| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Jodrell Laboratory, Royal Botanic Gardens, Kew Richmond, Surrey, TW9 3DS, United Kingdom
Received for publication September 22, 1997. Accepted for publication July 28, 1998.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: character evolution • internal transcribed spacer (ITS); Korthalsella • molecular phylogeny • species delimitation • trnL-F • Viscaceae
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
, for instance, recognized 23 species and Barlow (1983)
estimated 30, whereas Molvray (1997)
recognized eight. The disagreement between treatments is due to an absence of taxonomically informative characters resulting from the reduction of many organ systems, thus focusing attention on relatively few features, most of which are of dubious taxonomic utility. Reduction is most extreme within the vegetative portion of the plant. The vestigial leaves are represented by minute crests of tissue at the nodes (Fig. 2A, "br") and display no taxonomically significant variation among species. Korthalsella lacks roots, and its haustorium has not been examined for phylogenetically informative characters. Morphological variation is found in color and size of plants, cladotaxy, and shape of the branches and their internodes.
|
Individual flowers are reduced, unisexual, generally less than 1 mm in length, with the perianth reduced to three, somewhat fleshy subdeltoid sepals. Staminate flowers (Fig. 2D–E) have three confluent anthers forming a six-locular synandrium enclosed by the sepals. The synandrium is one of the diagnostic features of Korthalsella and a structure unique in the angiosperms, but it is invariant within the genus. The highly reduced pistillate flowers (Fig. 2C) have an umbonate stigma partially enclosed within the sepals and are similar among species of Korthalsella. Mature fruits are subclavate and may attain a size of 3.5 mm in larger plants, but they have no taxonomically useful characters associated with them.
All the variable characters mentioned above, plant size and color, branch shape and orientation, and location of inflorescences and number of flowers per cushion, have been used in previous treatments of Korthalsella (Van Tieghem, 1896
; Engler, 1897
; Danser, 1937
, 1940
), but their validity as taxonomically informative characters was never assessed. Van Tieghem (1896)
recognized Korthalsella and two other segregate genera, Bifaria and Heterixia, as distinct from Viscum on the basis of their inflorescence cushions, which are unique among Viscaceae. Van Tieghem was aware of the synandrium but did not explicitly discuss this structure. His three genera were characterized as follows: Korthalsella had "unspecialized" inflorescence branches and decussate stems; Bifaria also had "unspecialized" inflorescence branches, but distichous stems; and Heterixia had "specialized" inflorescence branches. Working largely from specimens at the Paris Herbarium, Van Tieghem described over 60 species in these three genera based in many cases on minute variations in color or internode shape.
The following year Engler (1897)
reduced Bifaria and Heterixia to sections of Korthalsella. He also published a number of new combinations within Korthalsella necessitated by these changes in rank but did not formally reevaluate any of Van Tieghem's species. However, he suggested that some of Van Tieghem's species might better be given infraspecific rank. Lecomte (1916)
also dealt with Korthalsella, although he did not propose a formal revision of the genus. His work has a number of taxonomic errors, but his lasting contribution was explicitly recognizing the significance of the synandrium.
The most recent revision of Korthalsella is that of Danser (1937
, 1940)
. He recognized that many of the fine distinctions drawn between species by previous workers could not be supported. Accordingly, he reduced the number of taxa to 23 species and ten varieties. Danser maintained the infrageneric classification established by Engler (1897)
and continued to use cladotaxy and the presence of "specialized" inflorescence branches to distinguish the three sections of the genus. An amended synopsis of Danser's treatment of Korthalsella appears in Table 1. By accepting many of the taxonomic criteria used by Van Tieghem, but broadening the circumscription of some species to accommodate the apparent plasticity of these characters, Danser effectively inherited the problems that plagued Van Tieghem's treatment. Danser's solution appears to have been to assume that sympatry meant conspecificity, hence specimens from one land mass would tend to be placed in the same species. Continents therefore had highly variable species found in numerous populations, whereas species found on islands often had lower variability and were found in only one or a few populations.
