| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2School of Biological Sciences, Washington State University, Pullman, Washington 99164-4236 USA; 3Botanical Gardens, University of Tokyo, Hausan, Bunkyo-ku, Tokyo 112-0001, Japan; 4Department of Biological Sciences, Idaho State University, Pocatello, Idaho 83209-8007 USA; 5Department of Botany, Kyoto University, Sakyo-ku, Kyoto-shi, Kyoto 606-8502, Japan; and 6Makino Herbarium, Faculty of Science, Tokyo Metropolitan University, Minami-Osawa, Hachioji-shi, Tokyo 192-0397, Japan
Received for publication March 7, 2000. Accepted for publication July 7, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: chromosomal evolution • Chrysosplenium • gynoecial diversification • matK sequences • phytogeography • Saxifragaceae
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Franchet, 1890
; Hara, 1957
; Packer, 1963
; Spongberg, 1972
). The genus is primarily restricted to the Northern Hemisphere with species occurring in eastern North America (two species), western North America (four species), Europe (two species), and eastern Asia, where the greatest number of species are present; for example, 27 species are restricted to the Sino-Himalayan region, and 10 are native to Japan (Spongberg, 1972
). Two species of the genus occur disjunctly in South America. Thus, Chrysosplenium exemplifies, in large part, a well-known floristic disjunction involving eastern and western North America, eastern Asia, and Europe (e.g., Gray, 1846
, 1859; Hu, 1935
; Chaney, 1947
; Li, 1952, 1972
; Graham, 1972
; Hara, 1972
; Wolfe, 1972, 1975, 1981, 1985
; Wood, 1972
; Raven and Axelrod, 1974
; Boufford and Spongberg, 1983
; Wu, 1983
; Tiffney, 1985a, b
; Boufford, 1992
; Hong, 1993
; Guo, Richlefs, and Cody, 1998
; Guo, 1999
). Based on morphological data and his views of primitive and derived character states, Hara proposed that Chrysosplenium arose in South America with subsequent migration to North America and the Old World. Recent molecular phylogenetic analyses have provided important biogeographical insights into taxa exhibiting this pattern of disjunction (e.g., Aesculus; Xiang et al., 1998
), as well as for taxa occurring only in the first three of these areas (e.g., Xiang, Soltis, and Soltis, 1998
).
Relationships within Chrysosplenium are problematic. The genus was divided into two subgenera (Gamosplenium and Dialysplenium) and several series by Maximowicz (1877)
. Franchet (1890)
divided the genus into two sections based on the presence of opposite vs. alternate leaves—sect. Oppositifolia and sect. Alternifolia. Hara (1957)
, in contrast, decided that the two sections recognized by Franchet were unnatural; he divided Chrysosplenium into 17 series.
Several broad phylogenetic studies of Saxifragaceae that included sequences of the chloroplast gene matK for species of Chrysosplenium suggested that matK sequence data would be useful for resolving relationships within the large, problematic genus Chrysosplenium (Johnson and Soltis, 1994, 1995
; Soltis et al., 1996
). More recently, Nakazawa et al. (1997)
used matK sequences to resolve species relationships in a study focused on Japanese members of Chrysosplenium.
We constructed a matK sequence data set for Chrysosplenium and included species from throughout the geographic range of the genus and representing 13 of the 17 series recognized by Hara (1957)
. Our study had four basic objectives: (1) to provide an initial phylogenetic hypothesis of species relationships for this enigmatic genus and begin to evaluate the monophyly of Hara's series; (2) to determine whether the genus consists of two major lineages corresponding to sections Alternifolia and Oppositifolia; (3) to assess biogeographic patterns among species of Chrysosplenium, with particular interest in Hara's proposed South American origin for the genus and in those species occurring in eastern and western North America, eastern Asia, and Europe; and (4) to examine the evolution of ovary position and chromosome number in light of the resultant phylogenetic hypothesis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
and Doyle and Doyle (1987)
as modified by Soltis et al. (1991)
or by a microprep procedure (Cullings, 1992
) that requires even smaller amounts (<0.1 g) of leaf tissue. The microprep procedure was used primarily to isolate DNA from herbarium specimens, which played a critical role as a source of DNA for several species.
Amplification of matK followed methods described earlier (e.g., Johnson and Soltis, 1994, 1995
). For manual sequencing, preparation of single-stranded DNAs and dideoxy sequencing followed Johnson and Soltis (1994, 1995)
; preparation of samples for automated sequencing followed Mort et al. (in press). For some taxa we sequenced 1080 bp (base pairs) per taxon representing over two-thirds of the gene beginning at the 5' end; for other taxa 898 bp of sequence data were obtained (following Nakazawa et al., 1997
). Several different PCR (polymerase chain reaction) primer combinations were used to amplify matK. The most commonly used combinations were: trnK-3914F and trnK-2R; trnK-253F and matK-2000R; trnK-710F and matK-2000R; matK-934 and matK-MR; matK-1412F and matK-1848R; matK-XF and matK-2200R. The sequencing primers used were: trnK-710JF, matK-1168, matK-1253R, matK-1470R, and matK-1412F. The sequences of all primers were provided previously (Johnson and Soltis, 1994, 1995
; Soltis et al., 1996
; Nakazawa et al., 1997
).
