| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
(American Journal of Botany. 2008;95:985-1005.) doi: 10.3732/ajb.2007313 © 2008 Botanical Society of America, Inc. |
What's this? |
Systematics and Phytogeography |
2 Departamento de Biologia, Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade de São Paulo, Av. Bandeirantes 3900 14040-901, Ribeirão Preto-SP, Brazil 3 Instituto de Biociências, Universidade de São Paulo, Caixa Postal 11461 05499 São Paulo, SP, Brazil 4 The New York Botanical Garden, Bronx, New York 10458-5126 USA
Received for publication 1 October 2007. Accepted for publication 8 April 2008.
ABSTRACT
Primarily known only by the edible fruits of Citrus, Rutaceae comprise a large (c. 160 genera and 1900 species), morphologically diverse, cosmopolitan family. Of its extraordinary array of secondary chemical compounds, many have medicinal, antimicrobial, insecticidal, or herbicidal properties. To assist with the much-needed suprageneric reclassification and with studies of evolution of chemical compounds and biogeographic history of the family, here we included sequence data (from two noncoding regions of the chloroplast genome—rps16 intron and trnL-trnF region) from 65 species in 59 genera (more than one third of those in the family) that represented all subfamilies and tribes and more genera of Toddalioideae and of neotropical groups than previous studies. Results confirmed that Cneorum, Ptaeroxylon, Spathelia, and Dictyoloma form a clade sister to the remaining Rutaceae, none of the subfamilies with more than one genus (except Aurantioideae) is monophyletic, and characters of the ovary and fruit are not reliable for circumscription of subfamilies. Furthermore, clades are better correlated with geographic distributions of the genera than with ovary and fruit characters. Circumscriptions of subfamilies and tribes (and some subtribes of Rutoideae) must be reevaluated. Results are discussed in light of geographic distributions, caryology, chemotaxonomy, and other molecular studies.
Key Words: Cneoraceae • cpDNA • phylogeny • Ptaeroxylaceae • rps16 • Rutaceae • trnL-trnF
Rutaceae are known to most by a single genus, Citrus, because of its commercially important, widely consumed fruits. It is, however, a very large family (c. 160 genera and 1900 species) with great diversity in morphological characters and a worldwide distribution. In addition, Rutaceae are famous among phytochemists for their extraordinary array of secondary chemical compounds—among them, limonoids, flavonoids, coumarins, volatile oils, and alkaloids. Considering their wide range of alkaloids, Price (1963)
labeled the Rutaceae as the most chemically versatile of all plant families. Many papers have been published on Rutaceous compounds; some compounds have proven to be medicinally useful (e.g., Holmstedt et al., 1979; Moraes et al., 2003), and others have potential, for example, as natural pesticides (e.g., Oliva et al., 2000), herbicides (e.g., Aliotta et al., 1996), and antimicrobials (e.g., Mandalari et al., 2007). A firm phylogeny of the family would facilitate not only the study of the evolution of chemical compounds but also the search for useful compounds.
Results of molecular studies have had a deep impact on the understanding of the relationships in many groups of angiosperms (see, e.g., Soltis et al., 2005; Judd et al., 2008). Several recent molecular studies of the Rutaceae (e.g., Chase et al., 1999; Scott et al., 2000; Samuel et al., 2001; Morton et al., 2003; Poon et al., 2007) have shed light on relationships within and among some of the groups of genera, but the phylogeny is incomplete. The current study contributes to a phylogeny of the family by including representatives of genera in all subfamilies and tribes, more genera from the problematic Toddalioideae, and, for the first time, some genera of the subtribe Galipeinae (Rutoideae), the most diverse group in the neotropics.
The family is positioned in Sapindales (APG, 2003), with strong support from molecular data (Gadek et al., 1996), with the major families Simaroubaceae, Sapindaceae, and Meliaceae. Morphological synapomorphies of the order include estipulate, compound leaves, and a well-developed nectary disk. Rutaceae share with Meliaceae and Simaroubaceae bitter, triterpenoid compounds (Waterman, 1983), but are distinguished from them by glandular-punctate leaves and secretory cavities containing aromatic ethereal oils scattered in almost all organs (Judd et al., 2008). Within the Rutaceae, habit varies from herbs to trees; leaves are simple or compound and alternate or opposite; corollas range from a few millimeters to several centimeters long and from actinomorphic to zygomorphic; fertile stamens and carpels range from many to two; carpels are united or free; the fruit is dry or fleshy, dehiscent or indehiscent, winged or not; and a carpel can contain two to several ovules.
Engler (1874) presented the first infrafamilial classification of Rutaceae in Flora Brasiliensis, with further additions in Die Natürlichen Pflanzenfamilien (Engler, 1896, 1931). In the classification of 1931 (maintained with slight modifications by Scholz, 1964, and summarized in Appendix 1 of the present paper), Engler defined seven subfamilies based mainly on the degree of connation and number of carpels, fruit characters (e.g., dehiscent vs. indehiscent, fleshy vs. dry, winged or not), and histology of the glands. A detailed discussion of the characteristics of the subfamilies can be found in Chase et al. (1999). Of these, Rhabdodendroideae has been excluded from Rutaceae (see Prance, 1968, 1972; Fay et al., 1997); three are small—Spathelioideae and Dictyolomatoideae (each with one genus) and Flindersioideae (two genera); and three are large—Aurantioideae (not Citroideae, see Mabberley, 1998, p. 333), Toddalioideae, and Rutoideae. Recent molecular (Chase et al., 1999; Samuel et al., 2001) and chromosomal (Stace, 1993; Guerra et al., 2000) data support the monophyly of Aurantioideae. In contrast, a preponderance of morphological (Hartley, 1974, 1981, 1982), molecular (Chase et al., 1999; Scott et al., 2000), chromosomal (Stace, 1993), and phytochemical (Da Silva et al., 1988) evidence indicates that Toddalioideae and Rutoideae are not monophyletic.
Although Engler divided the Toddalioideae into a single tribe with six subtribes (of which two are now considered to be synonymous with others), he divided the Rutoideae into five tribes and 15 subtribes. He defined these tribes of Rutoideae mainly by habit, presence/absence of endosperm, cotyledon characteristics, and geographic distribution (see Table 1). Data from studies of secondary metabolites (Da Silva et al., 1988) and DNA sequences (Samuel et al., 2001; Morton et al., 2003; Poon et al., 2007) indicate that many Englerian tribes and subtribes, even in Aurantioideae, are not monophyletic.
|
; Samuel et al., 2001; Morton et al., 2003; Poon et al., 2007).
The rps16 intron is a type II, first used for phylogenetic analysis by Oxelman et al. (1997). The trnL-trnFregion is composed of the trnL intron and the trnL-trnFintergenic spacer (Taberlet et al., 1991). Because noncoding regions have higher rates of evolution than coding regions (for example, see Gielly and Taberlet, 1994, and references therein), fragments such as rps16 intron and trnL-trnFregion have been employed at the infrafamilial level with good resolution (e.g., Andersson and Rova, 1999; Baker et al., 1999; Wallander and Albert, 2000; Asmussen and Chase, 2001).
The objective of the current study is to evaluate the Englerian suprageneric classification of the Rutaceae and to broaden the basis for the requisite new suprageneric classification. A more inclusive phylogeny of the Rutaceae will satisfy not only our scientific curiosity about relationships and character evolution in the family, but will also provide a means to focus the search for additional useful secondary compounds.
MATERIALS AND METHODS
DNA extraction
Total genomic DNA was extracted from 3–5 mg of leaf samples that were fresh, dried, or conserved in cetyltrimethylammonium bromide (CTAB)-NaCl. One sample (Halfordia) was extracted from a herbarium specimen. Nucleon Phytopure kit (Amersham Pharmacia Biotech, Little Chalfort, Buckinghamshire, UK) was used, following the manufacturer’s protocol, except for a longer time of precipitation in –20°C isopropanol (ca. 24 h). Twenty-one DNA samples, previously purified in CsCl2-ethidium bromide (1.55 g/mL, were obtained directly from the DNA Bank at the Royal Botanic Gardens, Kew, UK.
Amplification of rps16
The rps16 intron of 55 sequences from 50 genera (44 of Rutaceae sensu APG [2003]) was amplified using the rpsF> and rpsR2< primers described in Oxelman et al. (1997). The PCR reaction volume (50 µL) contained 30.75 µL water, 3 µL 1% polyvinyl pyrrolidone (PVP), 3 µL 50 mM MgCl2, 5 µL Taqbuffer (10x), 5 µL 10 mM dNTP, 0.25 µL Taqpolymerase, 0.25 µL each primer, and 2 µL DNA sample. Thermal cycling was performed in a PTC-100 Thermal Sequencer (MJ Research, Waltham, Massachusetts, USA), using initial denaturation at 95°C (2 min), followed by 33 cycles at 95°C (30 s), 57°C (1 min), 72°C (2 min); and ended with an elongation at 72°C (7min).
Amplification of trnL-trnF
The trnL-trnFregion of 52 sequences from 47 genera (41 of Rutaceae sensu APG [2003]) was amplified using trn-c> and trn-f< primers described in Taberlet et al. (1991). Some samples were also amplified using the trn-e> and trn-d< internal primers. The PCR reaction volume (50 µL) contained the same proportions of the same ingredients as that used to amplify the rps16 intron. Thermal cycling was performed using initial denaturation at 94°C (7 min), followed by 30 cycles at 94°C (1 min), 56°C (1 min), 72°C (1 min); and ended with an elongation at 72°C (7 min). Slight modifications in the amplification conditions, especially in the annealing temperature, were required for some samples.