|
undertook a morphometric study of shape and size characters previously used to delimit taxa in Korthalsella. Multivariate statistical analysis of quantitative characters demonstrated that variations in plant size and internode length and shape were continuous and could not be partitioned into discrete character states. Stevens (1991)
argued that continuous characters with no discontinuities in their states were best avoided in phylogenetic analyses. Of all the morphological characters used by previous workers, only the presence of differentiated inflorescence branches proved taxonomically reliable at the specific rank. Comparative anatomical studies also shed light on taxonomic relationships. Differences in bundle number could consistently differentiate among taxa: some had four or fewer bundles, others had eight or more (Touw, 1984
). Using the characters that morphological and anatomical studies indicated were reliable, Molvray (1997)
recognized eight species and proposed a new infrageneric classification based on presence or absence of differentiated inflorescence branches. Cladotaxy, the second character formerly used to delimit sections, was proven unreliable at any rank. Field work demonstrated that this character sometimes varies within populations.
Since the classification proposed for Korthalsella by Molvray (1997)
differs significantly from previous treatments and there are few reliable morphological or anatomical characters, DNA sequence data were used to evaluate species delimitations and examine infrageneric relationships. Portions of two genomes were used, the trnL-F region from the plastid and the internal transcribed spacer (ITS) regions of the nuclear ribosomal DNA (nrDNA). Both of these contain noncoding regions that are characterized by sometimes higher nucleotide substitution rates (Taberlet et al., 1991
; Baldwin, 1993
; Baldwin et al., 1995
) and have proven useful for elucidating phylogenetic relationships at the infrageneric level (Baldwin, 1993
; Hahn and Sytsma, 1993
; Bogler and Simpson, 1996
; Freemann and Ybarra, 1996
). In addition, ITS has already been used infragenerically within Viscaceae to clarify relationships in Arceuthobium (Nickrent, Schuette, and Starr, 1994
) and Phoradendron (Ashworth, 1996
).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), as well as three other genera of Viscaceae: Notothixos, Viscum, and Phoradendron. Unfortunately, two species of Korthalsella, K. dacrydii and K. geminata, are exceedingly rare. Repeated attempts over a 10-yr period to obtain additional collections for DNA extraction, as well as attempts to use existing herbarium material for this purpose have failed. As a result these taxa are not included here.
|
5% of the herbarium samples yielding high molecular weight DNA. Genomic DNA was extracted following the CTAB method described by Doyle and Doyle (1987) and purified by CsCl-ethidium bromide centrifugation. After centrifugation the portion of the sample containing the DNA was extracted with water-saturated n-butanol to remove the ethidium bromide and dialyzed to remove excess salts. The trnL-F and ITS regions were amplified from the purified genomic DNA using the polymerase chain reaction (PCR; Mullis and Faloona, 1987
). Primers used were as follows: trnL-F "c" and "f" (Taberlet et al., 1991
), which amplified the intron, 3' exon, and intergenic spacer, and "ITS5" and "ITS4" (White et al., 1990), which amplified ITS1, the 5.8S gene, and ITS2. For both trnL-F and ITS, all segments were included in separate as well as combined analyses. The double-stranded PCR products were purified using Promega Wizard PCRTM minicolumns (Promega Corp., Madison, Wisconsin) in accordance with the manufacturer's protocols, and 10 µL sequencing reactions were carried out in a Perkin-Elmer GenAmp model 9600 thermocycler using the Applied Biosystems Inc. (ABI) Taq DyeDeoxy Terminator Cycle Sequencing KitTM (Perkin-Elmer Corp., Norwalk, Connecticut). Sequencing was done on an ABI model 373 automated sequencer, and each region was sequenced for both strands. Sequences were assembled and edited electronically using SequencherTM software from Gene Codes Corporation (Ann Arbor, Michigan). The completed sequences were manually aligned prior to analysis.
Data analysis
Posted sequences of nrDNA for other genera of Viscaceae obtained from GenBank were used to determine the boundaries of the ITS1 and ITS2 regions. The boundaries of the trnL intron and trnL-F intergenic spacer were identified by comparing sequences with posted sequences from Genbank of the highly conserved trnL 5' and 3' exon and the trnF gene.