Earlier phylogenetic analyses of Saxifragaceae indicated that the sister group of Chrysosplenium is Peltoboykinia. This relationship receives high bootstrap support and is revealed by the analysis of rbcL, matK, and ITS sequences (Soltis et al., 1993, 1996
; Johnson and Soltis, 1994, 1995, 1998
). Peltoboykinia was therefore used as the outgroup in our analyses. The relationships of the Peltoboykinia/Chrysosplenium clade within Saxifragaceae are unclear, however, with as many as seven clades possible close relatives; hence, in this study no other outgroups other than Peltoboykinia were employed.
Phylogenetic analyses
The matK sequences were easily aligned by eye and varied in length due to the presence of several short insertions and deletions (occurring, as expected, in multiples of three base pairs) ranging in length from 3 to 15 bp (Table 2). The aligned data set is available at http://www.wsu.edu:8080/
soltilab/. Phylogenetic analyses were conducted using both parsimony and maximum likelihood methods. We constructed a data set and included all of the aligned base pairs with indels coded as missing (we then considered the phylogenetic distribution of each indel by mapping its occurrence on trees derived from analysis of base substitutions alone).
|
) using the heuristic search option, saving all minimal length trees (MULPARS on), tree bisection-reconnection (TBR) branch-swapping, and 100 replicates with random taxon addition; characters were equally weighted, and character states were specified as unordered. Bootstrap (Felsenstein, 1985
) and decay analyses (Bremer, 1988
; Donoghue et al., 1992
) were then conducted to obtain estimates of reliability for monophyletic groups. For the bootstrap analysis (1000 replicates), simple taxon addition and TBR branch swapping, with characters equally weighted, were used. The decay analysis was conducted following the method of Eernisse and Kluge (1993
; e.g., Xiang, Soltis, and Soltis, 1998
; Xiang et al., 1998
). The method involves examining each node of interest in turn using a constraint statement that specifies only the node of interest being monophyletic and saving the shortest trees that do not satisfy this criterion. The difference between the length of these trees and the shortest trees is used as the decay value for that node.
We conducted two maximum likelihood (MLE) analyses with PAUP* version 4.0 using heuristic search, TBR branch-swapping, and the Hasegawa-Kishino-Yano model (1985)
with different settings for ratio of transitions (ti)/transversions (tv), base frequencies, and rate heterogeneity among sites. The first analysis used a ti/tv ratio of 2.0 and empirical base frequencies and assumed equal rate of evolution for all sites. Trees resulting from this analysis were then used to calculate base frequencies, ti/tv ratio, and the shape parameter of the gamma distribution to estimate rate heterogeneity among sites. A second analysis with the latter values was then conducted.
Character evolution
The phylogenetic distribution of chromosome number, ovary position, and geographic distribution was investigated. Chromosome numbers were obtained from Fedorov (1969)
; ovary positions were based on Hara (1957)
and personal observations. Using MacClade 3.05 (Maddison and Maddison, 1992
), data were traced onto the strict consensus of shortest trees obtained (see below). To gain insights into the geographic history of Chrysosplenium from a phylogenetic perspective, we inferred the ancestral areas of clades recognized in the matK tree using MacClade following the general approach of Xiang, Soltis, and Soltis (1998)
and Xiang et al. (1998)
. As noted, Peltoboykinia is the sister group of Chrysosplenium and was therefore used as the outgroup. However, subsequent sister taxa to Chrysosplenium/Peltoboykinia are uncertain (Soltis et al., 1993, 1996
). This limits our ability to reconstruct unambiguously the ancestral state for characters in Chrysosplenium (Maddison, Donoghue, and Maddison, 1994).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), each of 371 steps, CI (consistency index) = 0.6852 excluding uninformative characters, RI (retention index) = 0.8509. One of these 18 shortest trees, picked at random, is depicted in Fig. 1 with the nodes that collapse in the strict consensus indicated. As noted in more detail below, the distribution of indels lends further support to relationships suggested by base substitutions.
|
All shortest trees recognize two large sister clades corresponding to the two groups of Franchet (1890)
: sect. Oppositifolia, with bootstrap and decay values of 92% and 5, respectively, and a poorly supported sect. Alternifolia (44%, 2). Within the Oppositifolia clade, C. pseudofauriei, C. grayanum (both from eastern Asia), C. oppositifolia (from Europe), C. americanum (eastern North American), and C. glechomaefolium (western North America) form a weakly supported clade (62%, 2). A second clade (72%, 3) within Oppositifolia contains C. delavayi (China and Taiwan) as sister to C. valdivicum (Chile) and C. ramosum (northeastern China). The remaining members of the Oppositifolia clade, all from eastern Asia, form a moderately supported (78%, 3) subclade, within which C. kamtschaticum is sister to the remaining members; they, in turn, form two subclades. The first is a well-supported (100%, 9) Macrostemon clade of C. fauriae, C. echinus, C. kiotense, C. nagasei, and C. macrostemon, all of series Macrostemon and occurring in Japan, and the second is a Pilosa clade (without support >50%) of C. maximowiczii, C. pilosum, C. rhabdospermum, and C. album, all of series Pilosa and occurring primarily in Japan.
The Alternifolia clade comprises two subclades, one of which is strongly supported (100%, 20) and contains C. carnosum, C. griffithii, C. davidianum, and C. henryi, all from eastern Asia. The second major subclade within Alternifolia does not receive bootstrap support >50%. It comprises C. flagelliferum and C. tosaense from eastern Asia as the well-supported (100%, 11) sister group to a clade (94%, 4) that includes C. japonicum (eastern Asia) as the sister to a clade of C. alternifolium and C. tetrandrum (both circumboreal).
