| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics |
2Section of Botany, Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213 USA; 3Department of Botany, School of Plant Sciences, Whiteknights, The University of Reading, Reading RG6 6AS UK; 4Regius Keeper, Royal Botanic Garden Edinburgh, 20a Inverleith Row, Edinburgh EH3 5LR UK Scotland
Received for publication February 28, 2002. Accepted for publication May 9, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Aurantioideae • Citreae • Citrus • Clauseneae • rps16 • Rutaceae • trnL-trnF
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). In the most recent classification (Swingle and Reece, 1967
), 33 genera were recognized and grouped into two tribes: the Clauseneae with five genera and the Citreae with 28 genera, including Citrus.
In general, Aurantioideae can be characterized as small trees, shrubs, and, rarely, vines that produce fruit with a leathery rind or hard shell and often with a juicy pulp. The leaves and fruits have schizolysigenous oil glands that release an aroma when touched, and the flowers are typically white and fragrant. The leaves are persistent except in three monotypic genera (Poncirus, Aegle, and Feronia), three species of Clausena, and one species of Murraya (Swingle and Reece, 1967
).
The Rutaceae are native to Africa, Australia, North and South America, and Asia. The genera of Aurantioideae grow in varied climates from equatorial, hot-humid conditions to cool maritime. The world citrus belt extends from over 40° N latitude to almost 40° S latitude, but most fruit production is concentrated between the latitudes of 20° N and 40° S. Generally, the trees and fruits are sensitive to frost and cold and require long, warm summers for the fruit to reach maturity. Twenty-nine of the 33 genera comprising Aurantioideae are indigenous to monsoon regions, which extend from West Pakistan to north-central China then south through the East Indian Archipelago to New Guinea, the Bismarck Archipelago, Australia, New Caledonia, Melanesia, and the western Polynesian islands. The remaining five genera are believed to be native to tropical Africa (Swingle and Reece, 1967
).
Tribal and subtribal classifications are in dispute and differ in the three most recent classifications of Aurantioideae: Engler (1931)
, Tanaka (1936)
, and Swingle and Reece (1967)
(Table 1). Tanaka (1936)
grouped the subfamily into eight tribes and eight subtribes including 28 genera. Tribes Micromeleae, Clauseneae, Aegleae, Lavangeae, Meropeae, Atalantieae, Microcitreae, and Aurantieae are divided by a suite of features such as the number of leaflets, venation of the leaf, origin and development of the winged rachis, presence and appearance of the thorns, the number of the floral organs (stamens, locules, and ovules), the development of the pulp vesicles, the texture of the rind of the fruits, and features of the cotyledons. In 1931, Engler grouped members of the subfamily into a single tribe, Aurantieae, including two subtribes, Hesperethusinae (16 genera) and Citrinae (13 genera). The subtribes are mainly based on the number of ovules per locule, with the Hesperethusinae having an ovary with one or two ovules per locule (except Wenzelia), whereas the Citrinae contain taxa with more than two ovules per locule. Swingle and Reece (1967)
is the classification most widely referred to by modern authors. These authors recognized the subfamily to include two tribes and six subtribes. Tribe Clauseneae contains three subtribes with 28 genera, while tribe Citreae contains three subtribes with five genera. Both Engler (1931)
and Swingle and Reece (1967)
believed that Clauseneae contains the more primitive genera of the subfamily. None of the species of the Clauseneae develop axillary spines, and the odd pinnate leaves have alternately attached leaflets. The fruits are usually small, semi-dry or juicy berries, except in Merrillia, which has large leathery fruits. In the tribe Citreae nearly all the species develop axillary spines, single or paired, sometimes curved as in Luvunga and Paramignya. The leaves are easily distinguished from those of the tribe Clauseneae by either being simple, unifoliolate, or trifoliolate, but a few genera have odd pinnate leaves with opposite leaflets. The subtribe Citrinae differs from all the other subtribes in the subfamily by having pulp vesicles, which arise from the dorsal wall of the locule, growing into the locular cavity and developing into sacs filled with large, thin-walled cells with watery juice. No close homologies are known in any of the higher plants.
|
using evidence from rbcL and atpB sequence variation. These authors found that the Aurantioideae were a well-defined natural group and that members of the Flindersioideae (Chloroxylon) and Rutioideae (Ruta) were basal to the main Aurantioideae clade. We have therefore used Chloroxylon and Ruta as the outgroups for this analysis.
