Neotropical Malvaceae. Sida 17: 69-76

Fryxell P. A. 1997 The American genera of Malvaceae. II. Brittonia 49: 204-269[CrossRef][Web of Science]

Fryxell P. A. 2002 An Abutilon nomenclator (Malvaceae). Lundellia 5: 79-118

Fryxell P. A. J. Valdés 1991 Observations on Fryxellia pygmaea (Malvaceae). Sida 14: 399-404

Fryxell P. A. J. Fuertes Aguilar 1992 A re-evaluation of the Abutilothamnus complex (Malvaceae). I. Two new species and two new genera, Sidasodes and Akrosida. Brittonia 44: 436-447[CrossRef][Web of Science]

Fryxell P. A. D. M. Stelly 1993 Additional chromosome counts in the Malvaceae. Sida 15: 639-647

Fuertes Aguilar J. P. A. Fryxell R. K. Jansen 2003 Phylogenetic relationships and classification of the Sida generic alliance based on nrDNA ITS evidence. Systematic Botany 28: 352-364[Web of Science]

Fuertes Aguilar J. M. F. Ray J. Francisco-Ortega A. Santos-Guerra R. K. Jansen 2002 Molecular evidence from chloroplast and nuclear markers for multiple colonizations of Lavatera (Malvaceae) in the Canary Islands. Systematic Botany 27: 74-83[Web of Science][Medline]

Fuertes Aguilar J. G. Nieto Feliner 2003 Additive polymorphisms and reticulation in an ITS phylogeny of thrifts (Armeria, Plumbaginaceae). Molecular Phylogenetics and Evolution 28: 430-447[CrossRef][Web of Science][Medline]

Givnish T. J. K. J. Sytsma 1997 Homoplasy in molecular vs. morphological data: the likelihood of phylogenetic inference. In T. J. Givnish and K. J. Sytsma [eds.], Molecular evolution and adaptive radiation, 55– 101. Cambridge University Press, Cambridge, UK

Hershkovitz M. A. E. A. Zimmer 1996 Conservation patterns in angiosperm rDNA ITS2 sequences. Nucleic Acids Research 24: 2857-2867[Abstract/Free Full Text]

Hill S. R. 1982 A monograph of the genus Malvastrum A. Gray (Malvaceae: Malveae). Rhodora 84: 1-83, 159–264, 317–409

Huelsenbeck J. P. F. Ronquist 2001 MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754-755[Abstract/Free Full Text]

Huelsenbeck J. P. B. Larget R. E. Miller F. Ronquist 2002 Potential applications and pitfalls of Bayesian inference of phylogeny. Systematic Biology 51: 673-688[CrossRef][Web of Science][Medline]

Hutchinson J. 1967 The genera of flowering plants. Clarendon Press, Oxford, England

Iltis H. H. S. Kawano 1964 Cytotaxonomy of Napaea dioica (Malvaceae). The American Midland Naturalist 72: 76-81[CrossRef]

Judd W. S. S. R. Manchester 1997 Circumscription of Malvaceae (Malvales) as determined by a preliminary cladistic analysis of morphological, anatomical, palynological, and chemical characters. Brittonia 49: 384-405[CrossRef][Web of Science]

Kearney T. H. 1949 Malvaceae: a new subtribe and genus, and new combinations. Leaflets of Western Botany 5: 189-191

Kearney T. H. 1951 The American genera of Malvaceae. American Midland Naturalist 46: 93-131[CrossRef][Web of Science]

Krapovickas A. 1945 Nota sobre el género Modiolastrum en la Argentina. Revista Argentina de Agronomía 12: 38-44

Krapovickas A. 1949 Relación entre número cromosómico y área en el género Modiolastrum (Malvaceae). Lilloa 19: 121-125

Krapovickas A. 1954 Sinopsis del género Tarasa (Malvaceae). Boletín de la Sociedad Argentina de Botánica 5: 113-143

Krapovickas A. 1960 Poliploidía y área en el género Tarasa. Lilloa 30: 233-249

Krapovickas A. 1969 Notas sobre el género Abutilon Mill. (Malvaceae). I. La sección Tetrasida (Ulbr.) Krapov. Bonplandia 3: 25-47

Krapovickas A. 1970 Dos géneros nuevos de Malvaceas: Dirhamphis y Hochreutinera, con notas sobre los afines Briquetia y Neobrittonia. Darwiniana 16: 219-232