Multiple populations of five species were sampled. In C. album, C. pilosum, and C. flagelliferum, the two populations sampled are sisters. However, the two populations sampled of C. fauriae are not immediate sisters, but rather successive, basal branches of the Macrostemon clade. Similarly, the populations of C. alternifolium sampled form a clade only with the inclusion of C. tetrandrum.
Some of the indels detected provide additional support for clades. Indels I-1, I-2, I-6, I-8, D-1, D-3, and D-5 are unique to individual taxa and will not be discussed further (Table 2). Other indels further support relationships indicated by base substitutions; these findings for Chrysosplenium further support a trend noted in other analyses employing matK sequence data—indels in matK are often phylogenetically informative (reviewed in Soltis and Soltis, 1998). Two insertions, I-4 and I-9, both unite C. album with C. rhabdospermum. The deletion D-2 is an additional synapomorphy for all members of the Macrostemon subclade, with the exception of the first-branching member, C. fauriae (Soltis collection); thus, D-2 unites the subclade consisting of C. echinus, C. fauriae (Wakabayashi & Nakazawa collection), C. nagasei, C. kiotense, and C. macrostemon (Table 2; Fig. 1). This same deletion provides additional evidence for the distinctness of the two populations of C. fauriae sampled. More populations of C. fauriae should be carefully studied; these two populations should perhaps ultimately be treated as distinct species. Insertion I-5 appears to have arisen twice independently, with each occurrence representing a synapomorphy for a distinct clade: (1) C. alternifolia, C. tetrandrum, and C. japonicum; (2) C. henryi, C. davidianum, C. griffithii, and C. carnosum (Table 2; Fig. 1). Insertion I-3 is present in C. oppositifolium and C. glechomaefolium, but not C. americanum (the sister of C. glechomaefolium). Hence, this insertion either arose independently in C. oppositifolium and C. glechomaefolium or arose in the ancestor of this clade of three species with a subsequent loss in C. americanum. Insertion I-7 occurs in two distantly related members of the Oppositifolia clade (C. grayanum and C. kiotense) and thus appears to be homoplasious. It is noteworthy that three of the nine bases involved in the insertion differ between the two species, possibly reflecting the independent nature of this insertion in these taxa. In addition, a deletion (D-5) occurs at the beginning of this same region in C. kiotense, suggesting that the history of insertion and deletion for this area may be more complex in this species than in C. grayanum.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. Phylogenetic analyses of matK sequence data for 33 collections representing much of the diversity of Chrysosplenium support the division of the genus into two major clades, one with opposite leaves and one with alternate leaves. Thus, sections Oppositifolia and Alternifolia reflect phylogenetic relationships, and leaf phyllotaxy is a diagnostic morphological character. Hara also recognized 17 series in Chrysosplenium, many of which contain only one or a few species; two of the series he recognized appear monophyletic in our analyses. Series Macrostemon, consisting of C. echinus, C. fauriae, C. kiotense, C. macrostemon, and C. nagasei, all from Japan, receives strong support (100%, 9) as a monophyletic group. Series Pilosa, represented in our study by C. maximowiczii, C. rhabdospermum, C. pilosum, and C. album also is monophyletic in all phylogenetic analyses, albeit without support >50%. The Pilosa subclade is also primarily centered in Japan, with C. pilosum also known from Korea. However, in contrast to our findings for series Macrostemon and Pilosa, matK sequence data do not provide support for the monophyly of Hara's series Oppositifolia, Nepalensia, or Alternifolia. Of the three species sampled of series Oppositifolia (Table 1), only C. americanum and C. oppositifolium are closely related; C. ramosum appears as a member of a distinct clade. Chrysosplenium glechomaefolium and C. grayanum of series Nepalensia are both members of the same subclade, but are not sister taxa. Instead, C. glechomaefolium, from western North America, is sister (with strong support) to C. americanum (series Oppositifolia ) from eastern North America. Series Alternifolia appears grossly polyphyletic with species spread among the three subclades present within the Alternifolia clade (Fig. 1).
Chromosome evolution
The most frequently reported haploid chromosome numbers for Chrysosplenium are n = 11 and n = 12, but several species have n = 4, n = 8, n = 9, and n = 21 (Fedorov, 1969
; Spongberg, 1972
). For example, C. delavayi has n = 4; C. henryi has n = 8; C. glechomaefolium, C. griffithii, and C. davidianum have n = 9; and C. oppositifolium has n = 21. These are unusual numbers for Saxifragaceae, a family in which most genera have n = 7. For the two most common numbers in Chrysosplenium (n = 11, 12), these numbers are known elsewhere only in some species of Saxifraga (Webb and Gornall, 1989
), with n = 11 also reported for Peltoboykinia, the sister group of Chrysosplenium. Given that n = 11 in Peltoboykinia, it is likely that the original base chromosome number for Chrysosplenium was also n = 11 (Fig. 2). Reconstructions of the evolution of chromosome numbers in Chrysosplenium (using MacClade version 3.05; Maddison and Maddison, 1992
) suggest that n = 12 arose independently several times. For example, if n = 11 is ancestral, considering first the large Oppositifolia clade, n = 12 has evolved independently in the Pilosa subclade, C. ramosum or the ancestor of the small clade to which it belongs, and either in the ancestor of the C. americanum/C. glechomaefolium/C. oppositifolium/C. pseudofauriei clade (with subsequent chromosomal evolution in C. oppositifolium and C. glechomaefolium), or perhaps multiple times just within this clade (Fig. 2). For the large Alternifolia clade, n = 12 characterizes the large subclade containing C. tosaense, C. flagelliferium, C. japonicum, C. tetrandrum, and C. alternifolium.
|
). An example of chromosome increase, probably involving polyploidy, is observed in C. oppositifolium, with n = 21. The first two branches of the clade to which C. oppositifolium belongs are C. grayanum and C. pseudofauriei, with n = 11 and n = 12, respectively. A likely scenario is a polyploid event, perhaps involving ancestor(s) with n = 11 leading to a polyploid with n = 22; this would be followed by aneuploid decrease to n = 21.