Phylogenetic analysis was performed using PAUP Version 3.1.1 (Swofford, 1993
) and Macintosh computers. Data were analyzed as separate ITS and trnL-F sets, and then as a combined data set. All transformations were unordered and weighted equally in both data sets (Fitch parsimony; Fitch, 1971
); gaps were coded as missing values, but also separately coded as binary characters using PaupGap (Cox, 1997
) and appended to the respective data sets. Heuristic analyses were performed on all data sets (ITS, trnL-F, and combined) with tree-bisection-reconnection (TBR) branch-swapping and the MULPARS option in effect. Tree space was searched by using 100 random taxon addition sequence replicates with a limit of five trees saved per replicate. All trees saved were then swapped to completion. This method is applied to increase the chances of finding the shortest tree(s) (Olmstead et al., 1993
). In trnL-F for which more than one most parsimonious tree was obtained, trees resulting from the heuristic search were successively weighted to reduce the influence of homoplasious characters (Farris, 1969
). Examination of downweighted characters showed these to be among the most variable in trnL-F with three to eight changes per site. The combined and ITS data sets were also successively weighted to facilitate comparison among tree scores. Successive weighting did not alter tree topology, but rather produced one of the trees found with Fitch weights. Relative support for the clades identified by the parsimony analysis was assessed by bootstrapping (Felsenstein, 1985
), with 1000 replicates for each matrix.
Potential rooting within Korthalsella was evaluated by comparing ITS and trnL-F sequences from Viscum album, Notothixos subaureus, and Phoradendron serotinum with those of Korthalsella. These ITS sequences were too divergent to be aligned with sequences from Korthalsella, thereby precluding the use of these genera as outgroups. However, the trnL-F region from these genera could be aligned with sequences from Korthalsella, and an analysis based upon these aligned sequences roots Korthalsella at the midpoint of the longest branch. (Midpoint rooting of the ITS tree provides the same tree topology.) Thus, the K. papuana clade was sister to the rest of the genus in the trnL-F analysis and was used as the outgroup in subsequent analyses of both trnL-F and ITS matrices. Since not all genera within Viscaceae were available for the trnL-F analysis, the sister group to Korthalsella was identified by a combined analysis of ten 18S rDNA and plastid rbcL sequences obtained from GenBank. These sequences represent all seven genera of Viscaceae. Outgroups were two genera of Santalaceae and one genus of Loranthaceae. The taxa used, their Genbank accession numbers and individuals who posted these sequences are listed in Table 3.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
1:2 ratios, i.e., an excess of transversions. The mean G + C content of the three regions within the trnL-F sequences is as follows: 32.7% for the intron, 48.3% for the 3' exon, and 26.5% for the intergenic spacer. Details of measures of variability can be seen in Table 4.
|
|
) under a number of different assumptions. Depending on the gap penalty the number of potentially informative characters varies from 259 to 300, whereas the branch length between Korthalsella and the other genera ranges from 235 to 237 steps. Regardless of which gap penalty is chosen, the number of changes in ITS between Korthalsella and the other genera included in this study is greater than the number of variable sites within Korthalsella and is too great to allow any valid conclusions to be drawn about intergeneric relationships. These findings agree with previous reports that the Viscaceae have highly divergent ITS sequences at the generic level (Nickrent, Schuette, and Starr, 1994
; Baldwin et al., 1995
). Furthermore, the topology within Korthalsella is unstable depending on the gap penalty. Since outgroup genera could not be aligned, Korthalsella papuana was used as the outgroup in ITS analyses based on results from trnL-F analysis.
The aligned region comprising ITS1, 5.8S, and ITS2 in Korthalsella consists of 717 bp. The shortest sequence (650 bp) is found in K. japonica from Mauritius. The longest sequence (679 bp) is the outgroup, K. papuana. This is the longest ITS region reported for angiosperms (Baldwin et al., 1995
). ITS1 extends for 278 bp, of which 76 (27.3%) are potentially phylogenetically informative, and contributes 183 steps in the combined analysis. The 5.8S gene is 168 bp long with nine (5.4%) informative characters, and provides 24 steps. The ITS2 region extends for 271 bp, yields 71 (26.2%) informative characters, and provides 188 steps. As the step lengths and numbers of informative sites indicate, the substitution rate for ITS is 2.5x that of the trnL-F region in Korthalsella when their size difference is taken into account. ITS1 and ITS2 have an 1:1 ratio of transitions to transversions, while the ratio in the 5.8S gene is 2:1, i.e., an excess of transitions. The mean G + C content of the ITS region is as follows: 55.2% for ITS1, 52.1% for the 5.8S gene, and 51.2% for ITS2. (Data are summarized in Table 4.)
