| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics and Phytogeography |
2Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan 650204, People's Republic of China; 3Department of Botany, Field Museum of Natural History, 1400 S. Lake Shore Drive, Chicago, Illinois 60605-2496 USA; Laboratory of Systematic and Evolutionary Botany, Institute of Botany, Chinese Academy of Sciences, Nanxinchun 20, Xiangshan, Beijing 100093, People's Republic of China; 4Department of Botany, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118-4599 USA
Received for publication March 1, 2004. Accepted for publication December 16, 2004.
ABSTRACT
Kelloggia Torrey ex Bentham (Rubiaceae) consists of two species disjunctly distributed in western North America (K. galioides Torrey) and the western part of eastern Asia (K. chinensis Franch.). The two species exhibit a high level of morphological divergence. To test its monophyly and to infer its biogeographic history, we estimated the phylogeny of Kelloggia and its relatives from sequences of three chloroplast DNA regions (rbcL gene, atpB-rbcL spacer, and rps16 intron). The monophyly of Kelloggia was strongly supported, and it forms a sister relationship with the tribe Rubieae. The divergence time between the two disjunct species of Kelloggia was estimated to be 5.42 ± 2.32 million years ago (mya) using the penalized likelihood method based on rbcL sequence data with fossil calibration. Our result does not support the Madrean-Tethyan hypothesis, which assumes an earlier divergence time of 20–25 mya. Ancestral area analysis, as well as dispersal-vicariance (DIVA) analysis, suggests the Asian origin of Kelloggia and the importance of Eurasia in the diversification of its close relatives in the Rubieae-Theligoneae-Paederieae group. The intercontinental disjunction in Kelloggia is suggested to have evolved via long-distance dispersal from Asia into western North America.
Key Words: disjunction • eastern Asia • Kelloggia • monophyly • Rubiaceae • western North America
The disjunct distribution of plants between eastern Asia and North America has attracted much attention from botanists, zoologists, mycologists, and biogeographers since the time of Linnaeus (Boufford and Spongberg, 1983
; Wen, 1999
; Mueller et al., 2001
; Sanmartín et al., 2001
). Most studies have focused on the eastern Asian and eastern North American subset of this disjunct pattern (Wen, 1999
). Another common pattern between the two continents is the eastern Asian and western North American disjunction (Li, 1952
; Thorne, 1972
; Wu, 1983
; Hong, 1993
; Wen, 1999
), which has been rarely examined using modern phylogenetic and biogeographic methods. The floristic migrations between eastern Asia and North America have been hypothesized to be via either the Bering land bridge across the northern Pacific (Hopkins, 1967
) or the North Atlantic land bridge indirectly across the Tethys Seaway, southwestern Europe, and the Thulean land bridge (Tiffney, 1985a
). Both migration pathways perhaps allowed the wide distribution of many disjunct plants in the Northern Hemisphere in the Tertiary. The modern disjunction of plants in eastern Asia and North America has largely resulted from the range restrictions of formerly widespread mesophytic forests (Leopold et al., 1972
; Wood, 1972
; Axelrod, 1975
; Tiffney, 1985b
; Boulter, 1994
; Collinson, 1994
; Manchester, 1999
; Wen, 1999
; Tiffney and Manchester, 2001
; Weber, 2003
). The mesophytic forest became widely discontinuous in distribution as late as the middle Miocene in eastern Asia and eastern and western North America because of orogenies and subsequent development of a xeric and higher-elevation coniferous flora in western North America and the gradual cooling of the climate in the Northern Hemisphere, which finally brought heavy glaciations in Europe and in other regions of higher northern latitudes (Graham, 1972
; Wolfe, 1972
; Hoey and Parks, 1991
; Wen, 1999
).
Another important disjunct pattern in the Northern Hemisphere is the disjunction among western North America, the Mediterranean region, and the Tibetan Plateau (Graham, 1999
; Sun and Li, 2003
). Evidence from fossil floras across the Northern Hemisphere indicates the presence of a rich, subtropical or warm temperate forest as the north-Tethyan floristic belt that extended across Laurasia from California in the west through southeastern North America, Europe, and Mediterranean to eastern Asia in the east, even to north China and Japan from the Cretaceous to the Eocene (Takhtajan, 1969
; Axelrod, 1975
; Raven and Axelrod, 1978
; Kubitzki and Krutzsch, 1996
). Recent studies on Malpighiaceae also show that this belt is a conduit of Laurasian migration between the Old and the New Worlds (Davis et al., 2002
). Sun and Li (2003)
suggest that the disjunction between eastern Himalayas and western North America may be related to the pattern between Mediterranean and western North America.
At present, molecular phylogenetic data have been generated only for two genera with an eastern Asian and western North American disjunction: Oplopanax (Miq.) Miq. (Lee and Wen, 2002
) of Araliaceae and Glehnia Schmidt ex Miq. of Apiaceae (Sun et al., 2004
). Oplopanax was suggested to have diverged between the two continents in the Pliocene about 5.6 million years ago (mya). The two species of Glehnia were found to be non-monophyletic. The times of divergence and the migration history of congeners between eastern Asia and western North America need to be analyzed in many more plant groups before any generalizations can be made on the evolution of such a disjunction.
Kelloggia Torrey ex Bentham of Rubiaceae shows such a disjunction between eastern Asia and western North America. It consists of only two species: K. chinensis Franch. (Franchet, 1892
) from the Hengduan Mountains of southwestern China (Smith, 1921
; Handel-Mazzetti, 1936
; Dempster, 1975
; Chen, 1999
; Fang, 2003
) and K. galioides Torrey from the western United States (Bentham and Hooker, 1873
; Torrey, 1874
; Hitchcock and Cronquist, 1973
; Hickman, 1993
). Kelloggia chinensis is distributed at the western edge of eastern Asia, close to central Asia and the Mediterranean region. Many close relatives of Kelloggia (such as Putoria Pers. and Plocama Aiton in the tribe Paederieae) also occur in the Mediterranean region (Puff, 1982
). The biogeographic diversification of Kelloggia and its close relatives needs to be examined in a broader phylogenetic framework including taxa across the Northern Hemisphere.