Purification, cycle sequencing, and edition of the sequences
The PCR products were purified with GFX PCR columns (Amersham Biociences, Piscataway, New Jersey, USA), following the manufacturer’s recommendations. The sequencing reaction volume was 10 µL, with better yields obtained with 3.25 µL of water, 2 µL of BigDye Terminator Ready Reaction, 0.5 µL (10 mM) of primers, and 4.25 µL of PCR product (60–150 ng of DNA). The reactions were performed in an ABI-3100 automatic sequencer (Applied Biosystems,-Hitachi, Tokyo, Japan), using the same cycle as Scott et al. (2000): 25 cycles at 96°C (10 s), 50°C (15 s), and 60°C (4 min). The obtained sequences were analyzed and edited using the Biological Sequence Aligment Editor software (BioEdit), v.5.0.9 (Ibis Bioscience, Carlsbad, California, USA). Each fragment was carefully examined to verify concordance among the sites. Limits of the trnL-trnFregion and the rps16 intron were determined by comparison with sequences deposited at GenBank. A total of one hundred and six new sequences for both rps16 and trnL-trnF were produced during this study.
Selection of taxa
Representatives of all subfamilies and tribes and almost all subtribes of Rutaceae (sensu Engler, 1931; Swingle and Reece, 1967, for tribes and subtribes of Aurantioideae) were sampled (Appendix 1). These representatives comprised almost one third of the estimated number of genera in the family, and given its hypothetical nonmonophyly, a large sampling of the subfamily Toddalioideae (including all recognized genera from subtribe Toddaliinae, see Appendix 1). Five species of Hortia, the only genus of Toddaliinae from South America, and six genera of the neotropical Galipeinae (a substitution for the illegitimate name Cuspariinae, cf. Kallunki and Pirani, 1998) were included, for the first time, in a phylogenetic study. Only one neotropical species (Zanthoxylum rhoifolium Lam.) of the pantropical genus Zanthoxylum (c. 200 species) was sampled. Most of the sequences of Aurantioideae were obtained directly from GenBank, particularly those produced by Morton et al. (2003). Sequences from Cneorum and Ptaeroxylon (both included recently in Rutaceae, cf. APG, 2003) were also included Carapa, Cedrela, and Guarea (Meliaceae), Simaba (Simaroubaceae sensu stricto, Fernando and Quinn, 1995), and Cupania and Allophylus (Sapindaceae), all from families consistently included in Sapindales (Gadek et al., 1996; APG, 2003), were used as outgroups. Thus, 71 terminals, representing 59 genera of Rutaceae were used in the rps16 intron analyses. Only 70 terminals were used in the analyses of trnL-trnFbecause it was not possible to produce a sequence from Leionema. DNA sequences produced in this study were deposited in GenBank (Appendix 1).
Sequence alignment and phylogenetic analyses
Initial automated alignments of the sequences were made with the CLUSTAL_X software (Thompson et al., 1997) and later largely refined by eye. Indels were coded as missing data. Regions with high homoplasy and long strings of A’s, T’s, or AT’s of different lengths (variable within species and/or caused by slip-strand mispairing, see Kelchner and Wendel, 1996) were excluded.
PAUP* version 4.0b10 (Swofford, 2002) was used for all analyses, with maximum-parsimony criterion and heuristic search. All characters were unordered and equally weighted (Fitch parsimony; Fitch, 1971). Searches were performed with the tree-bisection-reconnection (TBR) branch-swapping algorithm with steepest descent and multrees options, with 100 random-taxon addition replicates, and with 10 trees held in each replicate. All analyses were programmed to retain only 10000 trees. Bootstrap analyses (Felsenstein, 1985) were implemented to verify support for the clades, with 1000 pseudoreplicates (10 trees retained in each pseudoreplicate), simple addition of sequences, and subtree-pruning-regrafting (SPR) branch-swapping algorithm.
The congruence of the results obtained from the fragments was analyzed according to the partition homogeneity test (Farris et al., 1994) implemented in PAUP* to determine the combinability of the two data sets. The test was run with 1000 replicates, heuristic search, simple addition of sequences, and TBR.
RESULTS
rps16
The G+C content in the rps16 intron of Rutaceae species was ca. 35%. The final alignment resulted in a matrix with 993 characters, 338 (33.8%) of them parsimony-informative. More than 10000 equally parsimonious trees were produced, with 1398 steps (CI = 0.56, RI = 0.67, see Table 2). In the strict-consensus tree (Fig. 1), Rutaceae appears as monophyletic (BS = 68%) with the inclusion of Cneorum and Ptaeroxylon. These two genera form a clade with Dictyoloma and Spathelia (BS = 72%), sister to the remaining Rutaceae. The remaining Rutaceae constitute a strong clade (BS = 100%), including representatives of Aurantioideae, Toddalioideae, Rutoideae, and Flindersioideae, of which only Aurantioideae constitute a monophyletic group. Aurantioideae (BS = 99%) appear as sister to a clade formed by Ruta (Rutoideae) and Chloroxylon (Flindersioideae). Casimiroa and Skimmia (Toddalioideae) and Dictamnus (Rutoideae) form a clade (BS = 51%) sister to the bulk of Rutaceae, comprising Flindersia (Flindersioideae) and interdigitated members of Rutoideae and Toddalioideae. None of the Englerian tribes with more one genus (except Diosmeae, represented here by Agathosma and Coleonema) appears as a clade.
|
|
|
; Yoder et al., 2001; Dowton and Austin, 2002). It has been argued that combined data sets provide the best explanatory power in analyses that involve parsimony (see Nixon and Carpenter, 1996; Smith, 2000). Furthermore, visual inspection of consensus trees (and bootstrap values) obtained from different fragments has been used (Mason-Gamer and Kellogg, 1996; Whitten et al., 2000; Muellner et al., 2003). If data sets differ only by weakly supported clades (usually BS < 70% support), a combined analysis is appropriate. In the current study, the incongruity occurred in the positions of the genera in the clade of Aurantioideae, a subfamily that is strongly supported and monophyletic in analyses of both molecular regions. In fact, only one incongruence was strongly supported: Aegle was almost equally supported as sister to Triphasia in the trnL-trnFanalysis (BS = 96%, Fig. 2) and to Afraegle and Balsamocitrus in the rps16 intron analysis (BS = 95%, Fig. 1). Given the overall congruence, the two matrices were combined and the resulting trees examined to determine whether the internal resolution (in terms of bootstrap values, for example) increased or not.
The sequences of Leionema (represented only in the rps16 intron analysis) and Philotheca (represented by a different species in each of the separate analyses, see Appendix 1) were excluded from the combined analysis. The sequences of Flindersia, although derived from specimens identified as different species in the separate analyses, were included, because the genus appears to be monophyletic (Scott et al., 2000). Thus, 69 taxa were included in the combined analysis. The matrix comprised 2127 characters (650 informative, i.e., nearly 30%). More than 10000 equally parsimonious trees were produced, with 2570 steps (CI = 0.60, RI = 0.71, Table 2). The overall topology of the strict-consensus tree is largely congruent with those produced in the separate analyses (Fig. 3). Rutaceae is monophyletic with the inclusion of Cneorum and Ptaeroxylon (BS = 93%). These two genera appear as sister to Dictyoloma and Spathelia, forming a clade (BS = 67%), sister to all other Rutaceae. Within this clade of core Rutaceae (BS = 100%), Aurantioideae appear as monophyletic (BS = 100%) and sister to the clade of Ruta and Chloroxylon (BS = 85%). Casimiroa, Skimmia, and Dictamnus are sister to the remaining Rutaceae. Flindersia (Flindersioideae) and interdigitated members of Rutoideae and Toddalioideae constitute a strongly supported clade (BS = 94%). As in the separate analyses, none of the Englerian tribes with more one genus (except Diosmeae) appears as a clade.
|
The family Rutaceae has been the focus of several recent molecular phylogenetic studies to reevaluate the circumscription of the Englerian subfamilies (Chase et al., 1999; Scott et al., 2000; Poon et al., 2007) or to examine specific subclades, such as the Aurantioideae (Samuel et al., 2001; Morton et al., 2003). The previously published study with the largest taxon sampling of the family was that of Chase et al. (1999); they used rbcL and atpB sequences and sampled 35 genera of Rutaceae (including Cneorum, Harrisonia, and Ptaeroxylon), but only three genera from Toddalioideae (Casimiroa, Phellodendron, and Skimmia) and three neotropical genera (Dictyoloma, Pilocarpus, and Spathelia). Scott et al. (2000) used a less comprehensive data set than did Chase et al. (1999) to infer the monophyly of Englerian subfamilies; they used only the trnL-trnFregion and sampled 23 genera, mainly of Aurantioideae, from Australia and adjacent Asia. Most recently, Poon et al. (2007) used trnL-trnFand ITS sequences and focused mainly on the genera of "proto Rutaceae" (Phellodendron, Tetradium, and Zanthoxylum). Despite the fact that these three studies were based on different DNA sequences and different sets of genera, they consistently showed that Toddalioideae and Rutoideae were not monophyletic groups, that Ruta was closer to Aurantioideae than to other Rutoideae, and that in the case of Chase et al. (1999) and Scott et al. (2000), Dictyoloma and Spathelia formed a clade separate from the remaining Rutaceae. The broader sampling of genera of Rutaceae included in this study confirmed some relationships hypothesized in earlier studies, such as the position of Ruta close to Aurantioideae and the placement of Spathelia, Dictyoloma, Cneorum, and Ptaeroxylon in a clade sister to all other Rutaceae, and suggested some new relationships. In addition, because this study included a larger sampling of the subfamily Toddalioideae and of neotropical groups (e.g., the large subtribe Galipeinae) than did previous studies, we were able to correlate phylogeny with geographic distributions on a worldwide scale.