After alignment there are 81 indels in the ITS data matrix. These indels range in size from 1 to 26 bp; 67% are only 1 bp long, 15% are 2 or 3 bp, while the remaining 18% are >4 bp. Fifty-nine of these indels (72%) are phylogenetically informative. Six indels of particular interest to this study are mapped onto the cladogram in Fig. 5. One of these, a 15-bp deletion at positions 82 to 96 (I-1) is shared by the New Zealand sequences (clade "B" in Fig. 5). A second, a 3-bp deletion from position 99 to 101 (I-2), is found only in the Hawaiian sequences. A third, a 26-bp deletion from position 223 to 248 (I-4) is common to the Mascarene collections. The remaining three indels, a 5-bp insertion at positions 195 to 199 (I-3), a 2-bp insertion at positions 528 to 529 (I-5) and a 1-bp deletion at position 688 (I-6), are all shared by the sequences grouped as clade "C" in Fig. 5. As in the trnL-F study, there is no unique indel that is found only among all of the Australian samples.
|
Our analysis of the ITS data yielded one most parsimonious tree, 531 steps long, with a consistency index of 0.71 and a retention index of 0.84 (Fig. 5). Successive weighting results in the same tree (CI = 0.90, RI = 0.95, weighted length 311 618). If gaps are excluded, two minimal trees are obtained (not shown; CI = 0.76, RI = 0.87, length 392), which are maintained after successive weighting (CI = 0.90, RI = 0.95, weighted step length 260 453). The topology of the ITS tree is similar to those based on trnL. Korthalsella papuana is sister to all other species within the genus, and these species are further resolved into two well-supported subclades corresponding to the taxa with differentiated inflorescences branches and those lacking them (clades B and C in Fig. 5). The composition of all three clades, A, B, and C, is identical in the two data sets.
An analysis based upon both data sets echoes the results found in the separate analyses, but with stronger levels of bootstrap support for many of the clades. This analysis yielded one most parsimonious tree, 1108 steps long, with a consistency index of 0.91 and a retention index of 0.95 (Fig. 6). All of the clades are fully resolved, and it is evident in Fig. 6 that many of the species formerly recognized by Danser (1937,
1940
) and Barlow (1983)
are polyphyletic.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), using 25–30 individuals from each population, showed that almost every population is resolved as its own evolutionary unit. Because Korthalsella populations probably seldom interbreed since neither seeds nor pollen move far from the parent plant, such a pattern among populations is expected. Seeds are deposited near or even on the parent plant except in rare cases of animal-mediated dispersal, and the sticky pollen is carried on the feet of minute thrips, mites, and similar arthropods with limited mobility. Thus, the evolutionary units in the genus are considerably more restricted than the morphological lineages, a situation also found in cave spiders, which have similar problems bridging gaps between populations (Hedin, 1997
). The molecular data could be used to hypothesize the existence of microspecies, morphologically undiagnosable and recognizable only by their DNA sequences. Such a proliferation of unrecognizable species is both philosophically and practically unattractive (Olmstead, 1995
).
If species delimitations are to be drawn more broadly, however, there needs to be some consistency in circumscription. Without reliance on branch shape or other labile characters Molvray (1997)
found only eight diagnosable species in Korthalsella: K. geminata, K. papuana, K. lindsayi, K. salicornioides, K. dacrydii, K. taenioides, K. japonica, and K. cylindrica. However, two of these, K. cylindrica and K. japonica, need further molecular study to determine whether they can be supported. The existing molecular evidence indicates that both taxa may be polyphyletic. In our study K. cylindrica appears embedded within the larger clade comprising K. taenioides (clade C in Fig. 6), and a partial ITS sequence from a Tahitian specimen referable to K. cylindrica is not placed on the same clade as the Hawaiian material. Similarly, Molvray (1997)
distinguished K. japonica from K. taenioides on the basis of a difference in the number of internodal vascular bundles (two or four in the former as opposed to eight or more in the latter), but the molecular phylogeny indicates that a reduction in bundle number has occurred at least three times within the clade with undifferentiated inflorescence branches (indicated in Fig. 6).