Kelloggia is characterized by a combination of characters including well-developed creeping rhizomes, opposite, entire, and sessile leaves with small interpetiolar stipules, perfect and four- to five-merous flowers with short funnel-shaped corolla tubes, and small indehiscent fruits covered with hooked bristles. In addition, the whole plant becomes black after drying, which is similar to plants in Paederieae, but in contrast to those in Rubieae. It was recognized by some workers as a monotypic genus, consisting of K. galioides from western North America only (Davis, 1952
; Munz and Keck, 1959
; Abrams and Ferris, 1960
; Munz, 1974
; Demptster, 1975
). The taxonomic and phylogenetic status of the eastern Asian K. chinensis has not been examined since its original description by Franchet in 1892
. The two widely disjunct species of Kelloggia are morphologically highly distinct (Table 1 and Fig. 1). Ecologically K. chinensis occurs in alpine meadows or open places or shrublands along the stream of high mountains (about 3000 m a.s.l.), whereas K. galioides grows in somewhat open places in coniferous forests (1100–3000 m a.s.l.), sometimes among rocks or along streams (Hickman, 1993
).
|
|
originally placed it in the tribe Anthospermeae (also see Baillon, 1879
; Schumann, 1891
). Puff (1982)
placed all the wind-pollinated genera in the tribe Anthospermeae and insect-pollinated genera in the tribe Paederieae and transferred the insect-pollinated Kelloggia to the tribe Paederieae. Bremer and Manen (2000)
did not include Kelloggia in their broad molecular analysis of Rubioideae, yet they still recognized Kelloggia in the paraphyletic tribe Paederieae. Because the two species of Kelloggia exhibit a high level of morphological differentiation and are widely disjunct in two highly isolated biogeographic regions, we attempt to: (1) test the monophyly of Kelloggia using sequences of three DNA markers, (2) if they form a sister species relationship, estimate the divergence time of the two species, and (3) discuss the evolution of the disjunct pattern between eastern Asia and western North America in Kelloggia.
MATERIALS AND METHODS
Taxa and sequencing
Six populations representing both species of Kelloggia and one of Paederia scandens (Lour.) Merr. were sequenced (Appendix 1; see Supplemental Data accompanying the online version of this article). To evaluate the monophyly of genus and its phylogenetic position, sequences of the close relatives of Kelloggia in Rubiaceae, especially tribes Paederieae and Rubieae (Puff, 1982
; Andersson and Rova, 1999
; Bremer and Manen, 2000
), were obtained from GenBank (Appendix 2; see Supplemental Data accompanying the online version of this article). The outgroups were chosen in agreement with results from earlier investigations (Bremer and Manen, 2000
); three taxa (Hillia, Faramea, and Mitchella) were selected from groups outside of tribes Anthospermieae, Paederieae, and Rubieae in Rubiaceae (Fig. 2).
|
. Three chloroplast markers including rps16 intron, atpB-rbcL spacer, and rbcL gene were selected because many sequences of these markers are already available for Rubiaceae (e.g., Bremer et al., 1995
; Manen and Natali, 1996
; Andersson and Rova, 1999
; Bremer and Manen, 2000
). The DNA regions were amplified following the protocols in Bremer et al. (1995)
for rbcL, Andersson and Rova (1999)
for the rps16 intron, and Bremer and Manen (2000)
for the atpB-rbcL spacer. DNA amplifications were performed in 20 µL reactions with approximately 10–50 ng of total DNA, 20 mmol/L Tris buffer (pH 8.3, with 50 mmol/L KCl, 1.5 mmol/L MgCl2, and 0.1% Tween 20), 0.15 mmol/L of each dNTP, 5 µmol/L of each primer, 2 µL of Taq polymerase. The polymerase chain reactions (PCR) were run on a Peltier Thermal Cycler DNA engine DYAD (MJ Research, Watertown, Massachusetts, USA) for 37 cycles starting at 94°C for 2 min, followed by denaturation (94°C for 40 s), annealing (50°C for 2 min), and extension (72°C for 2 min). A 5-min final extension at 72°C followed the 37th cycle to ensure the completion of novel strands. To be cleaned for sequencing, the subsequent PCR products were then run on a 1% low-melting agarose gel, and the bands excised and gelased, using the GELase (Epicentre Technologies, Madison, Wisconsin, USA) method. Cycle sequencing was conducted using BigDye version 3 reagents (Applied Biosystems, Foster City, California, USA) with an ABI model 3100 automated sequencer (Applied Biosystems). Cycling conditions included an initial denaturing step at 94°C for 5 min, followed by 30 cycles of 96°C for 10 s, 50°C for 5 s, and 60°C for 4 min. Each sample was sequenced in the sense and antisense direction. The resulting sequences were assembled using Sequencher 4.1.10 (Gene Codes Corporation, Ann Arbor, Michigan, USA). Sequences obtained in this study have been assigned GenBank accession numbers (Appendix 1). They were then aligned by ClustalX version 1.83 (PC version) (Thompson et al., 1997
), followed by manual adjustments.
Phylogenetic analyses
Phylogenetic analyses were performed with PAUP* 4.0b10 (Swofford, 2003
) using maximum parsimony (MP) (Farris et al., 1970
; Fitch, 1971
) and maximum likelihood (ML) (Felsenstein, 1981
) methods. The heuristic searches for MP analysis with 100 random taxon addition replicates, tree bisection-reconnection (TBR) branch swapping, and collapse of zero-length branches were conducted, and character state changes were equally weighted in the analysis. Gaps were treated either as missing data or as new characters. The amount of support for the clades revealed in the maximally parsimonious tree(s) (MPTs) was examined with 1000 bootstrap replicates with the heuristic search options using parsimony. The appropriate model of DNA substitution for the maximum likelihood analysis was determined using Modeltest version 3.06 (Posada and Crandall, 1998
). Once the best-fit model was determined, maximum likelihood searches were performed for the data set (Felsenstein, 1981
).
Bayesian analyses (Rannala and Yang, 1996
; Mau et al., 1999
) were carried out using MrBayes version 3.0b3 (Huelsenbeck and Ronquist, 2001
, available at http://morphbank.ebc.uu.se/mrbayes/) with the same model parameters as the ML searches. Bayesian analyses started from random trees and employed four Markov chain Monte Carlo (mcmc) runs, monitored over 1 x 106 generations, resampling trees every 100 generations. Runs were repeated twice to confirm results. The resulting log likelihood and number of generations were plotted to determine the point after which the log likelihoods had stabilized. After discarding the trees saved prior to this point as burn-in, the remaining trees were imported into PAUP*, and a 51% majority-rule tree was produced to obtain posterior probabilities of the clades. Internodes with posterior probabilities 
95% were considered statistically significant.