The Aurantioideae are an important group of plants, with many species of commercial importance—for example, the fruits of Citrus and Fortunella. It is therefore important to understand the internal relationships among the different taxa of the subfamily for advancing breeding techniques and developing better conservation strategies. The most current classification by Swingle and Reece (1967)
is based on traditional taxonomic methods using morphology and anatomy. This study examines the phylogenetic relationships within the Aurantioideae using molecular characters and the mapping of selected morphological features onto these trees.
For this study, two noncoding chloroplast regions were selected, the trnL-trnF region and the intron of rps16. The trnL-trnF region consists of the trnL intron and the trnL-trnF intergenic spacer (Taberlet et al., 1991
). The intron of the rps16 is a group II intron that was first used for phylogenetic studies by Oxelman et al. (1997)
. Various workers have found both of these genes give good resolution at the generic and species levels (Baker et al., 2000
; Wallander and Albert, 2000
).
Using the classification of Swingle and Reece (1967)
and the characters just described, we attempted to confirm the monophyly of the subfamily and to investigate the monophyly of the tribal and subtribes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. Organic compounds were removed using 24 : 1 chloroform : isoamyl alcohol (SEVAG), followed by DNA precipitation with ice-cold isopropanol.
rps16
The rsp16 gene in 26 genera was amplified using the primer pair rpsF/rpsR2 (Oxelman et al., 1997
) to acquire the entire region. The final polymerase chain reaction (PCR) cocktail of 50 mL contained the following: 38 mL water, 5 mL 10% buffer, 3 mL Mg+2, 1 mL dNTPs, 0.25 mL Taq polymerase, and 0.5 mL of each primer. These amplifying reactions were run for 25 cycles of denaturing for 30 s at 95°C, primer annealing for 50 s at 57°C, and elongation for 2 min at 72°C. Some taxa required altered amplification conditions (e.g., magnesium chloride concentration, the addition of tetramethylammonium chloride, and annealing temperature).
trnL-trnF
The trnL intron and the trnL-trnF intergenic spacer for 26 genera were amplified. The PCR was performed using universal primers trn-c, trn-d, trn-e, and trn-f as described by Taberlet et al. (1991)
. For some samples, the entire trnL intron/trnL-trnF spacer region was amplified with trn-c and trn-f, whereas in others, two separate amplifications were performed, one to amplify the trnL intron with trn-c and trn-d, the other to amplify the trnL-trnF spacer with trn-e and trn-f. In general, each 50-µL amplification reaction contained the same proportions as in the rps16 reactions. The PCR amplification used a 7-min denaturing step at 94°C followed by 30 cycles of denaturing for 1 min at 94°C, primer annealing for 1 min at 45°C, and elongation for 1 min at 72°C, with a final 7-min elongation step at 72°C.
Cycle sequencing
The PCR products were cleaned using the QIAGEN QIAquick PCR purification kit (QIAGEN, Chatsworth, California, USA) following the protocols provided by the manufacturer. Cleaned products were then directly sequenced using the ABI PRISM Dye Terminator Cycle Sequencing Ready Kit with AmpliTaq DNA Polymerase (Applied Biosystems, Foster City, California, USA). Unincorporated dye terminators were removed using the QIAGEN DyeEx dye-terminator removal system following the recommendations provided by the manufacturer. Samples were then loaded into an ABI 3100 DNA Sequencer. The sequencing data was analyzed and edited using the Sequencher software program (Gene Codes, Ann Arbor, Michigan, USA).