It is worth stressing that species concepts have a predictive component (O'Hara, 1993
) in that they seek to discover lineages that will remain discrete in the future, not just those that have been so in the past. Hence, although it appears counterintuitive initially to group very different morphs, such groupings are consistently diagnosable, they are monophyletic, and they are unlikely to be mere accidents of geographic distance or the chance visits of birds. The high level of variability in some aspects of these species can be understood in terms of the considerable isolation of most populations.
Using broad species delimitations, data from the two gene trees yield the same species tree composed of a few morphologically plastic taxa. Supporting broad delimitation is some evidence of gene flow within species. Gene flow can result in differences in tree topology between the nuclear, biparentally inherited ITS and plastid, probably uniparentally inherited trnL-F that is more likely to show the effects of lineage sorting. There is at least one instance of a well- to moderately supported incongruence in terminal branches between the ITS and trnL-F trees: for instance, K. lindsayi (315) and K. lindsayi (352) are sister clades on a well-supported branch in the ITS tree but separated on other well-supported branches in the trnL-F tree (Figs. 4, 5). Incongruences such as these may be evidence of chloroplast capture or gene flow (Wolfe and Elisens, 1995
), which in Korthalsella is likely to be seed mediated.
Korthalsella geminata, an Indonesian-Malaysian taxon missing from our molecular data, is almost certain either to form a clade sister to the rest of the genus or to group with K. papuana on clade A. This conclusion is based on morphological analysis by Molvray (1990)
, the results of which have been largely confirmed by our molecular studies. The other unsampled taxon, K. dacrydii from Malaysia and Indonesia, will most likely be sister to other Asian populations and thus represent a highly derived terminal clade on a lineage generally showing reduced size and number of vascular bundles. The accuracy of the overall tree topology is likely to be unaffected due to the large amount of data supporting the tree, which should compensate for the missing elements. Thus, biogeographic and other implications discussed below are likely to remain valid.
It is possible that ITS sequences appear congruent due to paralogy (Buckler and Holtsford, 1995
). However, all sequences studied using the computer program RNAdraw (Matzura, 1995
) have similar secondary structure stability, which indicates that all are similarly constrained and therefore comparable. It is possible that skewed nucleotide composition in trnL-F might result in differential effects among lineages, but the generally high levels of support for the same topology in trees based on different regions indicate to us that such biases are not a factor.
The great variation in branch shape among closely related plants on different hosts raises the question whether there may be host-mediated changes in morphology of the parasite. Especially striking examples can be found in the Hawaiian terminal clades, in which plants with long thin internodes (e.g., "K. platycaula") are most closely related to plants with short and broad internodes (e.g., "K. latissima"; Fig. 6). An interesting area for further research would be to look at the degree to which the host itself may be influencing the shape of the parasite. If there is a host effect, the use of morphological characters subject to that effect may be suspect not only in Korthalsella but in any parasite similarly affected.
Another potentially significant trend is the degree of differentiation that may develop due to adaptation to a given host. Some K. taenioides found in Hawaii growing on Metrosideros (e.g., "K. cylindrica" 303, 310 in Figs. 4 and 5) have unusually high levels of both molecular and isozyme divergence from their conspecifics, and cluster together despite their geographic separation on different islands of the Hawaiian archipelago. Possibly, the physiological adaptations required to grow on Metrosideros are such that a host race is forming.
The Korthalsella taenioides clade (C in Figs. 4–6) illustrates the misconceptions that can arise when characters are not analyzed for their phylogenetic significance. When faced with a high level of variation in one character in an otherwise difficult group, the first tendency is to try to recognize that variation. If this is a phylogenetically uninformative character, it will lead only to incorrect assessments as with branch shape in Korthalsella. Another implication of general relevance is that continuous characters, such as shape and size, need quantitative evaluation (Stevens, 1991
). Without it, spurious groupings may be recognized, which either contain at best no information or at worst misleading patterns.