Using the MP and ML methods as described, a phylogenetic analysis was also performed using rbcL sequences to test the monophyly of Kelloggia with a broader sampling of taxa from Rubiaceae, because a large number of rbcL sequences are available in GenBank. A total of 46 rbcL sequences were included in the analysis, encompassing the three main subfamilies (Rubioideae, Ixoroideae, and Cinchonoideae) of Rubiaceae and two outgroups from Kopsia Blume of Apocynaceae and Ligustrum L. of Oleaceae, respectively (Appendix 2; Fig. 3). The ML tree of this broader analysis was also used for estimating the divergence times and inferring biogeographic ancestral areas.
|
) was carried out to test whether the rbcL sequence evolved in a clocklike fashion. This test resulted in P < 0.005, suggesting that rate constancy in this data set was not supported. The penalized likelihood (PL) analysis (Sanderson, 2002
, 2003
) was then used for estimating divergence times. The PL method is a semiparametric approach that uses rate smoothing to allow for robust estimation of node ages in the presence of rate variation between lineages (Sanderson, 2002
). Ages of nodes in the tree were estimated using penalized likelihood rate smoothing under a truncated newton algorithm with the program r8s version 1.60 (Sanderson, 2003
; available at http://ginger.ucdavis.edu/r8s) on a Linux platform. A cross-validation analysis was performed to obtain the most likely smoothing parameter.
The fossils of Rubiaceae are rather poorly known, with only a few unequivocal records in the Northern Hemisphere (Manchester, 1994
; Robbrecht, 1996
). Emmenopterys Oliver is the earliest record of Rubiaceae fruits from the middle Eocene (Manchester, 1994
, 1999
). Pinckneya Michx. fossil is a seed known from the Oligocene (Meyer and Manchester, 1997
). Pollen fossils of Faramea Aublet are known from late Eocene to recent (Graham, 1985
). Some other fossil records in this family were assigned to the family or certain genera on a tentative basis and are thus not used in our analysis (Muller, 1981
; Manchester, 1999
). These include Mitchella L. (Tiffney, 1985b
) and Galium L. (Muller, 1981
; Tiffney, 1985b
). Based on the available fossil data, nodes of the Faramea-Coussarea clade (node F1 in Fig. 3) and the Emmenopterys-Pinckneya clade (node F2 in Fig. 3) were constrained with minimum ages of 40 and 45 mya, respectively. Because of the first record of Rubiaceae fossils is from the early Eocene (Roth and Dilcher, 1979
), the root (node F3 in Fig. 3) of Rubiaceae was set to be 54 mya (also see Magallón-Puebla et al., 1999
; Malcomber, 2002
).
To estimate the standard errors associated with divergence times, we used a parametric bootstrapping strategy similar to that in Baldwin and Sanderson (1998)
and Davis et al. (2002)
: (1) 100 data sets were simulated on the maximum likelihood tree with the computer program Seq-Gen version 1.2.7 (Rambaut and Grassly, 1997
), (2) the divergence times were estimated on the original tree with the sequence model and parameters estimated from the original data, and (3) the resulting simulated data sets were imported into PAUP*, and maximum likelihood trees were generated. The divergence times were estimated on each tree using r8s version 1.60, and the resulting ages of the nodes from the simulated data sets were used to calculate the variance in divergence time estimates.
Ancestral area and DIVA analyses
We used the ancestral area analysis of Bremer (1992)
and dispersal-vicariance analysis (DIVA; Ronquist, 1996
, 1997
) to infer the biogeographic origin of Kelloggia and its close relatives based on their rbcL data (Fig. 4). Three areas of endemism were circumscribed according to the main geographic distribution of Kelloggia and its close relatives (Fig. 4), as well as results of previous biogeographic analysis of the Northern Hemisphere (Wen, 1999
; Donoghue et al., 2001
; Fritsch, 2001
; Sanmartín et al., 2001
; Sanmartín, 2003
). Because our main interests lie in the construction of dispersal/vicariance history of Kelloggia between eastern Asia and western North America, subdividing Eurasia is not necessary in our case. Such subdivision will be useful to gain important insights into the diversification history of Rubiaceae, which is beyond the scope of the present analysis. These three areas of endemisms in our analysis are North America, Eurasia, including the Mediterranean region, and Africa.
|
Dispersal-vicariance analysis (Ronquist, 1997
) is implemented in the computer program DIVA version 1.1 (Ronquist, 1996
). The program optimizes distributions for each node of the tree by favoring vicariance events and minimizing the number of assumed dispersals and extinctions under a parsimony criterion. In DIVA, vicariance events (allopatric speciation) and duplication events (sympatric speciation, i.e., speciation within the area) carry a cost of zero, whereas dispersal and extinction events cost one per unit area added or deleted from the distribution (Ronquist, 1996
). Optimizations minimizing the cost are favored in this analysis.
RESULTS
Phylogenetic analysis
The alignment of 20 rps16 intron sequences of Kelloggia and its close relatives generated a data matrix of 928 positions, 35 of which are indel positions. A total of 111 positions were parsimony-informative. The atpB-rbcL spacer matrix of 19 taxa comprised 769 aligned positions, 58 of which were parsimony-informative, and 26 of which were indels. The 18 aligned rbcL sequences contained 1331 positions without positional homology ambiguities or gaps. Of these, 99 characters were parsimony-informative. Because there is no recombination within the chloroplast genome, we combined all three cpDNA data sets (rps16 intron, atpB-rbcL spacer, and rbcL) of Kelloggia and its close relatives of the tribes Rubieae and Paederieae. The combined data set of 18 taxa comprised 3028 positions, 265 of which were parsimony-informative.
With gaps as missing data, the MP analyses of the combined data set generated one MPT with a total length of 981 steps, a consistency index (CI) excluding uninformative characters of 0.6221, a retention index (RI) of 0.7211, and a rescaled consistency index (RC) of 0.5631. Among the multiple populations sampled for Kelloggia, two clades were resolved with robust support, corresponding to the Asian clade of K. chinensis (100% bootstrap value) and the North American clade of K. galioides (97% bootstrap value). The monophyly of Kelloggia was strongly supported with a 95% bootstrap value and a 100% mcmc posterior probability value in the Bayesian analysis. Analysis of treating gaps as new characters did not change the topology of the trees. Modeltest suggested that the transversion model (TVM + I + G) was the best-fitted model (–ln likelihood score = 9744.8818). The base frequencies were A = 0.3074, C = 0.1704, G = 0.1887, T = 0.3336, and the substitution rate matrix was as follows, A/C = 1.1424, A/ G = 2.1375, A/T = 0.4323, C/G = 1.3059, C/T = 2.1375, G/T = 1.0000. The gamma distribution shape parameter was 0.7901. The maximum likelihood tree had an identical topology to the MPT.