The utility of rps16 and trnL-trnFregions in the inference of infrafamilial relationships in Rutaceae has been stressed by Morton et al. (2003). In the current study, the number of informative characters is larger in rps16 (338, 33.8% of the total characters) than in trnL-trnF (323, 27.9% of total). Because the rps16 intron is more variable and more difficult to align than the trnL-trnFregion, more regions were excluded from the definitive alignment. That the RI in the rps16 analysis (0.67) was lower than that in the trnL-trnFanalysis (0.77) indicates that rps16 furnished more homoplasious characters than the trnL-trnFregion. The trees produced by both analyses have similar numbers of clades with bootstrap values 
50% (37 in rps16 and 38 in trnL-trnF), but that from the trnL-trnFanalysis had more clades with values 85%. Incongruence between the two fragments was caused mainly by the short branch length of some clades (Fig. 4), such as the Diosmeae, Pilocarpus, and some genera of Aurantioideae (e.g., Swinglea, Triphasia, Aegle, Citrus, and Clausena). Despite the position of the clade ((Swinglea) Aegle, Triphasia) in the trnL-trnFanalysis (Fig. 2), all other clades with some incongruence between the two analyses had low bootstrap values (and consequently, low support). Variation in the rps16 intron and trnL-trnFregions is insufficient to solve infrageneric problems in groups with more recent diversification (less than 5 mya, see Small et al.,1998; Wallander and Albert, 2000; Richardson et al., 2001), which may be the case in Aurantioideae.
|
85% (Table 2) and better support of internal clades and is, therefore, considered superior to the separate analyses. Given the large number of shortest trees, the following discussion is largely based on the consensus tree from the combined analysis (Fig. 3). Other types of data possibly indicative of relationships in the Rutaceae are also discussed and shown in Fig. 5. Some clades in Figs. 3 and 5 are named or lettered to ease the discussion.
|
) and of secondary metabolites (Adams et al., 1973; Waterman, 1983; Vieira et al., 1988). If these two genera are included in Rutaceae, however, Cneorum (Cneoraceae) and Ptaeroxylon (Ptaeroxylaceae) must be included also to achieve monophyly of the family, as suggested by Chase et al. (1999) and by APG (2003). The genera of the remaining four subfamilies (Aurantioideae, Flindersioideae, Rutoideae, and Toddalioideae) are also included in Rutaceae with strong support.
Spathelia-Ptaeroxylon clade
Spathelia and Dictyoloma (both neotropical) form a strongly supported clade in both separate and combined analyses. Chemical data support the relationship between these two genera—alkaloids, limonoids, and chromones from Dictyoloma vandellianum are very similar to those extracted from Spathelia sorbifolia (Vieira et al., 1990). The recognition of these as two distinct monogeneric subfamilies does not seem warranted, given that they appear to be a part of a larger group that includes Cneorum (three species from the Mediterranean region, the Canary Islands, and Cuba) and Ptaeroxylon (one species, P. obliquum, from Africa), both either recognized either as distinct families (Cneoraceae and Ptaeroxylaceae) or as part of the Rutaceae sensu APG (2003). In studies based on rbcL and atpB sequences (Chase et al., 1999), these four genera also appeared as a group sister to the remaining Rutaceae and Harrisonia (with three species in tropical Africa, southwest Asia, and tropical Australia) was sister to a clade formed by Cneorum and Ptaeroxylon. Harrisonia was included (albeit weakly) with Cneorum and Ptaeroxylon in the study by Gadek et al. (1996). Species of Harrisonia produce limonoids and chromones of the ptaeroxylin group; this combination of limonoids and ptaeroxylin chromones occur together only in Cneorumand Spathelia (Taylor, 1983; Da Silva and Gottlieb, 1987; Gadek et al., 1996) and Dictyoloma (Vieira et al., 1990). In contrast, Ptaeroxylon itself produces these chromones and typical rutacean coumarins, but no limonoids (Waterman, 1983; Gadek et al., 1996). Morphological synapomorphies of genera in the Spathelia-Ptaeroxylon clade are not known, but compound leaves (except in Cneorum) are common, and unisexual flowers and basally appendaged staminal filaments are found in Dictyoloma and Spathelia.
Cedrelopsis and Bottegoa, not sampled here, are the two other genera of Ptaeroxylaceae (as delimited by Verdcourt and Davies, 1996). Cedrelopsis (with seven species restricted to Madagascar) does produce ptaeroxylin chromones (Taylor, 1983) but not limonoids or protolimonoids. Its seeds, like those of Ptaeroxylon, are winged. Bottegoa (one species from Ethiopia and Somalia) was originally described in Sapindaceae but transferred to Ptaeroxylaceae after detailed morphological studies (van der Ham et al., 1995). Although Bottegoa has unisexual flowers and compound leaves, its staminal filaments are not appendaged. Chemical characteristics of Bottegoa are still unknown. These two genera clearly seem to belong to the Spathelia-Ptaeroxylon clade.
Inclusion in a more broadly circumscribed Rutaceae of genera previously placed in Ptaeroxylaceae and Cneoraceae requires discussion of the presence/absence of secretory cavities containing aromatic ethereal oils in the family. The occurrence of such cavities in some organs, commonly visible as pellucid dots in leaves, is considered synapomorphic to the Rutaceae (Judd et al., 2008) and more or less "defines" the family. Such cavities are present in members of the core Rutaceae clade and in Spathelia and Dictyoloma, i.e., in Rutaceae sensu Engler (1931). Pellucid dots may occur in some Harrisonia (Forman, 1958; Nooteboom, 1962, 1966), but there is no evidence that they correspond to cavities. In contrast, cavities do not occur in Cneorum, Ptaeroxylon, Bottegoa, or Cedrelopsis (Metcalfe and Chalk, 1950; van der Ham et al., 1995). Some Cedrelopsis bear pellucid dots on the leaves (Leroy et al., 1990), but the structure of these dots has not been studied. In Rutaceae, solitary oil cells may or may not occur (Metcalfe and Chalk, 1950).
The hypothesis that secretory cavities were present in the ancestors of the (Cneorum, Ptaeroxylon) clade but subsequently lost and the hypothesis that such cavities arose independently in the (Spathelia, Dictyoloma) clade and in the core Rutaceae clade are equally parsimonious. Differences in the formation of these cavities (lysigenous, schizogenous, or schizolysigenous) could be evidence that such structures appeared at different times in the history of the family, favoring the second hypothesis. The developmental anatomy of these cavities must be studied to evaluate their usefulness as indicators of the phylogeny of the family.
Core Rutaceae clade
The second internal branch, the "core Rutaceae" is formed by representatives of the four other subfamilies. In this clade, the presence of secretory cavities is constant, and chromones are restricted to Flindersia, Skimmia, and Maclurodendron (Zakaria, 2001). Two large clades are internal to this core (Figs. 3, 5): clade A, formed by the Englerian Aurantioideae (Citrus and allies) and the Ruta-Chloroxylon clade (Ruta, Rutoideae, and Chloroxylon, Flindersioideae), and clade B, formed by members of Rutoideae, Toddalioideae, and Flindersia (Flindersioideae). In clade B, the Casimiroa-Dictamnus clade is sister to clade C (comprising genera of tropical America and South Africa) and clade D (comprising genera from the Old World and Oceania).
Aurantioideae
Citrus and allies
Genera from Englerian Aurantioideae appeared in all analyses as a clade with strong support. This result is supported by other studies based on molecular data (Chase et al., 1999; Scott et al., 2000; Samuel et al., 2001; Morton et al., 2003), chromosome number (Stace et al., 1993), alkaloids (Waterman, 1975), and other secondary metabolites (Waterman, 1975; Grieve and Scora, 1980; Da Silva et al., 1988).
The delimitation of the tribes and subtribes in the Aurantioideae is more complicated. The number of tribes recognized varies from one to eight (Engler, 1931; Tanaka, 1936; Swingle and Reece, 1967). Using data from a study of sequences from atpB, rbcL, and rps16 intron and the distribution of carbazolic alkaloids, Samuel et al. (2001) criticized the tribal delimitation of Swingle and Reece (1967) and suggested that Murraya and Merrillia (sister groups here) be removed from Clauseneae (which have carbazolic alkaloids) and included in "Citreae sensu lato" (which do not have these alkaloids). The move of these two genera to the Citreae is also supported by chromosomal data (Guerra et al., 2000).
Our results also indicate that the delimitation by Swingle and Reece (1967) of tribes in Aurantioideae may be not adequate. Instead, our data show Glycosmis and Micromelum (both Clauseneae) to be successive sisters to the remainder of Aurantioideae and Clausena to fit among the bulk of the Citreae. The interdigitation of genera traditionally assigned to Clauseneae and Citreae indicate that the two tribes could be not monophyletic, as shown also by Morton et al. (2003).
Other than the successive sister position of Glycosmis and Micromelum to the remainder of Aurantioideae discussed, the positions of genera in Aurantioideae are not well resolved. Branch lengths within the clade comprising all Aurantioideae, except Glycosmis and Micromelum, are relatively shorter than those in most other parts of the tree (see Fig. 4). Analyses of DNA fragments with faster evolutionary rates may clarify the relationships between the genera in Aurantioideae.
Flindersioideae and the position of Ruta
Results confirmed the placement of Flindersia and Chloroxylon (Flindersioideae) in Rutaceae, as suggested by studies of secondary metabolites (Taylor, 1982; Da Silva et al., 1988), pollen morphology (Erdtman, 1966), and molecular data (Chase et al., 1999; Scott et al., 2000), and refuted the suggestion made by some authors on the basis of floral and wood anatomy (cf. Lal and Narayana, 1994) that Flindersioideae was intermediate between Meliaceae and Rutaceae.
