| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Systematics and Phytogeography |
Institute for Systematic Botany, Friedrich Schiller University Jena, Philosophenweg 16, D-07743 Jena, Germany; Studiengemeinschaft Südamerikanische Kakteen e.V., Treeneblick 11, D-24852 Langstedt, Germany
Received for publication July 13, 2006. Accepted for publication June 20, 2007.
ABSTRACT
The tropical Andes harbor a major part of the world's plant biodiversity. The montane cacti of the tribes Browningieae, Cereeae, and Trichocereeae underwent extensive radiation and thus are well suited as a model group to study the diversification of Andean plants. We reconstructed their phylogeny employing three noncoding chloroplast regions and explained it in the context of the geological history of South America. We found that the clade of cephalia-bearing cacti with naked pericarpels is centered in northeastern Brazil, whereas almost all other clades comprise Andean species. The spatial split between the clades was probably caused by the Andean uplift and the concurrent formation of intracontinental marine basins in the Tertiary. The phylogenetic reconstructions based on parsimony and Bayesian approaches do not reflect the traditional delimitation of the tribes and of the large genera. Our results suggest that Rebutia s.l. and Echinopsis s.l. are not monophyletic and that Sulcorebutia, Weingartia, and Cintia should be united into one genus. Even though this "Weingartia-complex" and the genus Gymnocalycium are similar in size and morphological diversity, Gymnocalycium has a very high molecular divergence suggesting a comparably older radiation.
Key Words: atpB-rbcL intergenic spacer • biogeography • Cactaceae • Cereeae • South America • Trichocereeae • trnK-rps16 intergenic spacer • trnL-trnF intergenic spacer
The flora of the tropical Andes is rich in endemics. Seven percent of the world's vascular plant species are endemic to this region (Myers et al., 2000
). Some studies tried to explain this by recent speciation (Bell and Donoghue, 2005
; Hughes and Eastwood, 2006
). The explosive diversification of Andean clades was thought to be enabled by the emergence of heterogeneous ecological conditions in island-like habitats after the Andean uplift and quaternary climate fluctuations facilitating geographical isolation. Furthermore, the Andean uplift during the Miocene caused marine incursions into the South American continent (Horton and DeCelles, 1997
) leading to geographic isolation of large areas and might have triggered allopatric speciation (Nores, 1999
). The knowledge about phylogenetic relationships of Andean plants provides a rich source of new insights into evolutionary processes underlying present patterns of biodiversity (Young et al., 2002
). Cactaceae are well-suited as a model group to study diversification connected with the Andean uplift because of their species richness in the Andes and in northeastern Brazil.
The cacti have always inspired botanists and enthusiasts with their great diversity, unusual morphology, and remarkable flowers. As in other ornamental plants (e.g., orchids and roses), the attention of collectors resulted in an inflated number of described species and a flood of synonyms. Current classification of Cactaceae is based mainly on morphology. A group of cactus taxonomists within the International Organization for Succulent Plant Study has tried to achieve a consensus between the widely varying views concerning the number of genera and species within the family (e.g., Backeberg, 1977
; Hunt, 2006
). However, the high level of convergence within Cactaceae does not allow for reliable morphology-based classification. Despite that, the list of molecular phylogenetic studies within this family is rather short (Wallace, 1995b
; Hartmann et al., 2001
; Applequist and Wallace, 2002
; Butterworth et al., 2002
; Griffith, 2002
; Nyffeler, 2002
; Wallace and Dieckie, 2002
; Butterworth and Wallace, 2004
; Edwards et al., 2005
; Harpke and Peterson, 2006
).