The 46 rbcL sequences of Rubiaceae generated a data matrix of 1331 positions, 212 of which were parsimony informative. The MP analyses generated eight MPTs with a total length of 842 steps, a CI of 0.5107, an RI of 0.7578, and an RC of 0.3870. general time reversible (GTR + I + G) was suggested by Modeltest as the best-fitted model (–ln likelihood score = 6795.6343) for the ML analysis. The base frequencies were A = 0.2687, C = 0.1913, G = 0.2459, T = 0.2941, and the substitution rate matrix was as follows, A/C = 1.0248, A/G = 2.1070, A/T = 0.2963, C/G = 1.2865, C/ T = 3.0423, G/T = 1.0000. The gamma distribution shape parameter was 0.7038. Both the MPTs and MLT also supported the monophyly of Kelloggia, and MLT topology is presented in Fig. 3.
Sequence divergence
The pairwise distances between the two species and among populations of Kelloggia were estimated (Table 2). The highest divergence (1.722–1.825%) was between populations of K. chinensis and K. galioides of rps16 intron sequences. The lowest value (0.00%) was between populations of K. galioides of rps16 intron and atpB-rbcL spacer. The sequence divergence between the two species is much larger than that within the species. The Chinese species showed a higher level of infraspecific sequence variation.
|
The ancestral area analysis employed the rbcL ML tree based on the broad sampling of Kelloggia relatives from Fig. 3. Eurasia was suggested to have played an important role in the diversification of the following major clades of Kelloggia relatives: clade 2 (Kelloggia-Rubieae), clade 3 (Kelloggia-Rubieae-Theligonum), clade 4 (Kelloggia-Rubieae-Putoria-Plocama-Gaillonia), and clade 5 (Kelloggia-Rubieae-Paederieae), with the highest G/L values of 2.00, 2.50, 3.00, and 3.50, respectively (Table 3). As shown in Fig. 4, DIVA suggests that Eurasia was the ancestral area of clades 2–4, when the possible ancestral area was constrained to be two areas (the minimum allowable option). For clade 5, Eurasia and Eurasia plus Africa were possible ancestral areas.
|
Monophyly and phylogenetic position of Kelloggia
Our results based on the combined analysis of three chloroplast markers support the monophyly of the genus comprising these two widely disjunct species: Kelloggia galioides from western North America and K. chinensis from southwestern China (95% bootstrap value, Fig. 2). This Kelloggia clade is also supported by the rbcL analysis alone with a broader sampling (Fig. 3). Nevertheless, the two species have shown a high level of morphological divergence (Table 1; Fig. 1). They occupy highly distinct ecological habitats. Kelloggia galioides from western North America occurs in dry areas and has relatively tall stems and large leaves and is shrublike, whereas K. chinensis of western China is a high montane species in wet meadows of the eastern Himalayas and Hengduan Mountains. The low stems and small hairy leaves of the Chinese species are perhaps adaptations to the habitat of high altitudes. Adaptation of each species to its different habitats after the formation of the disjunction may have largely contributed to the high level of morphological differentiation.
Our analyses of the chloroplast DNA data show that Kelloggia is sister to the Rubieae clade (e.g., Rubia L. and Galium L.) with a bootstrap value of 100% (Fig. 2). Fruits of Kelloggia are similar to those of Galium, but the latter usually has verticillate leaves (four or more leaves on each node). The morphological similarity of Kelloggia and Galium led Smith (1921)
to describe Galium abbrevatus Smith, which is actually a synonym of Kelloggia chinensis (Handel-Mazzetti, 1936
). The tribe Rubieae has been suggested to be monophyletic in previous molecular analysis (Manen et al., 1994
; Manen and Natali, 1995
; Natali et al., 1995
) and pollen morphological studies (Huysmans et al., 2003
). It is characterized by a specific 50-bp deletion of the remaining atpB promotor sequence (Natali et al., 1995
). However, we found that such a deletion was absent in the atpB-rbcL sequences of Kelloggia. The pollen morphology of K. galioides is similar to that of taxa in the tribe Paederieae in having three-colpate, spherical to prolate grains (Robbrecht, 1982
). The relatively medium- to large-sized three-colpate pollen of Kelloggia and Paederieae is different from that of Rubieae, which is characterized by small pollen with 5–13 colpi (Huysmans et al., 2003
). Kelloggia thus perhaps diverged early from the common ancestor with Rubieae. Kelloggia may be best recognized as comprising a separate tribe, but a more comprehensive analysis of Paederieae is needed before a new formal taxonomic treatment of Kellogia at the tribal level is proposed.
Infraspecific variation
Both species of Kelloggia exhibit infraspecific sequence variation (Table 2). The Asian Kelloggia chinensis has a higher level of infraspecific divergence than the North American K. galioides. In Asia, K. chinensis is distributed in the eastern Himalayan region of the Hengduan Mountains, which have undergone a rapidly changing environment corresponding to the uplift of the Himalayas after the collision between the Eurasian and the Indian plates in the Paleogene (Tao, 2000
). The region is characterized by rich biodiversity, with many regional evolutionary radiations, and diversification of many species groups associated with the many isolated habitats and mountains (Wu, 1988
; Li and Li, 1993
; Sun, 2002
). Kelloggia galioides is distributed in California and extends to Arizona, Idaho, Nevada, Oregon, Utah, Washington, and Wyoming. This region also experienced drastic geologic changes in the Tertiary after the collision between the Pacific and the North American plates (Axelrod, 1973
; Wilken, 1993
). The geographic isolation of populations of both species in relatively high montane habitats is perhaps responsible for the observed infraspecific sequence variation.