Although nested within Rutaceae, Flindersia and Chloroxylon do not form a clade in either the separate or combined analyses. Chloroxylon (three species restricted to southern India and Madagascar) and Flindersia (17 species, in Moluccas, New Guinea, eastern Australia, and New Caledonia) are similar in several morphological respects, but different in others, as summarized by Chase et al. (1999). The two genera also have different base chromosome numbers (Chloroxylon x = 10, Flindersia x = 18; Stace et al., 1993). Thus, it appears that the similarities of the capsular fruits and winged seeds of the two genera used by Engler (1931) to define Flindersioideae are products of convergence and not adequate to define them as a subfamily.
As also shown by Chase et al. (1999), Flindersia in the current study is sister to Lunasia (Zanthoxyleae, Rutoideae, one species in Australia, Philippines, and the Malayan region) (Figs. 3 and 5, Flindersia-Lunasia clade). The two genera differ, however, in several morphological characteristics; Lunasia has simple leaves, trimerous flowers disposed in congested glomerules, three carpels per flower, follicular fruits with elastic endocarp, and unwinged seeds. Morphological synapomorphies of this clade are not known.
The placement of Ruta (ca. seven species in the Mediterranean region) closer to Aurantioideae than to the other Rutoideae was suggested in previous macromolecular studies (Fernando et al., 1995; Gadek et al., 1996), and its relationship with Chloroxylon in the Ruta-Chloroxylon clade (Figs. 3 and 5), which is sister to Aurantioideae, was shown also by Chase et al. (1999). Ruta is unlike Chloroxylon in its herbaceous or suffrutescent (vs. arborescent) habit, unwinged (vs. winged) seeds, and presence (vs. absence) of endosperm. They are similar, however, in their diplostemonous flowers, unguiculate petals with concavities in which the smaller antepetalous stamens are encased, urceolate disc, and more than two ovules per locule. More significantly, the base chromosome number in both genera is x = 10 (Stace et al., 1993), a number so far encountered elsewhere in Rutaceae only in Boenninghausenia, in the same subtribe Rutinae as Ruta. Thus, the base number x = 10 may be a true synapomorphy of this clade. This pairing requires closer examination, however, because the sequences of rps16 and trnL-trnFwere obtained from the same sample of Chloroxylon (Chase 1291, K) as that used by Morton et al. (2003) and Chase et al. (1999). The position of Ruta close to Aurantioideae, however, is confirmed because a different sample of this genus (Groppo 1151, SPF) was used in the current study. The close relationship of Ruta with Aurantioideae was obtained also by Poon et al. (2007) in a study of fewer genera of Aurantioideae.
The placement of Chloroxylon and Ruta close to Aurantioideae in clade A is not supported by known morphological data. Morawetz (1986), however, suggested that the base number x = 10 in Chloroxylon is a dysploid alteration from x = 9, a number known in Rutaceae only in Aurantioideae and in Haplophyllum (considered by some authors as a subgenus of Ruta) and Thamnosma (both in Rutinae). Further phylogenetic studies of these genera are necessary to clarify the polarity of the character states x = 10 vs. x = 9.
Rutoideae and Toddalioideae
Flindersia (Flindersioideae) and interdigitated genera of Rutoideae (except Ruta) and Toddalioideae formed a clade (B) in all analyses (with BS = 80% in the combined analysis, Fig. 3). As in previous studies (Chase et al., 1999; Scott et al., 2000), the current study shows that the Englerian subfamilial circumscriptions of Rutoideae and Toddalioideae, based mainly on fruit dehiscence and degree of carpel connation, is not adequate. Many genera with dehiscent fruits and (sub)apocarpous ovaries appear as sister to others with indehiscent fruits and syncarpous ovaries in very strongly supported clades, e.g., Toddalia-Zanthoxylum and Adiscanthus-Hortia (Fig. 3). Thus, connate or separate carpels and dehiscent or indehiscent fruits appear to have arisen several times in the evolutionary history of the group (see Fig. 6). Data from fruit development (Hartl, 1957; Zavaleta-Mancera and Engleman, 1991) and formation of the oil glands (Scholz, 1964; comments in Hartley, 2001) also argue against this delimitation.
|
to define some tribes (Fig. 6B), also appear to have evolved several times in different lineages in Rutaceae. Winged seeds are known, for example, in the two genera of the Flindersioideae, in Ptaeroxylon, and in Tetractomia (segregated from Melicope by Hartley, 1979). Apart from the Casimiroa-Dictamnus clade, of which Skimmia and Dictamnus occur in temperate and tropical regions of the northern hemisphere in the Old World and Casimiroa in Mexico and Costa Rica, the clades appear to correlate with the geographic distributions of the groups (Fig. 5). The sister of the Casimiroa-Dictamnus clade is divided into two large clades—clade D of genera primarily from the Old World and Oceania and clade C of genera primarily from tropical America. The exception in clade D is Zanthoxylum, which occurs also in tropical America, and that in clade C is the Diosmeae, which occur in South Africa, Tanzania, and Kenya. The African Diosmeae and the neotropical Pilocarpus appear with the Old World genera in the trnL-trnFanalysis (Fig. 2), but with the American genera in the rps16 and combined analyses (Figs. 1 and 3). Their differing positions (and the low bootstrap values <50% in the combined analysis) may be explained by the short branch lengths at the base of the clades comprising the genera from tropical America and the Diosmeae (see Fig. 4).
Casimiroa-Dictamnus clade
The grouping of Dictamnus (Ruteae, Rutoideae) with Casimiroa and Skimmia (Toddalioideae) in a clade is also supported by the results of Chase et al. (1999). Dictamnus (one or two species, Europe to northern China) and Skimmia (four species, east of the Himalayas to southern Vietnam and the Philippines) occur in temperate and tropical areas of the Old World. These three genera differ in several distinct morphological characteristics. Dictamnus differs from Skimmia by the apocarpous (vs. syncarpous) ovaries, capsular (vs. baccate) fruits, an herbaceous or suffrutescent (vs. shrubby or arborescent) habit, pinnate (vs. simple) leaves, zygomorphic (vs. actinomorphic) corolla, and more than two (vs. one or two) ovules per locule. Casimiroa is more similar to Skimmia, sharing a fully syncarpous ovary, indehiscent fruit, and one or two ovules per locule, but its leaves are generally palmate, instead of simple. The formation of endocarp in the two genera is different, however, and not homologous (Zavaleta-Mancera and Engleman, 1991). Chemical studies have not suggested a relationship between these genera (see articles in Waterman and Grundon, 1983, and Da Silva et al., 1988).
In the present study, the Ruteae (Rutoideae) are represented by Dictamnus (the only genus in the Dictamninae) and Ruta (one of six genera in the Rutinae). The similarities between these two subtribes of Ruteae are few other than those used by Engler (1931) to define the tribe: herbaceous or shrubby habit and temperate distribution (see Table 1). Dictamnus and Ruta differ in floral anatomy (Moore, 1936), structure of the seed (Corner 1976), distribution of secondary metabolites (Da Silva et al., 1988), and chromosome number (Stace et al., 1993). All these data corroborate that the Ruteae are not monophyletic, as observed in this study. Despite its apparent chemical cohesiveness (Waterman, 1983), the genera of the Rutineae are highly diverse in morphology, geographic distribution, and chromosome number (Stace et al., 1993).
Clade C: Balfourodendron-Metrodorea clade
Esenbeckia, Metrodorea, Balfourodendron, and Helietta appear in a clade strongly supported in all analyses. Although Engler positioned the first two in Rutoideae because of their dehiscent fruits (capsules) and the last two in Toddalioideae because of their indehiscent fruits, the ovary in all four genera is syncarpous (though in some species of Esenbeckia, carpels are united only at the base). The flowers of these four genera are small, actinomorphic, isostemonous, and grouped in terminal inflorescences (panicles or thyrses). The association of these genera is reinforced by wood structure, i.e., heterogeneous rays with 1–5 cells and crystals in the axial parenchyma (Record and Hess, 1940). Engler (1931) positioned Balfourodendron, Helietta, and Ptelea (1–3 species in the United States and Mexico) in subtribe Pteleinae (Toddalieae, Toddalioideae) because of the winged fruits and the trifoliolate leaves, but the position of Ptelea obtained in present analyses indicates that the subtribe is not a monophyletic group. The wing of the fruit in Ptelea is peripheral (or sometimes vestigial), but that of Balfourodenron and Helietta is terminal. Ptelea and the other two genera also have significant differences in the disc and ovary morphology (see Pirani, 1998).
Clade C: Pilocarpus
This genus, together with Esenbeckia and Metrodorea (and Raulinoa, which was not analyzed), constitute the subtribe Pilocarpineae, one of the two subtribes of Galipeae (Rutoideae). Pilocarpus shares with Esenbeckia and Metrodorea, in addition to the characteristics cited in Table 1, actinomorphic, isostemonous flowers, and free stamens. It differs from these two genera by racemose (vs. paniculate or thyrsoid) inflorescences, disc totally united with (vs. only apressed to) the ovary, and follicular (vs. capsular) fruit. Pilocarpus differs further by the possession of imidazole alkaloids, otherwise known in Rutaceae only in Casimiroa (Price, 1963). Because of these differences, Kaastra (1982, p. 21) suggested the exclusion of Esenbeckia, Metrodorea, and Raulinoa from Pilocarpineae, leaving the tribe monogeneric. The close relationship of Esenbeckia and Metrodorea with Helietta and Balfourodendron obtained here supports Kaastra’s view. The positions of Pilocarpus and Ptelea, however, must be further investigated.