Morphological character analysis is also necessary at the sectional level. Danser's section Bifaria, composed of distichously branching plants, is polyphyletic, as are his sections Korthalsella, with decussate branching, and Heterixia, with differentiated inflorescence branches. The molecular tree indicates that, in addition to clade C (Figs. 3–6) with undifferentiated branches, there are two sections with differentiated branches, one containing K. papuana (clade A, Figs. 4–6), and one that could continue to be called Heterixia (clade B, Figs. 4–6), containing K. salicornioides and K. lindsayi. Morphologically, Heterixia in the molecular sense appears paradoxical since K. salicornioides does not obviously have differentiated branches. However, the branch morphology of K. salicornioides has been misinterpreted. Korthalsella salicornioides bears inflorescences only on distal nodes, as does its sister species, K. lindsayi. Korthalsella salicornioides must, in effect, be viewed as a plant with differentiated inflorescence branches in which vegetative branches have secondarily acquired a terete shape. In contrast, plants with truly undifferentiated branches have flowers on almost every node. The terete vegetative branches of K. salicornioides, seemingly different from the flattened obovate segments found in K. lindsayi, have led to the mistaken assumption of a fundamental difference in inflorescence development between these two species. Instead, branch shape is highly plastic in the genus, and this is simply another instance in which different branch morphologies have no bearing on phylogenetic relationship.
The character evolution of differentiated and undifferentiated inflorescence branches Molvray postulated (1990) is also confirmed by the molecular data. Danser considered the differentiated state to be advanced and called it "specialized branches," whereas Molvray postulated that differentiation is a symplesiomorphy because of its association with the primitive condition of floral triads found in the extremely rare K. geminata, a species known only from three collections in Sabah and one collection from Sumatra. Ginalloa, found in Malesia and sister to Korthalsella (Fig. 1), has triads of flowers at the nodes of branch segments, with flowering tending to occur on the more distal segments. Although the mature plants are different in appearance, the position and arrangement of flowers in Ginalloa are similar to that found in Korthalsella, and Korthalsella-type inflorescence cushions could have evolved from such an ancestor by reduction of inflorescence-bearing branches. Korthalsella geminata has the longest inflorescence branches in the genus, with several tens of nodes on each branch. A compression of nodes during development would result in the formation of inflorescences with multiple flowers per node as is found in K. papuana, K. lindsayi, and K. salicornioides, which also have noticeably fewer nodes per flowering branch. Complete compression of all flowering nodes results in plants with flowers confined to inflorescence cushions at the nodes. Nodal anatomy is complex (Touw, 1984;
Molvray, 1990
), supporting the concept that the nodes are derived from larger structures.
The molecular and the morphological data both have implications for the biogeography of Korthalsella. The most closely related genus, Ginalloa, is found in Malesia, the species that is sister to the rest of the genus comes from the same area, and the levels of both sectional and species diversity are highest there. It is logical to hypothesize that dispersal proceeded from this core area outward. We speculate that the dispersal agents responsible for the patchy, disjunct distribution found in the genus are birds and possibly butterflies in some restricted locations. The southern hemisphere distribution has also been explained by vicariance events following the breakup of Gondwana (Barlow, 1981
; Barlow, personal communication, CSIRO), but the hypothesis appears superfluous given that Korthalsella has clearly dispersed across oceanic distances to islands no more than five million years old. Previous taxonomy of Korthalsella (Danser, 1937,
1940
) had the highest diversity, both of species and sections, in Hawaii. High diversity at the sectional level suggests the genus had a longer history in Hawaii than elsewhere, which is biogeographically improbable. The previous alignment of Korthalsella caused species found on different continents to be included in the same section, whereas some species on one island were put in different sections. Although not impossible, such a situation also seems improbable.
Korthalsella is a good model system to explore several important topics in evolutionary biology. It provides an example of the phylogenetic resolving power of combining independent data sets and especially of a comparative molecular/morphological approach. The interaction among ecology, physiology, and systematics is exceptionally clear in the genus, especially in potentially host-influenced branch shape and size variation. Most importantly, Korthalsella exemplifies the interdependence between critical character analysis and species delimitation, and ultimately phylogeny itself, which rests upon the foundation provided by our understanding of characters and species.
|
FOOTNOTES |
|---|
2 Author for correspondence, current address: Dept. of Botany and Microbiology, Oklahoma University, Norman, OK 73019 (mmolvray{at}ou.edu
; fax: +405 325 7619).