Historical biogeography
Although the intercontinental distributional pattern in Kelloggia was briefly mentioned in Hong (1993)
, Wu (1983)
, and Ying (1983)
, the evolution of the disjunction has never been critically examined. Our molecular study, together with fossil data, suggests the divergence time between the eastern Asian and western North American species to be about 5.42 ± 2.32 mya (in late Miocene to early Pliocene). The estimated time is similar to that (in the late Miocene) in Datisca (Datiscaceae), which is disjunct between western North America and southwest-central Asia (Liston et al., 1992
). Our estimate does not support vicariance via the North Atlantic land bridge as the explanation for the evolution of the disjunction in Kelloggia. The North Atlantic land bridge was no longer viable by the middle Miocene (Parks and Wendel, 1990
; Tiffney and Manchester, 2001
). Axelrod (1975)
employed the Madrean-Tethyan hypothesis to explain the many disjunctions between western North America and the Mediterranean region in the Old World. He suggested that the age of the disjunct elements was about 20–25 mya. Several recent studies of taxa with representatives in western North America and the Mediterranean region have demonstrated various levels of support for the Madrean-Tethyan hypothesis (Liston et al., 1989a
, b
, 1992
; Fritsch, 1996
, 2001
; Liston, 1997
; Hileman et al., 2001
). However, Raven (1971
, 1973
) and Wolfe (1975)
argued that plants common to the arid regions of western North America and Eurasia were the products of convergent evolution to arid climates from once more widespread mesophytic ancestors or were the result of long-distance dispersal (Köhler and Brückner, 1989
; Ray, 1995
; Manos et al., 1999
). Our estimate in Kelloggia was also apparently much more recent than the age of the Madrean-Tethyan elements. Long-distance dispersal is perhaps the most reasonable explanation for the intercontinental disjunction in Kelloggia. Fruits of Kelloggia possess hooked bristles similar to those in Galium. These hooked bristles may facilitate long-distance dispersal via birds and other animals. But why each of the two species of Kelloggia has so restricted distribution remains an intriguing enigma.
Both the ancestral area and DIVA analyses supported Eurasia as the most likely ancestral area of Kelloggia together with its close relatives (clades 2–4) (Table 3; Fig. 4). Of clade 2, Rubieae is centered in temperate and tropical mountain regions (Natali et al., 1995
; Devesa and Ortega-Olivencia, 2003
), especially highly diverse and rich in endemics in the Mediterranean region and Europe (Mabberley, 1997
). Eurasia thus seemed to be more important for the early diversification of the Kelloggia-Rubieae clade than North America. However, our current analysis included only nine genera of Rubieae with rbcL sequences available in GenBank. Additional biogeographic analysis is needed to further understand the diversification history of Rubieae and Kelloggia.
Clades 3–4 include all the close relatives of Kelloggia including Putoria, Gaillonia, and Plocama in tribe Paederieae and the monotypic tribe Theligoneae (Theligonum) (Rutishauser et al., 1998
; Bremer and Manen, 2000
). As pointed out by Puff (1982)
, Putoria, Gaillonia, and Plocama were more closely related to Kelloggia than to other genera of the tribe. This argument was also supported by the molecular data (clade 4 in Fig. 4). Theligonum was sister to the Kelloggia-Rubieae clade in our analyses (clade 3) and disjunct between the Mediterranean region and eastern Asia (Rutishauser et al., 1998
). Eurasia was also hypothesized to be the most likely ancestral areas for these clades by both the ancestral area and DIVA analyses.
The combined results of biogeographic evidence presented thus supported that Eurasia was highly important in the origin of Kelloggia and its close relatives in the Rubieae-Theligoneae-Paederieae group. Our data suggest that the ancestor of Kelloggia galioides was dispersed from the Old World to western North America at 5.42 ± 2.32 mya (Figs. 3 and 4). Recent biogeographic and phylogenetic analyses have also shown that dispersal was important in the evolution of Rubiaceae (Eriksson and Bremer, 1991
; Bremer and Eriksson, 1992
; Anderson et al., 2001
). A phylogenetic and biogeographic study of Gaertnera (Malcomber, 2002
) of Rubiaceae suggested long-distance dispersal from Africa to southeastern Asia about 5.21 mya ago. The current disjunct distribution of Kelloggia is best explained to be via long-distance dispersal from the Old World into North America after the uplift of the Himalayan Mountains and the western cordillera of North America.
FOOTNOTES
1 This study was supported by grants from the Chinese Natural Science Foundation (CNSF, grant numbers 40332021 and 30270121 to H. Sun), the Innovation Project of the Chinese Academy of Sciences (KSCX2–1–09 to H. Sun), the Excellent Overseas Chinese Research Grant of the Chinese Academy of Sciences (to J. Wen), the Pritzker Lab of the Field Museum, and the US National Science Foundation (DEB0108536 to J. Wen). We thank Dr. Sylvain Razafimandimbison and two anonymous reviewers for their most constructive comments, which greatly improved our manuscript. 
5 Author for correspondence (e-mail: hsun{at}mail.kib.ac.cn
)
LITERATURE CITED
Abrams L. R. S. Ferris 1960 An illustrated flora of the Pacific States: Washington, Oregon, and California, vol. IV. Stanford University Press, Stanford, California, USA
Anderson C. L. J. H. E. Rova L. Andersson 2001 Molecular phylogeny of the tribe Anthospermeae (Rubiaceae): systematic and biogeographic implications. Australian Systematic Botany 14: 231-244
Andersson L. J. H. E. Rova 1999 The rps16 intron and the phylogeny of the Rubioideae (Rubiaceae). Plant Systematics and Evolution 214: 161-186[CrossRef][ISI]
Andreasen K. B. Bremer 1996 Phylogeny of the subfamily Ixoroideae (Rubiaceae). Opera Botanica Belgica 7: 119-138