Clade C: Adiscanthus-Hortia clade
Hortia (10 species, almost all in Amazonia) and Adiscanthus (one species, also in Amazonia) constitute a very strong clade in all analyses. Because of its baccate fruit and syncarpous ovary, Engler (1931) included Hortia as the single neotropical genus in Toddalioideae subtribe Toddaliinae. Because of its follicular fruit and apocarpus ovary, he included Adiscanthus in Rutoideae, subtribe Galipeinae (as Cuspariinae). Because of chemical characteristics, Da Silva et al. (1988) positioned Hortia with the Galipeae in the "Cusparia informal tribe." Many characteristics of Adiscanthus, such as actinomorphic flowers, free petals, anthers without appendages, and cotyledons not plicate, are less common characters in Galipeinae. Despite the differences in fruit and in connation of carpels, Hortia and Adiscanthus share simple leaves; free, valvate, and adaxially pilose petals; and sparsely branched habit. The baccate fruits of Hortia are apparently derived.
Clade C: Angostura-Sigmatanthus clade
The other genera from subtribe Galipeinae (Galipeae, as Cuspariineae in Engler, 1931) included in the current study constitute a very strong clade in all analyses (see Fig. 5, clade C). This subtribe is the most diversified group of Rutaceae in the neotropics, with 28 genera and c. 130 species. Morphological characteristics include mostly zygomorphic flowers, a more or less tubular corolla, filaments united to the corolla tube, two rather than five fertile stamens, basally appendaged anthers, and plicate cotyledons. Base chromosome number is very diverse (Stace et al., 1993), and there are exceptions to all the morphological characteristics cited.
The genera of Galipeae (except Pilocarpus) and Balfourodendron, Helietta, and Hortia (all Toddalioideae) constitute a clade in several shortest trees in the combined analysis (see for example, Fig. 4) but not in the consensus. Studies of pollen (Barth, 1982; Morton and Kallunki, 1993) show a great diversification in which many genera are characterized by a particular pollen type. Chromosomal differentiation in Galipeinae and allied genera (e.g., Hortia) is substantial; the base number varies from 12 in Hortia (Forni-Martins and Martins, 2000) to 15, 18, or even, in Erythrochiton, 58 (Stace et al., 1993). These differences in morphology, palynology, and base chromosome number could be a result of a longer period of evolution or a faster differentiation in these groups, consonant with the longer branch lengths.
Choisya (four or five Mexican species), here in clade C (Fig. 5), and Medicosma (25 species in Australia, New Guinea, and New Caledonia) in clade D (Fig. 5), were included by Engler (1931) in subtribe Choisyinae (tribe Zanthoxyleae, Rutoideae). The characteristics used by Engler to define this subtribe (actinomorphic, relatively large, white flowers and deciduous sepals) appear to be unsatisfactory. The positions of these two genera in the present analyses are congruent with their geographic distributions—Choisya in the clade of genera from tropical America and Medicosma in that from the Old World and Oceania.
Clade C: Agathosma-Coleonema clade (=Diosmeae of recent classifications)
The association of Adenandra and Coleonema (Diosmeae) in a very strongly supported clade in all analyses is corroborated by diverse sources of data. Diosmeae are a morphologically well-characterized group centered in South Africa. They are in general perennial shrubs or subshrubs, always with simple leaves, actinomorphic flowers, seeds without endosperm, an embryo with fleshy cotyledons, and glands at the apex of the anthers (cf. Williams, 1984). This group is characterized by an absence of quinolones and alkaloids and a reduction of coumarins (Waterman, 1983; Da Silva et al., 1988). In the subtribes Diosminae (nine genera) and Empleurinae, plants are relatively small, secondary growth is reduced, and the leaves are coriaceous—characteristics found also in unrelated taxa occurring in the sclerophyllous vegetation of the Cape flora (Goldblatt et al., 1985). The morphological uniformity of the tribe is broken by Calodendrum, the one genus of the third subtribe Calodendrinae (two species in South Africa to Tanzania and Kenya), which is characterized by arborescent habit and a fully syncarpous gynoecium. The base chromosome number of Calodendrum (x = 17) contrasts with those of other genera of Diosmeae (x = 13, Stace et al., 1993). In the analysis of Scott et al. (2000), based on the trnL-trnFintergenic spacer, Calodendrum appears as a sister group to Coleonema only in the neighbor-joining tree. In Chase et al. (1999), Calodendrum was paired with Adenandra (Diosminae) in the atpB analysis, but was sister to a clade of Adenandra and Phellodendron (Toddalioideae) in the rbcL and combined analyses. Recent molecular studies have corroborated the monophyly of Diosmeae, but a closer examination of some generic circumscriptions within the tribe is needed (Trinder-Smith et al., 2007).
Clade D: Toddalia-Zanthoxylum clade
The position of Toddalia (Toddalioideae, a single species distributed from East Africa to the Philippines) and Zanthoxylum (Zanthoxyleae, Rutoideae) in a clade with strong support (Fig. 3) is corroborated by many morphological characteristics such as leaves and bark with spines and small, unisexual flowers with a reduced annular disc and grouped generally in panicles. Chemical characteristics (Fish and Waterman, 1973; Waterman, 1975, 1983, 1990; Da Silva et al., 1988) also support the proximity of Toddalia and Zanthoxylum. Waterman (1975) considered the presence of 1-BTIQ alkaloids as primitive in Rutaceae. The presence of such alkaloids in Toddalia, Zanthoxylum, Fagaropsis, Phellodendron, and Tetradium (Ng et al., 1987) led Waterman to create the term "proto-Rutaceae" for this group of genera. Although a "primitive" position of this group within Rutaceae has been suggested by Waterman and Grundon (1983) and by Da Silva et al. (1988), it is not supported by the genera included in the current study. Phellodendron (Toddalioideae, two species in Japan, China, and eastern Russia; Ma et al., 2006) is close to Zanthoxylum in the trnL-trnFstudy of Scott et al. (2000) only in the neighbor-joining tree and in the study of Chase et al. (1999) only in the atpB, not the rbcL, analysis. In a more recent study of ITS and trnL-trnFregions (Poon et al., 2007), Phellodendron, Tetradium, Toddalia, and Zanthoxylum (Fagaropsis was not sampled) indeed formed a clade, and, thus, the proximity of these genera suggested by the presence of 1-BTIQ alkaloids was supported.
Clade D: Vepris
The two African species of Vepris (Toddaliinae, Toddalioideae, more than 60 species in Africa and tropical Asia) included in the current study formed a strongly supported clade in all the analyses. One of these, Vepris simplicifolia, was recently transferred to that genus from Teclea, which is one of eight African genera (from three of Engler’s subtribes) synonymized with Vepris by Mziray (1992; see comments in Appendix 1). The well supported Vepris clade supports Mziray’s transfer.
Clade E
The genera concentrated in Oceania and the Malayan region are grouped in clade E with BS = 85% in the combined analysis (see Fig. 3). It contains two very strongly supported clades (F and G). Clade F, the Melicope-Acronychia clade, comprises Acronychia (Toddalioideae) and Medicosma, Melicope, and Sarcomelicope (Zanthoxyleae, Rutoideae) and is sister to Boronia.
These four genera were included by Hartley (2001) in a group of 32 nonboronioid and nonaurantioid Rutaceae from the Australasian-Malayan region, 22 of these characterized by sclerotestal seeds. Additionally, many of the 22 genera, including Halfordia, Zanthoxylum, and those in the Melicope-Acronychia clade, have seeds with a circular chalazal aperture (Hartley, 2003, p. 9). A circular aperture and sclerotestal seeds cooccur also in Zieria chevalieri Virot., a species, according Hartley, that clearly belongs to Boronieae. Hartley (2001) stated, however, that the nonboronioid, nonaurantioid group did not appear to be divisible into tribes based on morphological characters and that, if treated as a single group, they would belong to Zanthoxyleae.
The position of Zanthoxylum in the present analysis, however, argues against Hartley’s view of Zanthoxyleae, which to be monophyletic must consist of all genera in clade D, i.e., Toddalia, Flindersia, Lunasia, Vepris, and the Boronieae (see Figs. 3 and 5). The presence of a sclerotesta and a circular chalazar aperture in the seeds of genera other than those in Melicope-Acronychia clade must be investigated to establish homologies and assess their distributions among the genera of clade D and their relatives.
The position of Boronia (Boronieae, Rutoideae, 140 species mostly in Australia) closer to the members of the Melicope-Acronychia clade than to other members of the Boronieae was noted by Hartley (1985), according to whom, Boronia differs consistently from Medicosma (25 species in Australia, New Caledonia, and New Guinea) only in cotyledon shape (linear and about the same width as the hypocotyl in Boronia, elliptic and wider than the hypocotyl in Medicosma). Boronia, Medicosma, and Melicope (as well as Brombya and Euodia, both also in Zanthoxyleae, Rutoideae) possess modifications in the tracheal elements (brachy- and sclerotrachaeids) at the ends of the foliar veins (Hartley, 1985, p. 29). The position of Melicope and Sarcomelicope, far from Zanthoxylum (and its sister Toddalia), renders the subtribe Euodiinae sensu Engler (1931) polyphyletic. The presence of seeds pendulous by the funicle after dehiscence of the fruit in Melicope and Zanthoxylum must be interpreted as a convergence.
Clade G, the other internal branch of clade E, is formed by other Boronieae (Australia, New Caledonia, and New Zealand) and Halfordia (Toddalioideae) and is strongly supported (BS = 95%) in the combined analysis. Other Boronieae that group with these genera are Leionema in the rps16 analysis (Fig. 1) and Philotheca in both rps16and trnL-trnFanalysis (Figs. 1 and 2), both of which were excluded from the combined analysis for reasons mentioned previously. Clade "Boronieae without Boronia" was obtained by Scott et al. (2000). In Chase et al. (1999), the Boronieae emerged as a clade, but Boronia itself was not included in that study. In the current study, subtribe Eriostemoninae (represented by Leionema and Philotheca) appears unresolved in the rps16 analysis, while subtribe Nematolepidinae (represented by Nematolepis and Chorilaena) is, according our results, one of the Englerian subtribes with more than one genus that appeared as monophyletic, supported by overall morphology and basic chromosome number x = 14 (Stace et al., 1993). Together with four other boronioid genera, the last two are part of the "Phebalium group" of Wilson (1998) that includes also Leionema (Eriostemoninae).