The Cactaceae contain 1500–1800 species and are distributed in both Americas from southern Patagonia to Canada (Barthlott and Hunt, 1993
). The family is traditionally divided into three subfamilies: the Pereskioideae, the Opuntioideae, and the Cactoideae (Schumann, 1898
), but Wallace (1995a)
, Nyffeler (2002)
, and Anderson (2005) suggested a fourth monogeneric subfamily, Maihuenioideae. The monophyletic subfamily Cactoideae (Nyffeler, 2002
) contains more than 80% of the described species. Within this subfamily, the arrangement of the genera into different tribes is still disputed. For example, in the largely congruent descriptions of the tribes Cereeae and Trichocereeae the only differentiating trait is the surface of the pericarpel, which is usually naked except for a few scales in Cereeae and usually scaly or hairy in Trichocereeae (Barthlott and Hunt, 1993
; Anderson, 2005
). The molecular phylogenetic analysis by Nyffeler (2002)
based on chloroplast DNA sequences clearly shows that the tribes of any classification within the Cactoideae are not monophyletic. In the present study we mainly focus on cacti of the subtropical South American mountainous regions, which constitute a well-supported monophyletic group within Cactaceae in Nyffeler's (2002) study. He named this clade "BCT clade" according to the initial letters of the included tribes Browningieae Buxb., Cereeae Salm-Dyck, and Trichocereeae Buxb. We expanded the taxon sampling within this clade to get further insights into the phylogenetic relationships especially between the globose to short cylindrical cacti of the genera http://Gymnocalyciumhttp:// Pfeiff,Rebutia K.Schum., Sulcorebutia Backeb., and Weingartia Werderm. In recent classifications the latter three genera are merged into the genus Rebutia s.l. (Barthlott and Hunt, 1993
; Hunt, 1999
; Anderson, 2001
; Hunt, 2006
), although the relationships of the four genera are still controversial. Endler and Buxbaum (1974)
and Backeberg (1977) assumed a close relationship between Gymnocalycium and Weingartia on one hand and Rebutia and Sulcorebutia on the other, emphasizing the flowers that emerge from either older lateral or younger apical areoles. Brinkmann (1976)
, Augustin et al. (2000)
, and Augustin and Hentzschel (2002)
considered Sulcorebutia and Weingartia to be very closely related to each other, because of the similar form and position of areoles and the similar shape of flowers and fruits.
Previous studies have treated the genus Rebutia s.s. in different ways. Buining and Donald (1963
, 1965
) divided Rebutia into six sections classified in two subgenera: Rebutia and Aylostera Speg., the latter differing by fusion of flower tubes with styles and filaments. Krainz (1967)
rejected the two subgenera because he assigned little importance to this character; the fusion of organs may be explained by a stretching of the flower tube during different ontogenetic stages. Krainz kept five of Buining and Donald's sections, using the presence of hairs and bristles at the flower tube, the ability to self-fertilization, and the globose or cylindrical body shape as differentiating characters. On the basis of essentially the same characters as in Buining and Donald's study, Backeberg (1977)
recognized three genera: Aylostera, Mediolobivia Backeb., and Rebutia, which he even assigned to different higher informal taxonomic entities.
Another complex within the BCT clade contains the small globose day-flowering genus Lobivia Britton & Rose and the columnar night-flowering genera Echinopsis Zucc., Setiechinopsis Backeb., and Trichocereus (A. Berger) Riccob., which have recently been united in Echinopsis s.l. by Barthlott and Hunt (1993)
, Hunt (1999)
, Anderson (2005)
, and Hunt (2006). Wishing to gain further insights into the phylogenetic relationships of these genera, we sequenced three noncoding intergenic regions of the chloroplast DNA: the 5' region of the atpB-rbcL intergenic spacer (IGS), the trnL-trnF IGS, and the trnK-rps16 IGS. The trnK-rps16 IGS is used for the first time in phylogenetic analyses, and we designed the primers for this study. We did not sequence nuclear genes because accumulation of low copy genes for our 87 taxa would require a rather prohibitive amount of cloning effort. Approaches using the internal transcribed spacer of the nuclear ribosomal DNA (nrITS), which is easy to amplify because of its high copy number per genome, failed to yield reliable topologies of the Cactaceae because of extensive paralogy due to a lack of concerted evolution (Hartmann et al., 2001
; Harpke and Peterson, 2006
; C. M. Ritz, unpublished data).
MATERIALS AND METHODS
Taxon sampling
We assigned Austrocactus philippii, Copiapoa laui, Eriosyce napina, Neowerdermannia vorwerkii, Parodia magnifica, P. uebelmanniana (Notocacteae Buxb.), Neoraimondia arequipensis subsp. roseiflora, and Neoraimondia herzogiana (Browningiae) to the outgroup following conclusions of a recent molecular study (Nyffeler, 2002
). The ingroup taxa represent South American cacti of the "BCT clade" as defined by Nyffeler (2002)
with an emphasis on the genera Gymnocalycium, Rebutia, Sulcorebutia, and Weingartia. We only sampled specimens that could be assigned unambiguously to a species to avoid including hybrids in the analysis. All voucher specimens are deposited at the succulent collection of the Botanical Garden Jena. Sampling details and GenBank accession numbers are presented in the Appendix. We used the nomenclature of Anderson (2005)
, except for the genus Weingartia for which we followed Augustin and Hentzschel (2002)
.