Ovary position
Ovary position has been proposed to evolve in a unidirectional manner throughout the angiosperms from superior to greater inferiority, generally via congenital fusion of the hypanthium to the ovary wall (e.g., Eames, 1931, 1961
; Langdon, 1939
; Douglas, 1944, 1957
; Gauthier, 1950
; reviewed in Kuzoff, Hufford, and Soltis, 2001
). This model of unidirectional ovary evolution has been invoked to account for the remarkable range of ovary positions seen in some genera of Saxifragaceae, for example Saxifraga (Stebbins, 1974
) and Lithophragma (Taylor, 1965
). As is the case for Lithophragma, Saxifraga, and other genera and small clades within Saxifragaceae, the ovary position reported for species of Chrysosplenium varies widely. Ovary position has been reported as superior for C. album (a superior ovary has also been reported for C. ludlowii, but this species is known only from the type collection), with half-inferior and inferior ovaries present in the remaining taxa we sampled (Fig. 3). Peltoboykinia, the sister of Chrysosplenium, has an ovary that is half-inferior. The ancestor of Chrysosplenium is reconstructed as having an ovary that is either half-inferior or completely inferior. Considering first the opposite-leaved clade, the ancestor of this clade also had an ovary position that was either half or fully inferior. The common ancestor of the Pilosa/Macrostemon/C. kamtschaticum clade is reconstructed as having an inferior ovary. An inferior ovary characterizes the Macrostemon clade, with the evolution of a half-inferior ovary occurring in one member, C. macrostemon. Both inferior and half-inferior ovaries also evolved within the C. americanum/C. glechomaefolium/C. oppositifolium/C. grayanum/C. pseudofauriei clade, although the lack of resolution within this clade precludes an accurate interpretation of the transitions that occurred. A half-inferior ovary characterizes the Pilosa clade. It is noteworthy that C. album, with what has been termed a superior ovary, is derived within this clade from an ancestor with a half-inferior ovary.
|
Thus, MacClade reconstructions of character states indicate that the evolution of ovary position in Chrysosplenium has been dynamic, with several shifts between completely inferior and half-inferior ovaries (Fig. 3). It is particularly noteworthy that the superior ovary of C. album is derived from a half-inferior ancestor. The labile nature of gynoecial diversification in Chrysosplenium is in agreement with similar results based on analyses of other well-defined genera within Saxifragaceae in which substantial variation in ovary position has been observed (e.g., Kuzoff et al., 1999
; Mort and Soltis, 1999
). For example, ovary position in Lithophragma ranges from inferior to what appears to be completely superior, with a wide range of intermediate conditions. Kuzoff et al. (1999)
demonstrated that: (1) the ancestor of Lithophragma had an ovary that was half-inferior; (2) ovary position has evolved toward greater inferiority in some species (e.g., L. affine, L. parviflorum, and L. trifoliatum) and greater superiority in others (e.g., L. glabrum, L. heterophyllum, and L. campanulatum); and (3) variation in ovary position in Lithophragma is not the result of a unidirectional trend. Very similar conclusions can now be drawn for Chrysosplenium. That is, the ancestor of Chrysosplenium had an ovary that was half-inferior, and there has been evolution to increasing inferiority in some species and increasing superiority in others. Ontogenetic analyses of the early gynoecial development in the Saxifragaceae indicate that there are no truly superior ovaries in the family (Kuzoff, Hufford, and Soltis, 2001
; Soltis et al., unpublished data). That is, all ovaries appear to develop from what is termed an appendicular epigynous ground plan and hence are technically "inferior." Truly superior ovaries, in contrast, develop from a hypogynous ground plan, and this ground plan has not been observed in the family. Subtle allometric shifts in early ontogeny in the superior vs. inferior region of the ovary appear to be responsible for the diverse array of ovary positions present in Saxifragaceae (Kuzoff, Hufford, and Soltis, in press
). Thus, in Saxifragaceae, ovaries that appear superior are not homologous to truly superior ovaries but are best referred to as "superior mimics" (Kuzoff, Hufford, and Soltis, in press
).
Biogeography
Chrysosplenium exhibits a wide distribution in the arctic and north temperate zones, with two species disjunctly distributed in the Southern Hemisphere. Most species of the genus occur in eastern Asia, a few in circumpolar regions, two in eastern North America, and four in western North America.
Hara (1957)
considered the two species from South America, C. valdivicum and C. macranthum, to be ancestral in the genus because they seem to retain features that Hara considered "primitive," such as glabrous and isophyllous stems, opposite leaves, simple inflorescences, short stamens and styles, smooth seeds, and an inferior ovary. The two species from South America are similar morphologically, differing slightly in leaf morphology. We obtained material of C. valdivicum, and our phylogenetic trees indicate clearly that this species is not the first-branching member of Chrysosplenium, but instead occurs as a derived member of the genus in a small clade with C. delavayi and C. ramosum, both of which occur in eastern Asia; C. delavayi appears as the sister to C. valdivicum/C. ramosum. Rather than a South American origin of Chrysosplenium, our phylogenetic reconstructions suggest an Asian origin of the genus, with at least three separate instances of migration to the New World.