Phylogenetic analysis
Boundaries of the trnL intron and the trnL-trnF intergenic spacer and rps16 were determined by comparison with sequences in Genbank. The following alignment criteria and methodology were used. When two or more gaps were not identical but overlapping, they were scored as two separate events. Phylogenetically informative indels (variable in two or more taxa) were scored as one event at the end of the data set. All DNA sequences reported in the analyses have been deposited in Genbank (see Appendix avaliable as Supplementary Data accompanying the online version of this article).
All phylogenetic analyses employed maximum-parsimony with the heuristic search option in PAUP* 4.0b8 (Swofford, 2000
) with uninformative characters excluded. Searches were conducted with 100 random-taxon-addition replicates with tree bisection-reconnection (TBR) branch swapping, steepest descent, and MulTrees selected, and with all characters and states weighted equally and unordered. All trees from the replicates were then swapped onto completion, all shortest trees were saved, and a strict consensus tree was computed. Relative support for individual clades was estimated with the bootstrap method (Felsenstein, 1985
). One thousand pseudoreplicates were performed with uninformative characters excluded. Ten random-taxon-addition heuristic searches for each pseudoreplicate were performed, and all minimum-length trees were saved per search. To reduce bootstrap search times, branches were collapsed if their minimum length was zero ("amb-").
To determine the combinability of data sets, the data structure of the two data sets was compared by methods outlined by Mason-Gamer and Kellogg (1996)
. These authors discuss various ways in which conflict between data sets can be assessed. In one method, the combination of independent data sets is possible if the trees do not conflict or if conflict receives low bootstrap support. Therefore, each node on the independent trees is tested for congruence against the other. If the nodes do not contain conflicting information, they are congruent and the data sets are combinable. Where there are incongruent nodes, the bootstrap values for each node are examined. If the support is less than 70%, then there is no hard conflict and the incongruence is interpreted as being due to chance.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
trnL-trnF
The length of the trnL intron ranged from 509 to 592 base pairs (bp) among species of Aurantioideae and from 544 to 558 bp among the outgroups. The length of the trnL-trnF intergenic spacer ranged from 367 to 457 bp among the taxa of Aurantioideae and from 398 to 416 bp among the outgroups. Multiple sequence alignment of Aurantioideae and all outgroups resulted in a matrix of 1183 characters (651 characters in the intron, 532 in the spacer), of which 240 (20%) included at least one accession with a gap (118 of 651 positions [18%] in the intron and 122 of 532 positions [23%] in the spacer). Unweighted pairwise sequence divergence among species of Aurantioideae ranged from 0.2 to 5.1% in the intron and from 0 to 5.7% in the spacer; that between species of Aurantioideae and the outgroups ranged from 5.1 to 11.9% in the intron and 7.6 to 18.0% in the spacer. A total of 34 gaps were required for proper alignment of the trnL-trnF sequences. These gaps ranged from 1 to 18 bps, with the average size being 
5–6 bp. Sixteen gaps were scored as binary characters. Excluded regions included positions 1–40, 161–168, and 1141–1183, because of missing data and/or homoplasy within these regions. Mean percentage guanine and cytosine (G + C) content was 36% in the intron and 38% in the spacer.
Of the 1183 positions constituting the aligned trnL-trnF sequences, 269 (22.9%) were variable and 173 (14.6%) were parsimony-informative.
The analysis recovered 18 equally optimal trees of 139 steps (consistency index [CI] = 0.78, retention index [RI] = 0.79; Fig. 1).
|
rsp16
The length of the rps16 ranged from 814 to 895 bp among species of Aurantioideae and from 846 to 857 bp among the outgroups. Multiple sequence alignment of Aurantioideae and all outgroups resulted in a matrix of 974 characters, of which 167 (17%) include at least one accession with a gap. Unweighted pairwise sequence divergence among species of Aurantioideae ranged from 0 to 6.0%; that between species of Aurantioideae and the outgroups ranged from 7.5 to 11.7%. A total of 31 gaps were required for proper alignment of the rps16 sequences. These gaps ranged from 1 to 16 bp, with the average size being 1–2 bp. Eight gaps were scored as binary characters. Excluded characters included positions 1–33, 108–120, 420–460, and 935–974, from high homoplasy within these regions. Mean percentage G + C content was 34%.