Krapovickas A. 1971 Evolución del género Tarasa (Malvaceae). In R. H. Mejía and J. A. Moguilevsky [eds.], Recientes adelantos en biología, 232–241. Bona, Buenos Aires, Argentina

Krapovickas A. 2003 Tropidococcus Krapov., nuevo género de Malváceas. Bonplandia 12: 63-66

Krebs G. 1993 Zur taxonomischen Stellung von Napaea dioica L. Feddes Repertorium 104: 465-467

La Duke J. C. J. Doebley 1995 A chloroplast DNA based phylogeny of the Malvaceae. Systematic Botany 20: 259-271

Lander N. S. 1984 Revision of the Australian genus Lawrencia Hook. (Malvaceae: Malveae). Nuytsia 5: 201-271

Larget B. D. L. Simon 1999 Markov chain Monte Carlo algorithms for the Bayesian analysis of phylogenetic trees. Molecular Biology and Evolution 16: 750-759[Web of Science]

Liu J.-S. C. L. Schardl 1994 A conserved sequence in internal transcribed spacer 1 of plant nuclear rRNA genes. Plant Molecular Biology 26: 775-778[CrossRef][Web of Science][Medline]

Melville R. 1966 Contributions to the flora of Australia. VII. Generic delimitation in the Plagianthus complex. Kew Bulletin 20: 511-516[CrossRef]

Pfeil B. E. C. L. Brubaker L. A. Craven M. D. Crisp 2002 Phylogeny of Hibiscus and the tribe Hibisceae (Malvaceae) using chloroplast DNA sequences of ndhF and the rpl16 intron. Systematic Botany 27: 333-350[Web of Science]

Posada D. K. A. Crandall 1998 MODELTEST: testing the model of DNA substitution. Bioinformatics 14: 817-818[Abstract/Free Full Text]

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. Novon 8: 288-295[CrossRef][Web of Science]

Sanderson M. J. M. J. Donoghue 1989 Patterns of variation in levels of homoplasy. Evolution 43: 1781-1795[CrossRef][Web of Science]

Schumann K. 1890 Malvaceae. In A. Engler and K. Prantl [eds.], Die naturlichen pflanzenfamilien, 30–53. Wilhelm Engelmann, Leipzig, Germany

Seelanan T. A. Schnabel J. Wendel 1997 Congruence and consensus in the cotton tribe (Malvaceae). Systematic Botany 22: 259-290[CrossRef][Web of Science]

Swofford D. L. 2002 PAUP*: phylogenetic analysis using parsimony (* and other methods). Sinauer, Sunderland, Massachusetts, USA

Tamura K. M. Nei 1993 Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molecular Biology and Evolution 10: 512-526[Abstract]

Tate J. A. 2003 Andeimalva, a new genus of Malvaceae from Andean South America. Lundellia 6: 10-18

Tate J. A. B. B. Simpson 2003 Paraphyly of Tarasa (Malvaceae) and diverse origins of the polyploid species. Systematic Botany 28: 723-737[Web of Science]

Thompson J. D. T. J. Gibson F. Plewniak F. Jeanmougin D. G. Higgins 1997 The CLUSTAL X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 25: 4876-4882[Abstract/Free Full Text]

Wendel J. F. A. Schnabel T. Seelanan 1995 Bidirectional interlocus concerted evolution following allopolyploid speciation in cotton (Gossypium). Proceedings of the National Academy of Sciences, USA 92: 280-284[Abstract/Free Full Text]

Whittall J. A. Liston S. Gisler R. J. Meinke 2000 Detecting nucleotide additivity from direct sequences is a SNAP: an example from Sidalcea (Malvaceae). Plant Biology (Stuttgart) 2: 211-217[CrossRef]

Wilcox T. P. D. J. Zwickl T. A. Heath D. M. Hillis 2002 Phylogenetic relationships of the dwarf boas and a comparison of Bayesian and bootstrap measures of phylogenetic support. Molecular Phylogenetics and Evolution 25: 361-371[CrossRef][Web of Science][Medline]

Zimmer E. A. S. L. Martin S. M. Beverley Y. W. Kan A. C. Wilson 1980 Rapid duplication and loss of genes coding for the alpha chains of hemoglobin. Proceedings of the National Academy of Sciences, USA 77: 2158-2162[Abstract/Free Full Text]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Abstract Full Text (PDF) Supplementary Data Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Tate, J. A. Articles by Simpson, B. B. Search for Related Content PubMed Articles by Tate, J. A. Articles by Simpson, B. B. Agricola Articles by Tate, J. A. Articles by Simpson, B. B. Social Bookmarking                            What's this?