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Baldwin, B. G. 1993 Molecular phylogenetics of Calycadenia (Compositae) based on ITS sequences of nuclear ribosomal DNA: chromosomal and morphological evolution reexamined. American Journal of Botany. 80: 222–238.
———. M. J. Sanderson, J. M. Porter, M. F. Wojciechowski, C. S. Campbell, and M. J. Donoghue. 1995 The ITS region of nuclear ribosomal DNA: a valuable source of evidence on angiosperm phylogeny. Annals of the Missouri Botanical Garden 82: 247–277.[CrossRef][ISI]
Barlow, B. A. 1983 A revision of the Viscaceae of Australia. Brunonia 6: 25–57.
———. 1981 The Australian flora: its origin and evolution. Flora of Australia 1: 25–75.
Bogler, D. J., and B. B. Simpson. 1996 Phylogeny of Agavaceae based on ITS DNA sequence variation. American Journal of Botany 83: 1225–1235.[CrossRef][ISI]
Buckler, E., and T. P. Holtsford. 1995 Zea ribosomal repeat evolution and substitution patterns. Molecular Biology and Evolution 13: 623–632.[Abstract]
Cox, A. 1997 PaupGap. Computer program available from A. Cox, Jodrell Laboratory, Royal Botanic Gardens Kew, London.
Danser, B. H. 1937 A revision of the genus Korthalsella. Bulletin du Jardin Botanique Buitenzorg 14(2): 115–159.
———. 1940 A Supplement to the revision of the genus Korthalsella. Bulletin du Jardin Botanique Buitenzorg 16: 329–342.
Doyle, J. J., and J. L. Doyle. 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemistry Bulletin 19: 11–15.
Engler, A. 1897 In Engler and Prantl, Die Natürliche Pflanzenfamilien, Nachträge II-IV. p. 138.
Farris, J. S. 1969 A successive approximations approach to character weighting. Systematic Zoology 18: 374–385.[CrossRef]
———, V. A. Albert, M. Källersjö, D. Lipscomb, and A. G. Kluge. 1996 Parsimony jackknifing outperforms neighbor-joining. Cladistics 12: 99–124.[CrossRef][ISI]
Felsenstein, J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791.[CrossRef][ISI]
Fitch, W. M. 1971 Toward defining the course of evolution: minimum change for a specific tree topology. Systematic Zoology 20: 406–416.[CrossRef][ISI]
Freemann, C. E. and G. R. Ybarra. 1996 The utility of chloroplast trnL intron DNA sequences in inferring phylogeny within Keckiella (Scorphulariaceae). American Journal of Botany (supplement) 83: 590 (Abstract).
Gielly, L., and P. Taberlet. 1994 The use of chloroplast DNA to resolve plant phylogenies: noncoding versus rbcL sequences. Molecular Biology and Evolution 11: 769–777.[Abstract]
———. 1996 A phylogeny of the European gentians inferred from chloroplast trnL (UAA) intron sequences. Botanical Journal of the Linnaean Society 120: 57–75.[CrossRef]
———. 1996 Chloroplast DNA sequencing to resolve plant phylogenies between closely related taxa. In T. B. Smith and R. K. Wayne [eds.], Molecular genetic approaches in conservation, 143–153. Oxford University Press, Oxford.
Hahn, W. J., and K. J. Sytsma. 1993 Nuclear ribosomal internal transcribed spacer phylogenetics of the genus Clarkia (Onagraceae). American Journal of Botany (Supplement) 80: 152 (Abstract).
Hedin, M. C. 1997 Molecular phylogenetics at the population/species interface in cave spiders of the southern Appalachians (Araneae: Nesticidae: Nesticus). Molecular Biology and Evolution 14: 309–324.[Abstract]
Lecomte, P. H. 1916 Le genre Korthalsella. Bulletin du Museum d'Histoire Naturelle de Paris 22: 260–267.
Matzura, O. 1995 RNAdraw v. 1.1. Computer program available from the author, ole@mango.mef.ki.se, Karolinska Institut, Stockholm, Sweden.
Merrill, E. D. 1916 Korthalsella, Bifaria and Pseudixus. Botanical Magazine Tokyo, 30: 66–68.
Molvray, M. 1990 A revision of Korthalsella (Viscaceae). Ph.D. dissertation, Tulane University, New Orleans, LA.