Phylogenetic analyses of Chrysosplenium demonstrate that the eastern Asian species are distributed in both major clades (Alternifolia and Oppositifolia), as well as among all of the larger subclades. Our results indicate that the ancestral area of the genus is eastern Asia, suggesting that Chrysosplenium originated in eastern Asia, with subsequent movement to Europe and North and South America (Fig. 4). Our data also suggest that the genus had undergone diversification into the alternate- and opposite-leaved groups in eastern Asia before it spread into other parts of the world. This scenario of an Asian origin is consistent with the hypothesis of Savile (1975)
. Based on relationships among rust parasites (genus Puccinia) that use members of Saxifragaceae as hosts, Savile (1975)
proposed that several genera of Saxifragaceae having disjunct distributions in eastern North America, eastern Asia, and western North America actually originated in eastern Asia with subsequent migration from that area of origin to other regions of the Northern Hemisphere.
|
Two additional migration events out of eastern Asia are also apparent in the Oppositifolia clade. The opposite-leaved lineage diverged into three subclades, one of which is exclusively Asian in distribution (the Pilosa/Macrostemon clade); a second small clade of three species also occurs in Asia, with one species (C. valdivicum) in Chile. The third subclade consists of several species displaying a disjunct distribution in eastern Asia, eastern North America, western North America, and Europe (i.e., C. grayanum and C. pseudofauriei from eastern Asia, C. oppositifolium from Europe, C. americanum from eastern North America, and C. glechomaefolium from western North America) (Fig. 4). Within this clade, the two North American species (C. americanum/C. glechomaefolium) are sisters; their immediate sister is the European species, C. oppositifolium, followed by the two Asian species, C. grayanum and C. pseudofauriei. Thus, the data indicate an additional migration event out of eastern Asia in the Oppositifolia clade. Furthermore, within this subclade the Old World species are the sister taxa to the New World species, with the eastern and western North American taxa appearing as sisters. The phylogenetic pattern within this clade is identical to that found in Aesculus (Hippocastanaceae), which exhibits the same disjunct distribution in eastern Asia, eastern North America, western North America, and Europe (Xiang et al., 1998
). This phylogenetic pattern is also similar to that found for several other genera (e.g., Tiarella, Boykinia, Trautvetteria, Calycanthus, Aralia sect. Aralia, big-bracted species of Cornus) displaying disjunct distributions in eastern Asia, western North America, and eastern North America (Xiang, Soltis, and Soltis, 1998
). In all of these taxa, the North American species form a monophyletic group sister to the Asian species. This pattern suggests a closer relationship between the floras of eastern and western North America than between eastern Asia and eastern North America. This relationship is in contrast to the long-standing view (first proposed by Gray, 1846)
that a closer relationship exists between eastern Asian and eastern North American floras.
The second subclade in the Oppositifolia clade consists of three species showing a disjunct distribution in eastern Asia and the Southern Hemisphere (i.e., C. valdivicum from Chile, C. ramosum from northeastern China, and C. delavayi from eastern Asia; with the first two species being sisters) (Figs. 1, 4). Thus, the data imply still another migration event from Asia to the New World. The occurrence of Chrysosplenium in the southernmost portion of South America represents an intriguing disjunction. This disjunction pattern is rare and has been considered difficult to explain; long-distance dispersal or a relictual pattern resulting from vicariance have both been proposed to explain the disjunction (see Thorne, 1972
, Schnabel and Wendel, 1998
). For example, Qin (1997)
proposed that the disjunct distribution of Lardizabalaceae in eastern Asia and the Southern Hemisphere resulted from the fragmentation of a once worldwide distribution of the family that formed before the separation of Laurasia and Gondwana. In contrast, Thorne (1972)
proposed that Coriaria (Coriariaceae) obtained its distribution in the two hemispheres via long-distance dispersal by birds. Schnabel and Wendel (1998)
argue against vicariance and also favor long-distance dispersal to explain the distribution of Gleditsia (Fabaceae), a genus of 13 species found disjunctly in eastern North America and eastern Asia with a single species from South America. As in Chrysosplenium, a molecular-based phylogenetic tree demonstrated that the South American species of Gleditsia forms a clade with two species from eastern Asia. Sequence divergence between the single South American species and its close relatives in Asia suggested long-distance dispersal (within the last 8 million years) for this disjunction in Gleditsia (Schnabel and Wendel, 1998
).
Thus, long-distance dispersal is also one possible explanation for the disjunction in Chrysosplenium. As argued by Schnabel and Wendel (1998)
for Gleditsia, there is no evidence to suggest that this particular clade (C. valdivicum/C. ramosum/C. delavayi) within the opposite-leaved lineage of Chrysosplenium (a derived lineage within Saxifragaceae; Soltis et al., 1993, 1996
) is an ancient lineage that evolved before the separation of Laurasia and Gondwana. We cannot provide an estimate of divergence times for species of Chrysosplenium occurring in Asia and South America because no absolute substitution rate (or clock) has been estimated for matK and no suitable fossil record exists in Saxifragaceae, or a close relative, for calibration. A comparison of sequence divergence values suggests, however, that the disjunction is probably not of recent origin. The following sequence divergence values were estimated using total substitutions calculated using Jukes-Cantor distance: 0.025 between C. ramosum (eastern Asia) and C. delavayi (southwestern China), 0.025 between C. ramosum and C. valdivicum (Chile), and 0.034 between C. valdivicum and C. delavayi. In the sister clade, sequence divergence is 0.012 between C. americanum (eastern North America) and C. glechomaefolium (western North America), and 0.020 between C. oppositifolium (Europe) and the clade of C. americanum/C. glechomaefolium. Assuming that matK is evolving in a roughly clock-like manner in these two clades, the data suggest that the species in eastern Asia and Chile diverged before the isolation of the North American and European species. Although the data seem to preclude relatively recent long-distance dispersal as a likely explanation for this disjunction, ancient long-distance dispersal remains a likely explanation. The seeds of Chrysosplenium are small, as in other Saxifragaceae, so long-distance dispersal of seeds from Asia to South America is a possibility. However, these plants occur on forest floors, typically in very wet areas, under dense forest canopies, all of which would make long-distance dispersal on such a large geographic scale more difficult.