Of the 974 positions constituting the aligned rps16 sequences, 242 (24.8%) were variable and 167 (17.1%) were parsimony informative. The analysis recovered 390 equally optimal trees of 126 steps (CI = 0.67, RI = 0.71; Fig. 2). The Aurantioideae were supported as monophyletic in the strict consensus of these trees (BS 100). Triphasia was sister to a clade comprising the rest of the subfamily. Within this large clade, there were 10 minor clades. Five of these clades consisted of two taxa clades: Citropsis and Hesperethusa (BS 90) and Atalantia and Severinia (BS 97) from the subtribe Citrinae; Feroniella and Feronia (BS 98) from the subtribe Balsamocitrinae; Murraya and Merrillia (BS 81) from the tribe Clauseneae and subtribes Clauseninae and Merrilliinae; and Clausena and Wenzelia from both tribes and the subtribes Triphasiinae and Clauseninae. The sixth clade contained members of both tribes and the subtribes Balsamocitrinae and Clauseninae. The seventh clade was a mostly an unresolved polytomy of members of the Citreae and the Citrinae [(Microcitrus and Eremocitrus [BS 83] was sister to Clymenia [BS 73], which forms a polytomy with Citrus, Poncirus and Fortunella [BS 86]) while Swinglea, Pleiospermium, and Pamburus were interdigitated among the previous clades.
|
and recently applied by Eldenäs and Linder (2000)
, the data sets were considered combinable. Seven nodes were shared between the independent strict consensus trees. Bootstrap support for these seven nodes ranged from 51 to 100%. In both analyses, the subfamily was monophyletic with a 100% bootstrap support. In only two positions were there conflicts between the rps16 and trnL-trnF topologies. In the rps16 tree, Citropsis and Hesperethusa were held together with 90% bootstrap support, whereas in the trnL-trnF tree these two taxa grouped with Pleiospermium with only 51% bootstrap support. In addition, within the rps16 tree, Clymenia grouped with Microcitrus and Eremocitrus with a 73% bootstrap value, whereas in the trnL-trnF it was unresolved in this clade. There were no hard conflicts between these trees. Therefore, the incongruence was interpreted as being due to chance and the data sets were combined. The length of the combined matrix ranged from 1783 to 1896 bp among species of Aurantioideae and from 1799 to 1820 bp among the outgroups. Multiple sequence alignment of Aurantioideae and all outgroups resulted in a matrix of 2157 characters, of which 407 (18.9%) included at least one accession with a gap. Unweighted pairwise sequence divergence among species of Aurantioideae ranged from 0.7 to 4.7%; that between species of Aurantioideae and the outgroups ranged from 7.7 to 12.1%. Mean percentage G + C content was 35.8%.
Of the 2157 positions constituting the aligned sequences, 511 (23.7%) were variable and 340 (15.8%) were parsimony informative.
The analysis recovered two equally optimal trees of 319 steps (CI = 0.62, RI = 0.60; Figs. 3 and 4). The Aurantioideae were supported as monophyletic in the strict consensus of these trees (BS 100). The ingroup consisted of a ladder of four clades. The first clade contained Glycosmis and Murraya (BS 70) from the tribe Clauseneae and the subtribe Clauseninae. The second clade was made up of Wenzelia (Triphasiinae) and Atalantia (Citrinae) of the tribe Citreae. The third clade consisted of members of both tribes and from the subtribes Balsamocitrinae, Citrinae, and Clauseninae. The fourth clade contained members from both tribes and four subtribes. The first branch was a trichotomy consisting of Clade A, (Triphasia [Triphasiinae] and Hesperethusa [Citrinae], BS 84; Merrillia [Merrilliinae], BS 87, and Pamburus [Triphasiinae]. Clade A consisted of a trichotomy with Feroniella and Feronia, BS 70), Swinglea, and the remaining members from the Citrinae.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
70%) than that compared with the trnL-trnF region (8 BS 70%). Although the rps16 contained slightly more homoplasy, it provided more supported resolution with roughly the same number of informative characters as that of the trnL-trnF region. The rps16 had several characteristics, making it attractive for comparative sequencing studies. The pair of universal primers designed by Oxelman et al. (1997)
was sufficient to amplify and sequence the entire region. Although in some taxa, the parts rich in adenine and thymine pose sequencing problems, the addition of tetramethylammonium chloride in the PCR reactions usually resolved this problem. We found the rps16 was useful in phylogenetic studies below the family level. This gene provided more supported resolution than the trnL-trnF in this study.
|
, Scott et al. (2000)
, and Samuel et al. (2001)
. A number of morphological, cytological, and biochemical features support the circumscription of this subfamily.