———. 1997 A synopsis of Korthalsella (Viscaceae). Novon 7: 268–273.[CrossRef][ISI]
Mullis, K. B., and F. A. Faloona. 1987 Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Methods in Enzymology 155: 335–350.[ISI][Medline]
Nickrent, D. L., K. P. Schuette, and E. M. Starr. 1994 A molecular phylogeny of Arceuthobium (Viscaceae) based on nuclear ribosomal DNA internal transcribed spacer sequences. American Journal of Botany. 81: 1149–1160.
O'Hara, R. J. 1993 Systematic generalization, historical fate, and the species problem. Systematic Biology 42: 231–246.[CrossRef][ISI]
Olmstead, R. G. 1995 Species concepts and plesiomorphic species. Systematic Botany 20: 623–630.[CrossRef][ISI]
———, B. Bremer, K. M. Scott, and J. D. Palmer. 1993 A molecular systematic analysis of the Asteridae sensu lato based on rbcL sequences. Annals of the Missouri Botanical Garden 80: 700–722.[CrossRef][ISI]
Smith, A. C. 1985 Flora Vitiensis Nova Vol. 3, p. 747. Pacific Tropical Botanical Garden: Lawai, Kauai, HI.
Stevens, P. F. 1991 Character states, morphological variation, and phylogenetic analysis: a review. Systematic Botany 16: 553–583.[CrossRef][ISI]
Swofford, D. L. 1993 PAUP: phylogenetic analysis using parsimony. Version 3.1.1. Illinois Natural History Survey, Champaign.
Taberlet, P. T., L. Gielly, G. Pautou, and J. Bouvet. 1991 Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105–1109.[CrossRef][ISI][Medline]
Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1995 CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position, specific gap penalties and weight matrix choice. Nucleic Acids Research 22: 4673–4680.
Touw, M. (Now Molvray). 1984 Preliminary observations on Korthalsella (Viscaceae) with special reference to vascular patterns. Blumea 29: 525–545.[ISI]
Van Tieghem, P. 1896 Sur le groupement des éspèces en genres dans les Ginalloées, Bifariées, Phoradendrées et Viscées, quatre tribus de la famille des Loranthacées. Bulletin du Societé Botanique de France 43: 161–194.
White, T. L., T. Bruns, S. Lee, and J. Taylor. 1991 Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In M. A. Innis, D. H. Gelfand, J. J. Sninsky, and T. J. White [eds.], PCR protocols, 315–322. Academic Press, San Diego, CA.
Wolfe, A. D., and W. J. Elisens. 1995 Evidence of chloroplast capture and pollen-mediated gene flow in Penstemon sect. Peltanthera (Scrophulariaceae). Systematic Botany 20: 395–412.[CrossRef][ISI]
This article has been cited by other articles:
|
|
 |
|
  D. T. Harbaugh and B. G. Baldwin Phylogeny and biogeography of the sandalwoods (Santalum, Santalaceae): repeated dispersals throughout the Pacific Am. J. Botany, June 1, 2007; 94(6): 1028 - 1040. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Nickrent, M. A. Garcia, M. P. Martin, and R. L. Mathiasen A phylogeny of all species of Arceuthobium (Viscaceae) using nuclear and chloroplast DNA sequences Am. J. Botany, January 1, 2004; 91(1): 125 - 138. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. D. Hoggard, P. J. Kores, M. Molvray, and R. K. Hoggard The phylogeny of Gaura (Onagraceae) based on ITS, ETS, and trnL-F sequence data Am. J. Botany, January 1, 2004; 91(1): 139 - 148. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. C. Muschner, A. P. Lorenz, A. C. Cervi, S. L. Bonatto, T. T. Souza-Chies, F. M. Salzano, and L. B. Freitas A first molecular phylogenetic analysis of Passiflora (Passifloraceae) Am. J. Botany, August 1, 2003; 90(8): 1229 - 1238. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. T. Bakker, A. Culham, R. Gomez-Martinez*, J. Carvalho, J. Compton, R. Dawtrey, and M. Gibby Patterns of Nucleotide Substitution in Angiosperm cpDNA trnL (UAA)-trnF (GAA) Regions Mol. Biol. Evol., August 1, 2000; 17(8): 1146 - 1155. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