Halfordia (one species, New Guinea, eastern Australia, New Caledonia, and other Pacific islands) is morphologically very distinct from the Boronieae; Halfordia is a tree with large leaves from tropical forests, and the Boronieae are subshrubs or shrubs with small leaves from open, dry areas. In addition, Halfordia differs from them by its syncarpous (vs. more or less apocarpous) ovary, drupaceous (vs. capsular) fruit, elliptic-oblong (vs. terete, linear) cotyledons, and seeds without (vs. with copious) endosperm. Yet, in all three analyses here, it was grouped with "Boronieae without Boronia," in disagreement with Hartley (2001, 2003), who positioned Halfordia in his expanded Zanthoxyleae. Morphological characters uniting Halfordia and its boronioid allies are not known so far.
Engler (1931) defined Boronieae by the herbaceous or shrubby habit, by the small, narrow leaves, and, more importantly, by the terete, linear cotyledons immersed in copious endosperm (see Table 1). The habit and the shape of the leaves are analogous to those in Diosmeae and reflect the adaptation of these and the Boronieae to dry (but not desert-like) areas in Africa and Australia, respectively. The results obtained here suggest that seeds with a terete, linear embryo and copious endosperm and the herbaceous-suffrutescent habit arose at least twice in the evolutionary history of Boronieae sensu Engler (1931). Characters of endosperm and embryo are not known, however, for many of the more than 55 genera in the Australasian and Malayan region, and thus, conclusions about the polarity of the states of endosperm and embryo characters are premature. Nevertheless, the consistent position of Boronia (type genus of Boronieae) far from other Boronieae (obtained here and supported by other sources of data) shows a need for formal recognition of the clade "Boronieae without Boronia" and the allied genus Halfordia.
Conclusions
The results obtained here, which agree with those from other sources, can be summarized as follows: (1) Cneorum and Ptaeroxylon (and consequently Cneoraceae and the other Ptaeroxylaceae, i.e., Bottegoa and Cedrelopsis) and Harrisonia (Simaroubaceae) must be included in Rutaceae (as in APG, 2003), where they form, with Spathelia and Dictyoloma, a clade sister to the remaining Rutaceae. (2) None of the Englerian subfamilies with more than one genus (except Aurantioideae) is monophyletic. (3) None of the Englerian tribes with more than one genus (except Diosmeae), and almost none of the subtribes analyzed so far are monophyletic. (4) Traditional characteristics used to delimit the subfamilies, such as degree of connation of the carpels, fruit dehiscence, and presence/absence of endosperm are not useful at subfamilial and tribal levels. (5) The clades are better correlated with geographic distributions of the genera than are the characteristics cited in (4).
The current study and others have accumulated enough evidence to abandon the Englerian subfamilies of Rutaceae. The Spathelia-Ptaeroxylon clade could be recognized as a newly circumscribed Spathelioideae (based on the oldest subfamilial name), including Spathelia (Spathelioideae), Dictyoloma (Dictyolomatoideae), Cneorum (Cneoraceae), all Ptaeroxylaceae, and Harrisonia (Simaroubaceae); and the core Rutaceae clade could be recognized as a newly circumscribed subfamily Rutoideae, including the traditional Rutoideae, Aurantioideae, Flindersioideae, and Toddalioideae. Alternatively, the clade containing Ruta (the type genus of Rutoideae) and Aurantioideae could be recognized as Rutoideae (based on the oldest name), the Aurantioideae recognized as a tribe, and the clade containing the remaining Rutoideae, the Toddalioideae, and Flindersia (Flindersioideae) as a subfamily.
The current study provides a framework for revisions of the larger clades within Rutaceae and a reinterpretation of the biogeography of the family. The hypothesis of two major clades in the family, one comprising neotropical taxa and the other taxa from Oceania (including the Malayan region) is novel. The relationships of Boronia with genera from the Melicope-Acronychia clade and of Halfordia with the remaining Boronieae, the grouping of genera of both Rutoideae and Toddalioideae in the Balfourodendron-Metrodorea clade, and the paraphyly of Galipeinae are also novel hypotheses.
Molecular studies, as well as morphological, anatomical, chemotaxonomical, and chromosomal studies, of a broader sample of genera are needed for a better understanding of phylogeny of the complex Rutaceae, especially at tribal and subtribal levels. The authors’ ongoing studies of neotropical groups of Rutaceae, with emphasis on Galipeinae (Rutoideae), are expected to contribute new insights into the phylogenetic history of the family.
Appendix 1. Voucher information and GenBank accession numbers for taxa used in this study. Infrafamilial classification of Rutaceae (excluding Rhabdodendroideae =Rhabdodendraceae) follows Engler (1931), and that of Aurantioideae follows Swingle and Reece (1967). Approximate number of genera in each group is noted in parentheses. Genera are those included in this study, each followed by citations of voucher specimens or references and GenBank accession numbers. Subtribes from which genera were not included in this study are included on the list. A dash indicates the molecular region was not sampled. Herbarium acronyms follow Holmgren et al. (1990).
|
FOOTNOTES
1 The authors thank T. G. Hartley for sending samples of Australian plants (Acronychia, Flindersia, Halfordia, Melicope, Sarcomelicope); J. Mafezolli for the sample of Sigmatanthus; E. Pansarin for sequence-editing programs; A. C. Marcato for the use of his Macintosh and help with some analyses; E. Kapinos of the Jodrell Laboratory, Kew, for technical support during a visit by M.G.; M. C. Oliveira and M.-A. Van Sluys for the use of the laboratory at Universidade de São Paulo; the Brazilian institutions and people who assisted in the field work, especially R. Sarquis and S. Sarquis (IEPA, Macapá), R. C. Viana (EMBRAPA, Belém), C. A. Cid-Ferreira and R. Gribel (INPA, Manaus), A. A. Barbosa (UFU, Uberlândia), M. C. T. B. Messias (UFOP, Ouro Preto), A. M. Carvalho (CEPEC, Ilhéus, in memoriam), W. Sciarra (Fazenda Fischer, Onda Verde), H. Boudet-Fernandes (MBML, Santa Teresa), and R. de Jesus (Reserva Ecológica da Vale do Rio Doce, Linhares); R. Olmstead and an anonymous reviewer for their constructive criticisms of the text; and J. Jernstedt, A. McPherson, and S. Balcomb for editorial assistance. This work was supported by grants to the first author from CAPES and FAPESP (00/07401-0, 05/50758-7) and the Margareth Mee Foundation. 
5 Author for correspondence (e-mail: groppo{at}ffclrp.usp.br)
LITERATURE CITED
Adams, C. D., D. R. Taylor, AND J. M. Warner. 1973. N-methylflindersine from Spathelia sorbifolia. Phytochemistry 12: 1359–1360.[CrossRef][Web of Science]
Aliotta, G., G. Cafeiro, V. De Feo, A. D. Palumbo, AND S. Strumia. 1996. Control of purslane weed by a simple infusion of rue: Biological and chemical aspects. Allelopathy Journal 3: 207–216.
Andersson, L., AND J. H. E. Rova. 1999. The rps16 intron and the phylogeny of the Rubioideae. Plant Systematics and Evolution 214: 161–186.[CrossRef][Web of Science]
APG [Angiosperm Phylogeny Group]. 2003. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG II. Botanical Journal of the Linnean Society 141: 399–436.[CrossRef][Web of Science]
Asmussen, C. B., AND M. W. Chase. 2001. Coding and noncoding plastid DNA in palm systematics. American Journal of Botany 88: 1103–1117.
Bailey, V. L. 1962. Revision of the genus Ptelea (Rutaceae). Brittonia 14: 1–45.[CrossRef]
Baker, W. J., C. B. Asmussen, S. Barrow, J. Dransfield, AND T. A. Hedderson. 1999. A phylogenetic study of the palm family (Palmae) based on chloroplast DNA sequences from the trn L-trnF region. Plant Systematics and Evolution 219: 111–126.[CrossRef][Web of Science]
Barth, M. O. 1982. Variações polínicas em espécies brasileiras da família Rutaceae. Boletim do Instituto de Geociências da Universidade de São Paulo 13: 129–134.
Chase, M. W., C. M. Morton, AND J. A. Kallunki. 1999. Phylogenetic relationships of Rutaceae: A cladistic analysis of the subfamilies using evidence from rbc L and atpB sequence variations. American Journal of Botany 86: 1191–1199.
Chiang, F. 1989. Casimiroa gregii, formerly in Sargentia (Rutaceae). Taxon 38: 116–119.[CrossRef][Web of Science]
Corner, E. J. H. 1976. The seeds of dicotyledons. Cambridge University Press, London, UK.
Cowan, R. S., AND G. K. Brizicky. 1960. Taxonomic relationships of Diomma Engler ex Harms. Memoirs of the New York Botanical Garden 10: 38–64.
Da Silva, M. F. G. F., AND O. R. Gottlieb. 1987. Evolution of quassinoids and limonoids in the Rutales. Biochemical Systematics and Ecology 15: 85–103.[CrossRef][Web of Science]
Da Silva, M. F. G. F., O. R. Gottlieb, AND F. Ehrendorfer. 1988. Chemosystematics of the Rutaceae: Suggestions for a more natural taxonomy and evolutionary interpretation of the family. Plant Systematics and Evolution 161: 97–134.[CrossRef][Web of Science]
Dowton, M., AND A. D. Austin. 2002. Increased congruence does not necessarily indicate increased phylogenetic accuracy—The behavior of the incongruence length-difference-test in mixed-model analyses. Systematic Biology 51: 19–31.[CrossRef][Web of Science][Medline]
Engler, A. 1874. Rutaceae. In C. F. P. Martius, and A. G. Eichler [eds.], Flora brasiliensis, vol. 12, pt. 2, 75–196. Typographia Regia, Munich, Germany.