Andreasen K. B. Bremer 2000 Combined phylogenetic analysis in the Rubiaceae Ixoroideae: morphology, nuclear and chloroplast DNA data. American Journal of Botany 87: 1731-1748
Axelrod D. I. 1973 History of the Mediterranean ecosystem in California. In F. Di Cristi and H. A. Mooney [eds.], Mediterranean type ecosystems: origin and structure, 225–277. Springer-Verlag, New York, New York, USA
Axelrod D. I. 1975 Evolution and biogeography of the Madrean-Tethyan sclerophyll vegetation. Annals of the Missouri Botanical Garden 62: 280-334[CrossRef][ISI]
Baillon H. 1879 Sur les Gaertnera et sur la valeur du groupe des Gaertnerées. Bulletin Mensuel de la Société Linnéenne de Paris 1: 209-210
Baldwin B. M. J. Sanderson 1998 Age and rate of diversification of the Hawaiian silversword alliance. Proceedings of the National Academy of Sciences, USA 95: 9402-9406
Bentham G. J. D. Hooker 1873 Genera plantarum, vol. 2. Reeve, London, UK
Boufford D. E. 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
Boulter M. C. 1994 Towards a review of Tertiary palaeobotany in the boreal realm. In M. C. Boulter and H. C. Fisher [eds.], Cenozoic plants and climates of the Arctic, 1–13. Springer-Verlag, Heidelberg, Germany
Bremer B. 1996 Phylogenetic studies within Rubiaceae and relationships to other families based on molecular data. Opera Botanica Belgica 7: 33-50
Bremer B. 1997 Combined and separate analyses of morphological and molecular data in Rubiaceae. Cladistics 12: 21-40[ISI]
Bremer B. K. Andreasen D. Olsson 1995 Subfamilial and tribal relationships in the Rubiaceae based on rbcL sequence data. Annals of the Missouri Botanical Garden 82: 383-397[CrossRef][ISI]
Bremer B. O. Eriksson 1992 Evolution of fruit characters and dispersal modes in the tropical family Rubiaceae. Biological Journal of the Linnean Society 47: 79-95
Bremer B. R. K. Jansen B. Oxelman M. Backlund H. Lantz K. J. Kim 1999 More characters or more taxa for a robust phylogeny in a case study from the coffee family (Rubiaceae). Systematic Biology 48: 413-435[CrossRef][ISI][Medline]
Bremer B. J. F. Manen 2000 Phylogeny and classification of the subfamily Rubioideae (Rubiaceae). Plant Systematics and Evolution 225: 43-72[CrossRef][ISI]
Bremer B. M. Thulin 1998 Collapse of Isertieae, re-establishment of Mussaendeae, and a new genus of Sabiceeae (Rubiaceae); phylogenetic relationships based on rbcL data. Plant Systematics and Evolution 211: 71-92[CrossRef][ISI]
Bremer K. 1992 Ancestral areas: a cladistic reinterpretation of the center of origin concept. Systematic Biology 41: 436-445[CrossRef][ISI]
Chen W. C. 1999 Flora Reipublicae Popularis Sinicae, vol. 71, Rubiaceae (Rubioideae). Science Press, Beijing, China
Collinson M. E. 1994 Arcto-Tertiary '93: perspectives and prospects. In M. C. Boulter and H. C. Fisher [eds.], Cenozoic plants and climates of the Arctic, 383–388. Springer-Verlag, Heidelberg, Germany
Davis C. C. C. C. Bell S. Mathews M. J. Donoghue 2002 Laurasian migration explains Gondwanan disjunctions: evidence from Malpighiaceae. Proceedings of the National Academy of Sciences, USA 99: 6833-6837
Davis R. J. 1952 Flora of Idaho. William C. Brown, Dubuque, Iowa, USA
Dempster L. T. 1975 An Asian Kelloggia (Rubiaceae). Madroño 23: 100-101
Devesa J. A. A. Ortega-Olivencia 2003 A new species of Valantia (Rubiaceae) from Spain. Botanical Journal of the Linnean Society 143: 331-335[CrossRef]
Donoghue M. J. C. D. Bell J. Li 2001 Phylogenetic patterns in Northern Hemisphere plant geography. International Journal of Plant Sciences 162: S41-S52[CrossRef]
Doyle J. J. J. L. Doyle 1987 A rapid isolation procedure from small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15
Eriksson O. B. Bremer 1991 Fruit characteristics, life forms, and species richness in the plant family Rubiaceae. American Naturalist 138: 751-761[CrossRef][ISI]
Fang R. Z. 2003 Flora Yunnanica, vol. 15. Science Press, Beijing, China
Farris J. S. A. G. Kluge M. J. Eckardt 1970 A numerical approach to phylogenetic systematics. Systematic Zoology 19: 172-191[CrossRef]
Felsenstein J. 1981 Evolutionary trees from DNA sequences: a maximum likelihood approach. Journal Molecular Evolution 17: 368-376[CrossRef]
Felsenstein J. 1988 Phylogenies from molecular sequences: inference and reliability. Annual Review of Genetics 22: 521-565[CrossRef][ISI][Medline]
Fitch W. M. 1971 Toward defining the course of evolution: minimal change for a specific tree topology. Systematic Zoology 20: 406-416[CrossRef][ISI]
Franchet A. 1892 Notes sur un Kelloggia de la Chine. Journal de Botany (Morot) 6: 10-13
Fritsch P. 1996 Isozyme analysis of intercontinental disjuncts within Styrax (Styracaceae): implications for the Madrean-Tethyan hypothesis. American Journal of Botany 83: 342-355[CrossRef][ISI]
Fritsch P. 2001 Phylogeny and biogeography of the flowering plant genus Styrax (Styracaceae) based on chloroplast DNA restriction sites and DNA sequences from the internal transcribed spacer region. Molecular Phylogenetics and Evolution 19: 387-408[CrossRef][ISI][Medline]
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, Netherlands
Graham A. 1985 Studies of Neotropical paleobotany. IV. The Eocene communities of Panama. Annals of the Missouri Botanical Garden 72: 504-534
Graham A. 1999 Late Cretaceous and Cenozoic history of North American vegetation. Oxford University Press, Oxford, UK
Handel-Mazzetti H. 1936 Symbolae Sinicae, Botanische Ergebnisse der Expedition der Akademie der Wissenschaften in wien nach Südwest-China 1914/1918, vol. 7. Springer, Wien, Germany
Hickman J. C. 1993 The Jepson manual: higher plants of California. University of California Press, Berkeley, California, USA
Hileman L. C. M. C. Vasey V. T. Parker 2001 Phylogeny and biogeography of the Arbutoideae (Ericaceae): implications for the Madrean-Tethyan hypothesis. Systematic Botany 26: 131-143[ISI]
Hitchcock C. L. A. Cronquist 1973 Vascular plants of the Pacific Northwest. University of Washington Press, Seattle, Washington, USA