Alternatively, this disjunct distribution could represent remnants of a once more continuous geographic distribution of this particular subclade of the genus Chrysosplenium, involving migration from Asia down the western Cordillera of North America and South America with subsequent extinction in much of the intervening area. Interrupted migration of plants between North and South America was possible via the West Indies as early as the Oligocene, and direct migration between North and South America became feasible 5.7 million years ago when the union of North and South America occurred (Raven and Axelrod, 1974
). One difficulty with this hypothesis is the absence of members of this C. valdivicum/C. ramosum/C. delavayi alliance in the area between Asia and South America. Suitable habitats seem to be present given that other species of Chrysosplenium (from other subclades) are found throughout the Pacific Northwest of North America, extending as far south as California. Thus, these data would also seem to favor ancient long-distance dispersal.
Chrysosplenium represents one of several genera in Saxifragaceae having a largely Northern Hemisphere distribution with highly disjunct members in South America. For example, the monotypic Hieronymusia is known only from a small area of northern Argentina and southern Bolivia; its putatively closest relatives (species of Suksdorfia) occur in the Pacific Northwest of North America (Gornall and Bohm, 1985
). Species of Saxifraga (in the strict sense; see Soltis et al., 1996
) are similarly widely distributed in North America, Europe, and Asia, with a few disjunct species (typically considered to represent distinct genera, Saxifragoides and Saxifragella; see Engler, 1930
) known from the southern tip of South America. In contrast to our results for Chrysosplenium, however, phylogenetic analyses of matK sequences suggest that Saxifragella, endemic to Tierra del Fuego, actually represents an early branch of the Saxifraga lineage (Soltis et al., unpublished data), with the other early-branching species occurring in Asia and the Pacific Northwest of North America (the relationships of Saxifagoides are less clear). Thus, South American disjuncts in different genera of Saxifragaceae may have different relationships with congeners, and perhaps also different origins. At this point, however, either ancient long-distance dispersal or migration from Asia down the western Cordillera of North America and South America with subsequent extinction in much of the intervening area could explain the disjunctions in Chrysosplenium and Saxifraga. More examples of these disjunctions involving North America and South America and eastern Asia and South America should be critically examined.
|
FOOTNOTES |
|---|
7 Current address: Department of Botany and the Genetics Institute, University of Florida, Gainesville, FL 32611-8526 USA.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
———, and S. A. Spongberg. 1983 Eastern Asian–eastern North American phytogeographical relationships—a history from the time of Linnaeus to the twentieth century. Annals of the Missouri Botanical Garden 70: 423–439
Bremer, K. 1988 The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42: 795–803[CrossRef][ISI]
Chaney, R. W. 1947 Tertiary centers and migration routes. Ecological Monographs 17: 139–148
Cullings, K. W. 1992 Design and testing of a plant-specific PCR primer for ecological and evolutionary studies. Molecular Ecology 1: 233–240
Donoghue, M. J., R. G. Olmstead, J. F. Smith, and J. D. Palmer. 1992 Phylogenetic relationships of Dipsacales based on rbcL sequences. Annals of the Missouri Botanical Garden 79: 333–345
Douglas, G. E. 1944 The inferior ovary I. Botanical Review 10: 125–186
———. 1957 The inferior ovary II. Botanical Review 23: 1–46[ISI]
Doyle, J. J., and J. J. Doyle. 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemistry Bulletin 19: 11–15
Eames, A. J. 1931 The vascular anatomy of the flower with refutation of the theory of carpel polymorphism. American Journal of Botany 18: 147–188[CrossRef][ISI]
———. 1961 Morphology of the angiosperms. McGraw-Hill, New York, New York, USA
Eernisse, D. J., and A. G. Kluge. 1993 Taxonomic congruence versus total evidence, and amniote phylogeny inferred from fossils, molecules and morphology. Molecular Biology and Evolution 10: 1170–1195[Abstract]
Engler, 1930. Saxifragaceae. In A. Engler and K. Prantl [eds.], Die Naturlichen Pflanzenfamilien, 2nd ed., vol. 18a, 74–226. Engelmann, Leipzig, Germany
Fedorov, A. [ed.]. 1969 Chromosome numbers of flowering plants. Academy of Sciences, Leningrad, Russia
Felsenstein, J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791[CrossRef][ISI]
Franchet, A. R. 1890 Monographie du genere Chrysosplenium Tournfort. Nouvelles Archives du Muséum d'histoire naturelle publiées par les professeurs-administrateurs de cet établissement. Troisiéme Série. Tome deuxiéme. (Paris) iii. 2: 87–114