Engler, A. 1896. Rutaceae. In H. G. A. Engler, and K. Prantl [eds.], Die natürlichen Pflanzenfamilien, ed. 1, Teil 3, Abteilung 5, 95–201. Wilhelm Engelmann, Leipzig, Germany.
Engler, A. 1931. Rutaceae. In H. G. A. Engler, and K. Prantl [eds.], Die natürlichen Pflanzenfamilien, ed. 2, Teil 19a, 187–359. Wilhelm Engelmann, Leipzig, Germany.
Erdtman, G. 1966. Pollen morphology and plant taxonomy: Angiosperms. E. J. Brill, Leiden, Netherlands.
Farris, J. S., M. Källersjö, A. G. Kluge, AND C. Bult. 1994. Testing significance of incongruence. Cladistics 10: 315–319.[CrossRef][Web of Science]
Fay, M. F., K. M. Cameron, G. T. Prance, AND M. W. Chase. 1997. Familial relationships of Rhabdodendron (Rhabdodendraceae): Plastid rbc L sequences indicate a caryophyllid placement. Kew Bulletin 52: 923–932.[CrossRef]
Felsenstein, J. 1985. Confidence limits on phylogenetics: An approach using the bootstrap. Evolution 39: 783–791.[CrossRef][Web of Science]
Fernando, E. S., P. A. Gadek, AND C. J. Quinn. 1995. Simaroubaceae, an artificial construct: evidence from rbc L sequence variation. American Journal of Botany 82: 92–103.[CrossRef][Web of Science]
Fernando, E. S., AND C. J. Quinn. 1995. Picramniaceae, a new family and a recircumscription of Simaroubaceae. Taxon 44: 177–181.[CrossRef][Web of Science]
Fish, F., AND P. G. Waterman. 1973. The chemosystematics of the Zanthoxylum/Fagara complex. Taxon 22: 177–203.[CrossRef]
Fitch, W. M. 1971. Toward defining the course of evolution: Minimum change for specific tree topology. Systematic Zoology 20: 406–416.[Abstract]
Forman, L. L. 1958. The identity of Feroniella pubescens Tanaka (Rutaceae). Kew Bulletin 1957: 503–504.
Forni-Martins, E. R., AND F. R. Martins. 2000. Chromosome studies on Brazilian cerrado plants. Genetics and Molecular Biology 23: 947–955.[Web of Science]
Gadek, P. A., E. S. Fernando, C. J. Quinn, S. B. Hoot, T. Terrazas, M. C. Sheahan, AND M. W. Chase. 1996. Sapindales: Molecular delimitation and infraordinal groups. American Journal of Botany 83: 802–811.[CrossRef][Web of Science]
Gielly, L., AND P. Taberlet. 1994. The use of chloroplast DNA to resolve plant phylogenies: Non-coding versus rbc L sequences. Molecular Biology and Evolution 11: 769–777.[Abstract]
Goldblatt, P., H. Tobe, S. Carlquist, AND V. Patel. 1985. Familial position of the cape genus Empleuridium. Annals of the Missouri Botanical Garden 72: 167–183.[CrossRef][Web of Science]
Grieve, C. M., AND R. W. Scora. 1980. Flavonoid distribution in the Aurantioideae (Rutaceae). Systematic Botany 5: 39–53.[CrossRef][Web of Science]
Guerra, M., K. G. B. Santos, A. E. B. Silva, AND F. Ehrendorfer. 2000. Heterochromatin banding pattern in Rutaceae-Aurantioideae—A case of parallel chromosomal evolution. American Journal of Botany 87: 735–747.
Hartl, D. 1957. Struktur und herkunft des endokarps der Rutaceen. Beiträge zur Biologie der Pflanzen 34: 35–49.
Hartley, T. G. 1974. A revision of the genus Acronychia (Rutaceae). Journal of the Arnold Arboretum 55: 469–523, 525–567.[Web of Science]
Hartley, T. G. 1979. A revision of the genus Tetractomia (Rutaceae). Journal of the Arnold Arboretum 60: 127–153.[Web of Science]
Hartley, T. G. 1981. A revision of the genus Tetradium (Rutaceae). Gardens' Bulletin (Singapore) 34: 91–131.
Hartley, T. G. 1982. A revision of the genus Sarcomelicope (Rutaceae). Australian Journal of Botany 30: 359–372.[CrossRef][Web of Science]
Hartley, T. G. 1985. A revision of the genus Medicosma (Rutaceae). Australian Journal of Botany 33: 27–64.[CrossRef][Web of Science]
Hartley, T. G. 2001. Morphology and biogeography in Australasian-Malesian Rutaceae. Malayan Nature Journal 55: 197–219.
Hartley, T. G. 2003. Neoschmidia, a new genus of Rutaceae from New Caledonia. Adansonia série 3, 25: 7–12.
Holmgren, P. K., N. H. Holmgren, AND L. C. Barnett. 1990. Index Herbariorum, part 1, The herbaria of the world. New York Botanical Garden Press, Bronx, New York, USA.
Holmstedt, B., S. H. Wassén, AND R. E. Schultes. 1979. Jaborandi: An interdisciplinary appraisal. Journal of Ethnopharmacology 1: 3–21.[CrossRef][Web of Science][Medline]
Judd, W. S., C. S. Campbell, E. A. Kellogg, P. F. Stevens, AND M. J. Donoghue. 2008. Plant systematics: A phylogenetic approach, 3rd ed. Sinauer, Sunderland, Massachusetts, USA.
Kaastra, R. C. 1982. Pilocarpinae (Rutaceae). Flora Neotropica Monograph 13. New York Botanical Garden, Bronx, New York, USA.
Kallunki, J. A., AND J. R. Pirani. 1998. Synopses of Angostura Roem. & Schult. and ConchocarpusJ. C. Mikan. Kew Bulletin 53: 257–334.[CrossRef]
Kelchner, S. A., AND L. G. Wendel. 1996. Hairpins create minute inversions in non-coding regions of chloroplast DNA. Current Genetics 30: 259–262.[CrossRef][Web of Science][Medline]
Lal, S., AND L. L. Narayana. 1994. Floral anatomy and systematic position of Flindersia R. Br. Feddes Repertorium (Berlin)105: 31–36.
Leroy, J. F., D. Lobreau-Callen, AND M. Lescot. 1990. Les Ptaeroxylaceae: Espèces nouvelles du genre malgache Cedrelopsis et palynologie de la famille. Adansonia 12: 43–57.
Ma, J. S., W. Cao, Q. R. Liu, M. Yu, AND L. J. Han. 2006. A revision of the genus Phellodendron (Rutaceae). Edinburgh Journal of Botany 63: 131–151.[CrossRef]
Mabberley, D. J. 1998. Australian Citreae with notes on other Aurantioideae (Rutaceae). Telopea 7: 333–344.
Mandalari, G., R. N. Bennett, G. Bisignano, D. Trombetta, A. Saija, C. B. Faulds, M. J. Gasson, AND A. Narbad. 2007. Antimicrobial activity of flavonoids extracted from bergamot (Citrus bergamia Risso) peel, a byproduct of the essential oil industry. Journal of Applied Microbiology 103: 2056–2064.[CrossRef][Medline]
Mason-Gamer, R. J., AND E. A. Kellogg. 1996. Testing for phylogenetic conflict among molecular data sets in the tribe Triticeae (Gramineae). Systematic Biology 45: 524–545.
Metcalfe, C. R., AND L. Chalk. 1950. Anatomy of the dicotyledons. Clarendon Press, Oxford, UK.
Moore, J. A. 1936. Floral anatomy and phylogeny in the Rutaceae. New Phytologist 35: 318–322.[CrossRef]
Moraes, V. R. de S., D. M. Tomaleza, R. J. Ferracin, C. F. Garcia, M. Sannomiya, M. del P. C. Soriano, M. F. G. F. da Silva et al. 2003. Enzymatic inhibition studies of selected flavonoids and chemosystematic significance of polymethoxylated flavonoids and quinoline alkaloids in Neoraputia. Journal of the Brazilian Chemical Society 14: 380–387.[Web of Science]
Morawetz, W. 1986. Remarks on karyological differentiation patterns in tropical woody plants. Plant Systematics and Evolution 152: 49–100.[CrossRef][Web of Science]
Morton, C. M., M. Grant, AND S. Blackmore. 2003. Phylogenetic relationships of the Aurantioideae inferred from chloroplast DNA sequence data. American Journal of Botany 90: 1463–1469.
Morton, C. M., AND J. A. Kallunki. 1993. Pollen morphology of the subtribe Cuspariinae (Rutaceae). Brittonia 45: 286–314.[CrossRef][Web of Science]
Muellner, A. N., R. Samuel, S. A. Johnson, M. Cheek, T. D. Pennington, AND M. Chase. 2003. Molecular phylogenetics of Meliaceae (Sapindales) based on nuclear and plastid DNA sequences. American Journal of Botany 90: 471–480.
Mziray, W. 1992. Taxonomic studies in Toddalieae Hook. f. (Rutaceae) in Africa. Symbolae Botanicae Upsalienses 30: 1–95.