Hoey M. T. C. R. Parks 1991 Isozyme divergence between eastern Asia, North American and Turkish species of Liquidambar (Hamamelidaceae). American Journal of Botany 78: 938-947[CrossRef][ISI]
Hong D. Y. 1993 Eastern Asian-North American disjunctions and their biological significance. Cathaya 5: 1-39
Hooker J. D. 1873 Rubiaceae. In G. Bentham and J. D. Hooker [eds.], Genera plantarum 2, 7–15. Reeve, London, UK
Hopkins D. M. 1967 The Bering Land Bridge. Stanford University Press, Stanford, California, USA
Huelsenbeck J. P. R. Ronquist 2001 MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754-755
Huysmans S. S. Dessein E. Smets E. Robbrecht 2003 Pollen morphology of NW European representatives confirms monophyly of Rubieae (Rubiaceae). Review of Palaeobotany and Palynology 127: 219-240[CrossRef][ISI]
Köhler E. P. Brückner 1989 The genus Buxus (Buxaceae): aspects of its differentiation in space and time. Plant Systematics and Evolution 162: 267-283[CrossRef][ISI]
Kubitzki K. W. Krutzsch 1996 Origins of east and south Asian plant diversity. In A. L. Zhang and S. G. Wu [eds.], Floristic characteristics and diversity of East Asian plants, 25–27. Higher Education Press, Beijing, China
Lee C. J. Wen 2002 Phylogeny and biogeography of the Asian clade of Araliaceae: insights from the nuclear ribosomal and chloroplast DNA data. Abstract, Botany 2002, annual meeting of Botanical Society of America, Madison, Wisconsin, USA. Available at website, http:// www.botany2002.org/section12/abstracts/201.shtml
Leopold E. B. H. D. MacGinitie 1972 Development and affinities of Tertiary floras in the Rocky Mountains. In A. Graham [ed.], Floristics and paleofloristics of Asia and eastern North America, 147–200. Elsevier, Amsterdam, Netherlands
Li H. L. 1952 Floristic relationships between eastern Asia and eastern North America. Proceedings of the Academy of Natural Sciences of Philadelphia 42: 371-429
Li X. W. J. Li 1993 A preliminary floristics study on the seed plants from the region of Hengduan Mountain. Acta Botanica Yunnanica 15: 217-231
Liston A. 1997 Biogeographic relationships between the Mediterranean and North American floras: insights from molecular data. Lagascalia 19: 323-330
Liston A. L. H. Rieseberg T. Elias 1989a Morphological stasis and molecular divergence in the intercontinental disjunct genus Datisca (Datiscaceae). Aliso 12: 525-542
Liston A. L. H. Rieseberg T. Elias 1989b Genetic similarity is high between intercontinental disjunct species of Senecio (Asteraceae). American Journal of Botany 76: 383-388[CrossRef][ISI]
Liston A. L. H. Rieseberg M. A. Hansen 1992 Geographic partitioning of chloroplast DNA variation in the genus Datisca (Datiscaceae). Plant Systematics and Evolution 181: 121-132[CrossRef][ISI]
Mabberley D. J. 1997 The plant-book: a portable dictionary of the vascular plants, 2nd ed. Cambridge University Press, Cambridge, UK
Magallón-Puebla S. P. R. Crane P. S. Herendeen 1999 Phylogenetic pattern, diversity, and diversification of eudicots. Annals of the Missouri Botanical Garden 86: 297-372[CrossRef][ISI]
Malcomber S. T. 2002 Phylogeny of Gaertnera Lam. (Rubiaceae) based on multiple DNA markers: evidence of a rapid radiation in a widespread, morphologically diverse genus. Evolution 56: 42-57[ISI][Medline]
Manchester S. R. 1994 Fruits and seeds of the Middle Eocene Nut Beds flora, Clarno Formation, North Central Oregon. Palaeontographica Americana 58: 1-205
Manchester S. R. 1999 Biogeographical relationships of North American Tertiary floras. Annals of the Missouri Botanical Garden 86: 472-522[CrossRef][ISI]
Manen J. F. A. Natali 1995 Comparison of the evolution of rbcL and atpB-rbcL noncoding spacer sequences in a recent plant group, the tribe Rubieae (Rubiaceae). Journal Molecular Evolution 41: 920-927
Manen J. F. A. Natali 1996 The chloroplast atpB-rbcL spacer in Rubiaceae. Opera Botanica Belgica 7: 51-57
Manen J. F. A. Natali F. Ehrendorfer 1994 Phylogeny of Rubiaceae-Rubieae inferred from the sequence of a cpDNA intergene region. Plant Systematics and Evolution 190: 195-211[CrossRef][ISI]
Manos P. S. J. J. Doyle K. C. Nixon 1999 Phylogeny, biogeography, and process of molecular differentiation in Quercus subgenus Quercus (Fagaceae). Molecular Phylogenetics and Evolution 12: 333-349[CrossRef][ISI][Medline]
Mau B. M. Newton B. Larget 1999 Bayesian phylogenetic inference via Markov chain Monte Carlo methods. Biometrics 55: 1-12[CrossRef][ISI][Medline]
Meyer H. W. S. R. Manchester 1997 The Oligocene Bridge Creek flora of the John Day Formation, Oregon. University of California Publications in Geological Science 141: 1-195
Muller J. 1981 Fossil pollen records of extant angiosperms. Botanical Review 47: 1-142
Mueller G. M. Q. X. Wu Y. Q. Huang S. Y. Guo R. Aldana Gomez R. Vilgalys 2001 Assessing biogeographic relationships between North American and Chinese macrofungi based on molecular data. Journal of Biogeography 28: 271-281[CrossRef][ISI]
Munz P. A. 1974 A flora of southern California. University of California Press, Berkeley, California, USA
Munz P. A. D. D. Keck 1959 A California flora. University of California Press, Berkeley, California, USA
Natali A. J. F. Manen F. Ehrendorfer 1995 Phylogeny of the Rubiaceae-Rubioideae: evidence from a non-coding chloroplast sequence. Annals of the Missouri Botanical Garden 82: 428-439[CrossRef][ISI]
Nepokroeff M. B. Bremer K. J. Sytsma 1999 Reorganization of the genus Psychotria and tribe Psychotrieae (Rubiaceae) inferred from ITS and rbcL sequence data. Systematic Botany 24: 5-27
Olmstead R. G. B. Bremer K. Scott J. D. Palmer 1993 A parsimony analysis of the Asteridae sensu lato based on rbcL sequences. Annals of the Missouri Botanical Garden 80: 700-722[CrossRef][ISI]
Olmstead R. G. H. J. Michaels K. M. Scott J. D. Palmer 1992 Monophyly of the Asteridae and identification of their major lineages inferred from DNA sequences of rbcL. Annals of the Missouri Botanical Garden 79: 249-265