Gauthier, R. 1950 The nature of the inferior ovary in the genus Begonia. Contributions of the Institute of Botany of the University of Montreal 66: 1–93
Gornall, R. J., and B. A. Bohm. 1985 A monograph of Boykinia, Peltoboykinia, Bolandra, and Suksdorfia (Saxifragaceae). Botanical Journal of the Linnean Society 90: 1–71
Graham, A. 1972 Outline of the origin and historical recognition of floristic affinities between Asia and eastern North America. In A. Graham [ed.], Floristics and paleofloristics of Asia and eastern North America, 1–18. Elsevier, Amsterdam, The Netherlands
Gray, A. 1846 Analogy between the flora of Japan and that of the United States. American Journal of Science and Arts, Series 2, 2: 135–136
———. 1859 Diagnostic characters of phanerogamous plants, collected in Japan by Charles Wright, botanist of the U.S. North Pacific Exploring Expedition, with observations upon the relationship of the Japanese flora to that of North America and of other parts of the northern Temperate Zone. Memoirs of the American Academy of Arts 6: 377–453
Guo, Q.-F. 1999 Ecological comparisons between eastern Asia and North America: historical and geographical perspectives. Journal of Biogeography 26: 199–206
———, R. E. Richlefs, and M. L. Cody. 1998 Vascular plant diversity in eastern Asia and North America: historical and ecological explanations. Botanical Journal of the Linnean Society 128: 123–136
Hara, H. 1957 Synopsis of the genus Chrysosplenium L. Journal of the Faculty of Science University of Tokyo Section III Botany 7: 1–90
———. 1972 Patterns of differentiation in flowering plants. In A. Graham [ed.], Floristics and paleofloristics of Asia and eastern North America, 50–60. Elsevier, Amsterdam, The Netherlands
Hasegawa, M., H. Kishino, and T. Yano. 1985 Dating of the human–ape splitting by a molecular clock of mitochondrial Dna. Journal of Molecular Evolution 21: 160–174
Hong, D.-Y. 1993 Eastern Asian–North American disjunctions and their biological significance. Cathaya 5: 1–39
Hu, H. H. 1935 A comparison of the ligneous flora of China and eastern North America. Bulletin Chinese Botanical Society 1: 79–97
Johnson, L. A., and D. E. Soltis. 1994 matK DNA sequences and phylogenetic reconstruction in Saxifragaceae s. str. Systematic Botany 19: 143–156
———, and ———. 1995 Phylogenetic inference in Saxifragaceae sensu stricto and Gilia (Polemoniaceae) using matK sequences. Annals of the Missouri Botanical Garden 82: 149–175[CrossRef][ISI]
———, and ———. 1998 Assessing congruence: empirical examples from molecular data. In D. E. Soltis, P. S. Soltis, and J. J. Doyle [eds.], Molecular systematics of plants, 297–348. Kluwer, Boston, Massachusetts, USA
Kuzoff, R. K., L. Hufford, and D. E. Soltis. 2001 Structural homology and developmental transformations associated with ovary diversification in Lithophragma (Saxifragaceae). American Journal of Botany 88: 196–205
———, D. E. Soltis, L. Hufford, and P. S. Soltis. 1999 Phylogenetic relationships within Lithophragma (Saxifragaceae): hybridization, allopolyploidy, and ovary diversification. Systematic Botany 24: 598–615[CrossRef][ISI]
Langdon, L. D. M. 1939 Ontogenetic and anatomical studies of the flower and fruit of Fagaceae and Juglandaceae. Botanical Gazette 101: 301–327[CrossRef][ISI]
Li, H. L. 1952 Floristic relationships between eastern Asia and eastern North America. Transactions of the American Philosophical Society 42: 371–429
———. 1972 Eastern Asia–eastern North America species-pairs in wide-ranging genera. In A. Graham [ed.], Floristics and paleofloristics of Asia and eastern North America, 65–78. Elsevier, Amsterdam, The Netherlands
Maddison, D. R. 1991 The discovery and importance of multiple islands of most-parsimonious trees. Systematic Zoology 40: 315–328[CrossRef]
Maddison, W. P., M. J. Donoghue, and D. R. Maddison. 1984 Outgroup analysis and parsimony. Systematic Zoology 33: 83–103
———, and D. R. Maddison. 1992 MacClade: analysis of phylogeny and character evolution. Sinauer, Sunderland, Massachusetts, USA
Maximowicz, C. J. 1877 Diagnoses plantarum novarum asiaticarum scripsit. Bulletin de Académie Impériale des Sciences de Saint-Pétersbourg 23: 340–350
Mort, M. E., and D. E. Soltis. 1999 Phylogenetic relationships and the evolution of ovary position in Saxifraga section Micranthes. Systematic Botany 24: 139–147[CrossRef][ISI]
Nakazawa, M., M. Wakabayashi, M. Ono, and J. Murata. 1997 Molecular phylogenetic analysis of Chrysosplenium (Saxifragaceae) in Japan. Journal of Plant Research 110: 265–274[CrossRef][ISI]
Packer, J. G. 1963 The taxonomy of some North American species of Chrysosplenium L., section Alternifolia Franchet. Canadian Journal of Botany 41: 85–103
Qin, H.-N. 1997 A taxonomic revision of the Lardizabalaceae. Cathaya—Annals of the Laboratory of Systematic and Evolutionary Botany, & Herbarium 8–9: 1–214
Raven, P. H., and D. I. Axelrod. 1974 Angiosperm biogeography and past continental movements. Annals of the Missouri Botanical Garden 77: 539–673