The subfamily can be characterized by features such as leaves and fruit with schizolysigneous oil glands, flowers usually white and aromatic, fruit a berry (or hesperidium), and seeds without endosperm, sometimes with two or more nucellar embryos. A survey of the cytological literature by Stace et al. (1993)
indicated that the Aurantioideae have a basic chromosome number of x = 9, which was taxonomically informative since it is markedly distinct from the other subfamilies of Rutaceae. Biochemical studies conducted by Waterman (1975)
using alkaloids, by Grieve and Score (1980)
using flavonoids, and by Samuel et al. (2001)
using coumarins and flavonioids also strongly support this subfamilial grouping. Most of the characters described, except for chromosome numbers, cannot be used in a cladistic analysis because of the lack of information or the need for reevaluation at the generic level.
Circumscription of the tribes
Swingle and Reece's (1967)
delimitation of Clauseneae from Citreae was not supported by rps16 or the combined analyses and was unresolved from the lack of support in the trnL-trnF analysis. In the rps16 analysis, Clauseneae occurred within clades containing genera of the Citreae. Clausena (Clauseneae) and Wenzelia (Citreae) formed a clade, and Glycosmis (Clauseneae), Balsamocitrus, Afraegle, and Aegle (Citreae) formed a weakly supported clade (BS 62). Within the trnL-trnF strict consensus tree, similar major clades to that of the rps16 were found, although no clades had intermixing of taxa from the Citreae and Clauseneae in the rps16 tree. In the strict consensus tree of the combined analysis, both Merrillia and Clausena of the Clauseneae formed part of two separate clades with members of Citreae, thus supporting that the two tribes were not monophyletic. All morphological characters that Swingle and Reece (1967)
used in delimiting the two tribes broke down upon thorough examination. Characters used by Swingle and Reece to circumscribe the Clauseneae included mostly odd-pinnate leaves, lack of axillary spines, rachis usually not winged, and ovary having 2–5 locules with only one or two ovules in each locules (except in Merrillia). The Citreae were characterized by leaves mostly simple, unifoliolate or trifoliolate, the development of axillary spines in almost all species, most taxa with winged rachis, and ovary having 2–20 locules each with 1–18 ovules in each locule.
Several of these traditionally used morphological features, such as the number of locules, the number of ovules per locule, the number of stamens and the number of petal and sepals per flower, were examined in a phylogenetic context. Because none of these features was discrete, they were not mapped onto the combined tree. Morphological features including pulp present or absent, spines present or absent, and chromosome numbers were mapped on the combine tree. In examining Fig. 5C, it becomes clear that the transition from one state into another took place repeatedly and was not unidirectional except for chromosome number, which indicated that the Aurantioideae do have a basic chromosome number of x = 9. All three of the described characters were synapomorphies at the family level (Fig. 5). An attempt to use chemical characters such as carbazole (present or absent) was made; however, over 40% of the genera had missing data, so this character was not mapped. A reexamination of these and additional morphological features is needed to provide discrete characters for a phylogenetic analysis.
|
The three subtribes of the Clauseneae were distinguished based on petals being valvate or imbricate, cotyledon thickness, size of flowers and fruits, number of ovaries and ovules per locule, and the appearance of exocarp and seed covering.
The subtribes of Citrinae were separated mainly based on fruit size, thickness and texture of the exocarp, the presence of well-developed pulp vesicles, and the number of stamens. Most of these characters are in need of reexamination, which is not possible at this time because of the lack of material, to assess the homology and pattern of variation before they can be used in a cladistic analysis. As part of this research program, an investigation of pollen morphology has already been undertaken (Grant et al., 2000
).