Ng, K. M., P. P. But, A. I. Gray, T. G. Hartley, Y. Kong, AND P. G. Waterman. 1987. The biochemical systematics of TetradiumEuodia and Melicopeand their significance in the Rutaceae. Biochemical Systematics and Ecology 15: 587–593.[CrossRef][Web of Science]
Nixon, K. C., AND J. M. Carpenter. 1996. On simultaneous analysis. Cladistics 12: 221–241.[CrossRef][Web of Science]
Nooteboom, H. P. 1962. Simaroubaceae. Flora Malesiana, series 1 6: 193–226.
Nooteboom, H. P. 1966. Flavonols, leuco-anthocyanins, cinnamic acids and alkaloids in dried leaves of some Asiatic and Malesian Simaroubaceae. Blumea 14: 309–315.
Oliva, A., B. Di Blasio, G. Cafiero, G. Aliotta, R. Iacovino, AND V. De Feo. 2000. Allelochemicals from rue (Ruta graveolens L.) and olive (Olea europaeaL.) oil mill waste waters as potential natural pesticides. Current Topics in Phytochemistry 3: 167–177.
Oxelman, B., M. Liden, AND D. Berglund. 1997. Chloroplast rps16 intron phylogeny of the tribe Sileneae (Caryophyllaceae). Plant Systematics and Evolution 206: 393–410.[CrossRef][Web of Science]
Perrier de la Bathie, H. 1950. Famille 104. Rutacées. In H. Humbert Flore de Madagascar et des Comores, 1–89. Typ. Firmin-Didot and Co. Mesnil, France.
Pirani, J. R. 1998. A revision of Helietta and Balfourodendron (Rutaceae, Pteleinae). Brittonia 50: 348–380.[CrossRef][Web of Science]
Poon, W. S., P. C. Shaw, M. P. Simmons, AND P. P. H. But. 2007. Congruence of molecular, morphological, and biochemical profiles in Rutaceae: A cladistic analysis of the subfamilies Rutoideae and Toddalioideae. Systematic Botany 32: 837–846.[Web of Science]
Prance, G. T. 1968. The systematic position of Rhabdodendron Gilg. and Pilg. Bulletin du Jardin Botanique National de Belgique 38: 127–146.[CrossRef]
Prance, G. T. 1972. Rhabdodendraceae. Flora Neotropica Monograph 11. Hafner, New York, USA.
Price, J. R. 1963. The distribution of alkaloids in the Rutaceae. In T. Swain [ed.], Chemical plant taxonomy, 429–452. Academic Press, New York, New York, USA.
Record, S. J., AND R. W. Hess. 1940. American woods of the family Rutaceae. Tropical Woods 64: 1–28.
Reeves, G., M. W. Chase, P. Goldblatt, T. De Chies, B. Lejeune, B. M. F. Fay, A. V. Cox, AND P. J. Rudall. 2001. Molecular systematics of Iridaceae: Evidence from four plastid DNA region. American Journal of Botany 88: 2074–2087.
Richardson, J. E., R. T. Pennington, T. D. Pennington, AND P. M. Hollingsworth. 2001. Rapid diversification of a species-rich genus of neotropical rain forest trees. Science 293: 2242–2245.
Samuel, R., F. Ehrendorfer, M. W. Chase, AND H. Greger. 2001. Phylogenetic analyses of Aurantioideae (Rutaceae) based on non-coding plastid DNA sequences and phytochemical features. Plant Biology 3: 77–87.[CrossRef]
Scholz, H. 1964. Rutales. In H. Melchior [ed.], A. Engler’s Syllabus der Pflanzenfamilien, vol. 12, part 2, 262–277. Borntraeger, Berlin, Germany.
Scott, K. D., C. L. McIntyre, AND J. Playford. 2000. Molecular analyses suggest a need for a significant rearrangment of Rutaceae subfamilies and a minor reassessment of species relationships within Flindersia. Plant Systematics and Evolution 223: 15–27.[CrossRef][Web of Science]
Small, R. L., J. A. Ryburn, R. C. Cronn, T. Seelanan, AND J. F. Wendel. 1998. The tortoise and the hare: Choosing between non-coding plastome and nuclear ADH sequences for phylogeny reconstruction in a recently diverged plant group. American Journal of Botany 85: 1301–1315.
Smith, J. F. 2000. Phylogenetic signal common to three data sets: Combining data which initially appear heterogeneous. Plant Systematics and Evolution 221: 179–198.[CrossRef][Web of Science]
Soltis, D. E., P. S. Soltis, P. K. Endress, AND M. W. Chase. 2005. Phylogeny and evolution of angiosperms. Sinauer, Sunderland, Massachusetts, USA.
Stace, H. M., J. A. Armstrong, AND S. H. James. 1993. Cytoevolutionary patterns in Rutaceae. Plant Systematics and Evolution 187: 1–28.[CrossRef][Web of Science]
Swingle, W. T., AND P. C. Reece. 1967. The botany of Citrus and its wild relatives. In W. Reuther, H. J. Webber, and L. D. Bachelor [eds.], The Citrusindustry, vol. 1, History, world distribution, botany, and varieties, 190–430. University of California, Berkeley, California, USA.
Swofford, D. L. 2002. PAUP*: Phylogenetic analyses using parsimony (and other methods), version 4. 0b10. Sinauer, Sunderland, Massachussetts, USA.
Taberlet, P., L. Gielly, G. Pautou, AND J. Bouvet. 1991. Universal primers for amplifications of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105–1109.[CrossRef][Web of Science][Medline]
Tanaka, T. 1936. The taxonomy and nomenclature of Rutaceae-Aurantioideae. Blumea 2: 101–110.
Taylor, D. A. H. 1982. The occurrence of limonoids in the Meliaceae. Flora Neotropica Monograph 28, 450–459. New York Botanical Garden, Bronx, New York, USA.
Taylor, D. A. H. 1983. Biogenesis, distribution and systematic significance of limonoids in the Meliaceae, Cneoraceae and allied taxa. In P. G. Waterman, and M. F. Grundon [eds.], Chemistry and chemical taxonomy of the Rutales, 353–375. Academic Press, London, UK.
Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, AND 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.
Trinder-Smith, T. H., H. P. Linder, T. van der Niet, G. A. Verboom, AND T. L. Nowell. 2007. Plastid DNA sequences reveal generic paraphyly within Diosmeae (Rutoideae, Rutaceae). Systematic Botany 32: 847–855.[Web of Science]
van der Ham, R. W. J. M., P. Baas, M. E. Barker, F. D. Boesewinkel, F. Bouman, B. J. van Heuven, AND R. K. W. M. Klaassen. 1995. Bottegoa Chiov. transferred to the Ptaeroxylaceae. Kew Bulletin 50: 243–265.[CrossRef]
Verdcourt, B., AND F. G. Davies. 1996. Ptaeroxylaceae. In R. M. Polhill [ed.], Flora of tropical East Africa. A. A. Balkema, Rotterdam, Netherlands.
Vieira, P. C., A. R. Lázaro, J. B. Fernandes, AND F. G. F. Da Silva. 1988. The chemosystematics of Dictyoloma. Biochemical Systematics and Ecology 16: 541–544.[CrossRef][Web of Science]
Vieira, P. C., A. R. Lázaro, J. B. Fernandes, AND F. G. F. Da Silva. 1990. Limonoids, alkaloids and chromones from Dictyoloma vandellianum, and their chemosystematic sigificance. Química Nova 13: 287–288.
Wallander, E., AND V. A. Albert. 2000. Phylogeny and classification of Oleaceae based on rps16 and trnL-Fsequence data. American Journal of Botany 87: 1827–1841.
Waterman, P. G. 1975. Alkaloids of the Rutaceae: Their distribution and systematic significance. Biochemical Systematics and Ecology 3: 149–180.[CrossRef]
Waterman, P. G. 1983. Phylogenetic implications of the distribution of secondary metabolites within the Rutales. In P. G. Waterman, and M. G. Grundon [eds.], Chemistry and chemical taxonomy of Rutales, 377–400. Academic Press, London, UK.
Waterman, P. G. 1990. Chemosystematics of the Rutaceae: Comments on the interpretation of Da Silva & al. Plant Systematics and Evolution 173: 39–48.[CrossRef][Web of Science]
Waterman, P. G., AND M. G. Grundon. 1983. Chemistry and chemical taxonomy of Rutales. Academic Press, London, UK.
Whitten, W. M., N. H. Williams, AND M. W. Chase. 2000. Subtribal and generic relationships of Maxillarieae (Orchidaceae) with emphasis on Stanhopeinae: Combined molecular evidence. American Journal of Botany 87: 1842–1856.
Williams, I. 1984. Studies on the genera of the Diosmeae (Rutaceae). 16. A key to the genera of Diosmeae Benth. & Hook. (Rutaceae) and a description of a new species of Agathosma (Rutaceae). South African Journal of Botany 51: 149–151.[Web of Science]
Wilson, P. G. 1998. New species and nomenclatural changes in Phebalium and related genera (Rutaceae). Nuytsia 12: 267–288.
Yoder, A. D., J. A. Irwi, AND B. A. Payseur. 2001. Failure of the ILD to determine data compatibility for slow loris phylogeny. Systematic Biology 50: 408–424.
Zakaria, M. B. 2001. The phytochemistry of Rutaceae species with special reference to Melicope. Malayan Nature Journal 55: 241–250.
Zavaleta-Mancera, H. A., AND E. M. Engleman. 1991. Anatomía del fruto de Casimiroa edulis (Rutaceae), "zapote blanco", durante su desarrollo. Boletín de la Sociedad Botánica de México 51: 53–65.
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  R. J. Bayer, D. J. Mabberley, C. Morton, C. H. Miller, I. K. Sharma, B. E. Pfeil, S. Rich, R. Hitchcock, and S. Sykes A molecular phylogeny of the orange subfamily(Rutaceae: Aurantioideae) using nine cpDNA sequences Am. J. Botany, March 1, 2009; 96(3): 668 - 685. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