Parks C. R. J. F. Wendel 1990 Molecular divergence between Asian and North American species of Liriodendron (Magnoliaceae) with implications of fossil floras. American Journal of Botany 77: 1243-1256[CrossRef][ISI]
Posada D. K. A. Crandall 1998 MODELTEST: testing the model of DNA substitution. Bioinformatics 14: 817-818
Puff C. 1982 The delimitation of the tribe Anthospermeae and its affinities to the Paederieae (Rubiaceae). Botanical Journal of the Linnean Society 84: 355-377
Rambaut A. N. C. Grassly 1997 Seq-Gen: an application for the Monte Carlo simulation of DNA sequence evolution along phylogenetic trees. Computer Applications in the Biosciences 13: 235-238
Rannala B. Z. H. Yang 1996 Probability distribution of molecular evolutionary trees: a new method of phylogenetic inference. Journal of Molecular Evolution 43: 304-311[ISI][Medline]
Raven P. H. 1971 The relationships between ‘Mediterranean’ floras. In P. H. Davis, P. C. Harper, and I. C. Hedge [eds.], Plant life of south-west Asia, 119–134. Botanical Society of Edinburgh, Aberdeen, UK
Raven P. H. 1973 The evolution of Mediterranean floras. In F. Di Cristi and M. A. Mooney [eds.], Mediterranean type ecosystems: origin and structure, 213–224. Springer-Verlag, New York, New York, USA
Raven P. H. D. I. Axelrod 1978 Origin and relationships of the California flora. University of California Press, Berkeley, California, USA
Ray M. F. 1995 Systematics of Lavatera and Malva (Malvaceae, Malveae)—a new perspective. Plant Systematics and Evolution 198: 29-53[CrossRef][ISI]
Razafimandimbison S. B. Bremer 2002 Phylogeny and classification of Naucleeae s.l. (Rubiaceae) inferred from Molecular (ITS, trnT-L, rbcL) and morphological data. American Journal of Botany 89: 1027-1041
Robbrecht E. 1982 Pollen morphology of the tribes Anthospermeae and Paederieae (Rubiaceae) in relation to taxonomy. Bulletin du Jardin Botanique National de Belgique 52: 349-366[CrossRef]
Robbrecht E. 1996 Generic distribution patterns in sub-Saharan African Rubiaceae (Angiospermae). Journal of Biogeography 23: 311-328[CrossRef][ISI]
Ronquist F. 1996 DIVA: dispersal–vicariance analysis, version 1.1. Uppsala University, Uppsala, Sweden
Ronquist F. 1997 Dispersal–vicariance analysis: a new approach to the quantification of historical biogeography. Systematic Biology 46: 195-203[CrossRef][ISI]
Roth J. L. D. L. Dilcher 1979 Investigations of angiosperms from the Eocene of North America: stipulate leaves of the Rubiaceae including a probable polyploid population. American Journal of Botany 66: 1194-1207[CrossRef][ISI]
Rutishauser R. L. P. Ronse Decraene E. F. Smets I. Mendoza-Heuer 1998 Theligonum cynocrambe: developmental morphology of a peculiar rubiaceous herb. Plant Systematics and Evolution 210: 1-24[CrossRef][ISI]
Sanderson M. J. 2002 Estimating absolute rates of molecular evolution and divergence times: a penalized likelihood approach. Molecular Biology and Evolution 19: 101-109
Sanderson M. J. 2003 r8s: analysis of rates ("r8s") of evolution (and other stuff), version 1.60. http://ginger.ucdavis.edu/r8s/
Sanmartín I. 2003 Dispersal vs. vicariance in the Mediterranean: historical biogeography of the Palearctic pachydeminae (Coleoptera, Scarabaeoidea). Journal of Biogeography 30: 1883-1897[CrossRef][ISI]
Sanmartín I. H. Enghoff F. Ronquist 2001 Patterns of animal dispersal, vicariance and diversification in the Holarctic. Biological Journal of the Linnean Society 73: 345-390
Schumann K. 1891 Rubiaceae. In A. Engler and K. Prant [eds.], Die Natürlichen Pflanzenfamilien 4(4). Engelmann, Leipzig, Germany
Smith W. W. 1921 Galium aberrans W. W. Sm. Notes from the Royal Botanica Garden Edinburgh 13: 161-162
Sun F. J. S. R. Downie R. L. Hartman 2004 An ITS-based phylogenetic analysis of the perennial, endemic Apiaceae subfamily Apioideae of western North America. Systematic Botany 29: 419-431[CrossRef][ISI]
Sun H. 2002 Tethys retreat and Himalayas-Hengduanshan Mountains uplift and their significance on the origin and development of the Sino-Himalayas elements and alpine flora. Acta Botanica Yunnanica 24: 273-288
Sun H. Z. M. Li 2003 Qinghai-Tibet Plateau uplift and its impact on Tethys flora. Advances in Earth Sciences 18: 852-862
Swofford D. L. 2003 PAUP*: phylogenetic analysis using parsimony (* and other methods), version 4.0b 10. Sinauer, Sunderland, Massachusetts, USA
Takhtajan A. 1969 Flowering plants, origin and dispersal. Oliver & Boyd, Edinburgh, UK
Tao J. R. 2000 The evolution of the late Cretaceous-Cenozoic floras in China. Science Press, Beijing, China
Thompson J. D. T. J. Gibson F. Plewniak F. Eanmougin D. G. Higgins 1997 The ClustalX Windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 24: 4876-4882
Thorne R. F. 1972 Major disjunctions in the geographical ranges of seed plants. Quarterly Review of Biology 47: 365-411[CrossRef]
Tiffney B. H. 1985a The Eocene North Atlantic land bridge and its importance in Tertiary and modern phytogeography of the Northern Hemisphere. Journal of the Arnold Arboretum 66: 243-273[ISI]
Tiffney B. H. 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]
Tiffney B. H. S. R. Manchester 2001 The use of geological and paleontological evidence in evaluating plant phylogeographic hypotheses in the North Hemisphere Tertiary. International Journal of Plant Sciences 162: S3-S17[CrossRef]
Torrey J. 1874 Kelloggia. In A. Gray [ed.], United States Exploring Expedition, during the years 1838, 1839, 1840, 1841, 1842, under the command of Charles Wilkes, U.S.N., vol. 17, 332. Sherman, Philadelphia, Pennsylvania, USA
Weber W. A. 2003 The middle Asian element in the southern Rocky Mountain Flora of the western United States: a critical biogeographical review. Journal of Biogeography 30: 649-685[CrossRef][ISI]
Wen J. 1999 Evolution of eastern Asian and eastern North American disjunct distributions in flowering plants. Annual Review of Ecology and Systematics 30: 421-455