Conclusion
The Aurantioideae as traditionally circumscribed are monophyletic. The two tribes, Citreae and Clauseneae, are not monophyletic, and the five sampled subtribes are most likely also not monophyletic. From this analysis, it is apparent that more studies are needed to provide information about the homology and development of the axillary spines and the various flower and fruit characteristics traditionally used in the classification of this group. Furthermore, additional research is needed to determine the number of potentially useful morphological characters. Many morphological features (i.e., ontogeny of the locule wall) have not been revised across all genera. Before taxonomic alignments can be proposed, studies using more genera, inclusion of genes from the nuclear region, and anatomical and morphological characters are needed to resolve the circumscription and generic relationships within the tribes and subtribes.
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Chase M. W. C. M. Morton J. A. Kallunki 1999 Phylogenetic relationships of Rutaceae: a cladistic analysis of the subfamilies using evidence from rbcL and atpB sequence variation. American Journal of Botany 86: 1191-1199
Doyle J. J. J. L. Doyle 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15
Eldenäs P. K. H. P. Linder 2000 Congruence and complementarity of morphological and trnL-trnF sequence data and the phylogeny of the African Restionaceae. Systematic Botany 25: 692-707[CrossRef][ISI]
Engler A. 1931 Rutaceae. In A. Engler and K. Prantl [eds.], Die natürlichen Pflanzenfamilien, 2nd ed., vol. 19a, 187–359. Engelmann, Leipzig, Germany
Felsenstein J. 1985 Confidence limits of phylogenies: an approach using the bootstrap. Evolution 39: 783-791[CrossRef][ISI]
Grant M. S. Blackmore C. M. Morton 2000 Pollen morphology of the subfamily Aurantioideae (Rutaceae). Grana 39: 8-20[CrossRef][ISI]
Grieve C. M. R. W. Score 1980 Flavonoid distribution in the Aurantioideae (Rutaceae). Systematic Botany 5: 39-53
Mason-Gamer R. J. E. A. Kellogg 1996 Testing for phylogenetic conflict among molecular data sets in the tribe Triticeae (Gramineae). Systematic Botany 45: 524-545
Oxelman B. M. Liden D. Berglund 1997 Chloroplast rps16 intron phylogeny of the tribe Sileneae (Caryophyllaceae). Plant Systematics and Evolution 206: 393-410[CrossRef][ISI]
Samuel R. F. Ehrendorfer M. W. Chase H. Greger 2001 Phylogenetic analyses of Aurantioideae (Rutaceae) based on non-coding plastid DNA sequences and phytochemical features. Plant Biology 2001: 77-87
Scott K. D. C. L. McIntyre J. Playford 2000 Molecular analyses suggest a need for a significant rearrangement of Rutaceae subfamilies and a minor reassessment of species relationships within Flindersia. Plant Systematics and Evolution 223: 15-27[CrossRef][ISI]
Stace H. M. J. A. Armstrong S. H. James 1993 Cytoevolutionary patterns in Rutaceae. Plant Systematics and Evolution 187: 1-28[CrossRef][ISI]
Swingle W. T. P. C. Reece 1967 The botany of Citrus and its wild relatives. In W. Reuther, H. J. Webber, and L. D. Batchelor [eds.], The citrus industry, vol. 1, History, world distribution, botany, and varieties, 190–430. University of California, Berkeley, California, USA
Swofford D. L. 2000 PAUP: phylogenetic analysis using parsimony, version. 4 beta8. Sinauer, Sunderland, Massachusetts, USA
Taberlet P. L. Gielly G. Pautou J. Bouvet 1991 Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105-1110[CrossRef][ISI][Medline]
Tanaka T. 1936 The taxonomy and nomenclature of Rutaceae-Aurantioideae. Blumea 2: 101-110
Wallander E. V. A. Albert 2000 Phylogeny and classification of Oleaceae based on rps16 and trnL-F sequence 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]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