Saghai-Maroof, M. A., K. M. Soliman, R. A. Jorgensen, and R. W. Allard. 1984 Ribosomal DNA spacer-length polymorphisms in barley: Mendelian inheritance, chromosomal location, and population dynamics. Proceedings of the National Academy of Sciences, USA 81: 8014–8018
Savile, D. B. O. 1975 Evolution and biogeography of Saxifragaceae with guidance from their rust parasites. Annals of the Missouri Botanical Garden 62: 354–361[CrossRef][ISI]
Schnabel, A., and J. F. Wendel. 1998 Cladistic biogeography of Gleditsia (Leguminosae) based on ndhF and rpl16 chloroplast gene sequences. American Journal of Botany 85: 1753–1765
Soltis, D. E., R. K. Kuzoff, R. Gornall, and K. Ferguson. 1996 matK and rbcL gene sequence data indicate that Saxifraga (Saxifragaceae) is polyphyletic. American Journal of Botany 83: 371–382[CrossRef][ISI]
———, D. R. Morgan, A. Grable, P. S. Soltis, and R. Kuzoff. 1993 Molecular systematics of Saxifragaceae sensu stricto. American Journal of Botany 80: 1056–1081[CrossRef][ISI]
———, P. S. Soltis, T. G. Collier, and M. L. Edgerton. 1991 Chloroplast DNA variation within and among genera of the Heuchera group: evidence for chloroplast capture and paraphyly. American Journal of Botany 78: 1091–1112[CrossRef][ISI]
Spongberg, S. A. 1972 The genera of Saxifragaceae in the southeastern United States. Journal of the Arnold Arboretum 53: 409–494[ISI]
Stebbins, G. L. 1974 Flowering plants: evolution above the species level. Harvard University Press, Cambridge, Massachusetts, USA
Swofford D. L. 1999 PAUP*: phylogenetic analysis using parsimony, version 4.0. Sinauer, Sunderland, Massachusetts, USA
Taylor, R. L. 1965 The genus Lithophragma (Saxifragaceae). University of California Publications in Botany 37
Thorne, R. F. 1972 Major disjunctions in the geographic ranges of seed plants. Quarterly Review of Biology. 47: 365–411[CrossRef]
Tiffney, B. H. 1985a The Eocene North Atlantic land bridge: its importance in Tertiary and modern phytogeography of the Northern Hemisphere. Journal of the Arnold Arboretum 66: 243–273[ISI]
———. 1985b Perspectives on the origin of the floristic similarity between eastern Asia and eastern North America. Journal of the Arnold Arboretum 66: 73–94[ISI]
Webb, D. A., and R. J. Gornall. 1989 A manual of Saxifragas and their cultivation. Timber Press, Portland, Oregon, USA
Wolfe, J. A. 1972 An interpretation of Alaskan Tertiary floras. In A. Graham [ed.], Floristics and paleofloristics of Asia and eastern North America, 201–233. Elsevier, Amsterdam, The Netherlands
———. 1975 Some aspects of plant geography of the Northern Hemisphere during the late Cretaceous and Tertiary. Annals of the Missouri Botanical Garden 62: 264–279[CrossRef][ISI]
———. 1981 Vicariance biogeography of angiosperms in relation to paleobotanical data. In G. Nelson and E. E. Rosen [eds.], Vicariance biogeography, a critique, 413–427. Columbia University Press, New York, New York, USA
———. 1985 Distribution and major vegetation types during the Tertiary. In E. T. Sundquist and W. S. Broecker [eds.], The carbon cycle and the atmospheric CO2: natural variations Archean to present, 357–375, vol. 32. American Geophysical Union, Washington, D.C., USA
Wood, C. E. 1972 Morphology and phytogeography: the classical approach to the study of disjunctions. Annals of the Missouri Botanical Garden 59: 107–124[CrossRef][ISI]
Wu, Z. Y. 1983 On the significance of Pacific intercontinental discontinuity. Annals of the Missouri Botanical Garden 70: 577–590[CrossRef][ISI]
Xiang, Q.-Y., D. J. Crawford, A. D. Wolfe, Y.-C. Tang, and C. W. Dpamphillis. 1998 Origin and biogeography of Aesculus L. (Hippocastanaceae): a molecular phylogenetic perspective. Evolution 52: 988–997[CrossRef][ISI]
Zhengyi, W. 1983 On the significance of Pacific intercontinental discontinuity. Annals of the Missouri Botanical Garden 70: 577–590
This article has been cited by other articles:
|
|
 |
|
  M. E. Mort, J. K. Archibald, C. P. Randle, N. D. Levsen, T. R. O'Leary, K. Topalov, C. M. Wiegand, and D. J. Crawford Inferring phylogeny at low taxonomic levels: utility of rapidly evolving cpDNA and nuclear ITS loci Am. J. Botany, February 1, 2007; 94(2): 173 - 183. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. J. Kenicer, T. Kajita, R. T. Pennington, and J. Murata Systematics and biogeography of Lathyrus (Leguminosae) based on internal transcribed spacer and cpDNA sequence data Am. J. Botany, July 1, 2005; 92(7): 1199 - 1209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. J. Barfuss, R. Samuel, W. Till, and T. F. Stuessy Phylogenetic relationships in subfamily Tillandsioideae (Bromeliaceae) based on DNA sequence data from seven plastid regions Am. J. Botany, February 1, 2005; 92(2): 337 - 351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Samuel, H. Kathriarachchi, P. Hoffmann, M. H. J. Barfuss, K. J. Wurdack, C. C. Davis, and M. W. Chase Molecular phylogenetics of Phyllanthaceae: evidence from plastid MATK and nuclear PHYC sequences Am. J. Botany, January 1, 2005; 92(1): 132 - 141. [Abstract] [Full Text] [PDF]
|
|
|
|