DNA isolation
Total DNA was extracted from silica gel-dried material of living plants using DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) and following the user's protocol, except that all centrifugation steps were performed at maximum speed (20,000 x g) because of the high polysaccharide content of the plant tissue.
Amplification
Amplification of double-stranded DNA was performed in 25 µL containing 2.5 µL 10-fold polymerase buffer (100 mM Tris-HCl pH 9.0, 500 mM KCl, 15 mM MgCl2, 1.0% Triton X-100, 2.0 mg/mL BSA), 2.5 µL 2 mM dNTP, 1 µL of each primer (10 pmol/µL), one unit of Taq polymerase (MBI Fermentas, St. Leon-Rot, Germany), and 1 µL DNA template from the diluted extracts. Primer sequences for the 5' region of the atpB-rbcL IGS were taken from Savolainen et al. (1994)
: "2" 5'-GAAGTAGTAGGATTGATTCTC-3' and "10" 5'-CATCATTATTGTATACTCTTTC-3'. Primer sequences for the amplification of the trnL-trnF IGS were taken from Taberlet et al. (1991)
: "E" 5'-GGTTCAAGTCCCTCTATCCC-3' and "F" 5'-ATTTGAACTGGTGACACGAG-3'. Primers for the amplification of the trnK-rps16 IGS "TKR16-for" 5'-GCCGAGTACTCTACC-3' and "TKR16-rev" 5'-CGAATCGTTGCAATTG-3' were developed from an alignment of published chloroplast sequences of angiosperms. The amplification was performed for 180 s at 95°C; followed by 28 cycles of 30 s at 95°C, 60 s at 42°C, and 120 s at 72°C; and a final extension for 180 s at 72°C.
Sequencing
PCR products were directly sequenced from both ends using the same primers as for the amplification. Cycle-sequencing was performed using the Thermo Sequenase labelled primer cycle sequencing kit with 7-deaza-dGTP (Amersham Pharmacia, Uppsala, Sweden). The cycle sequencing profile consisted of an initial denaturation at 95°C for 120 s followed by 25 cycles of 95°C for 15 s, TA for 15 s (TA = 57°C for the atpB-rbcL IGS and the trnL-trnF IGS, TA = 42°C for the trnK-rps16 IGS), and 70°C for 15 s. The resulting DNA fragments were separated on a 6% polyacrylamide gel, using an automated DNA sequencer (LICOR 4000L; Lincoln, Nebraska, USA).
Phylogenetic analyses
DNA sequences of the three markers were aligned manually. Although all regions are located in the plastid genome, which is inherited as a unit without recombination, we assessed incongruence between the three spacers with the incongruence length difference (ILD) test (Farris et al., 1995
), implemented as the partition homogeneity test in PAUP* version 4.0b10 (Swofford, 2002
), using a full heuristic search, simple taxon addition sequence, tree-bisection-reconnection (TBR) branch swapping, and with MaxTrees set to 100. The final alignment has been deposited at TreeBase http://www.herbaria.harvard.edu/treebase/ (matrix accession number: M3183, M3184; study accession number: S1744).
Phylogenetic analyses were performed using Bayesian and maximum parsimony approaches. Bayesian inference of phylogeny using Monte Carlo Markov chains (MCMC) was done with MrBayes 3.0b4 (Huelsenbeck and Ronquist, 2001
). Four incrementally heated simultaneous MCMC were run over 2 000 000 generations, under assumptions of a GTR + I + G model, using random starting trees and default starting values of the model. GTR + I + G was chosen as the best fitting model for our data by both the Akaike information criterion and the likelihood ratio tests implemented in MrModeltest 2.0 (Nylander, 2004
). Trees were sampled every 100 generations resulting in an overall sampling of 20 001 trees. The first 2000 trees were discarded as "burn-in." From the remaining trees, a majority rule consensus tree showing all compatible partitions was computed to obtain estimates for the posterior probabilities (PP). Branch lengths were estimated as mean values over the sampled trees. This Bayesian analysis was repeated four times, always using random starting trees and random starting values for the model parameters to test the independence of the results from the revisiting of the prior topologies during chain growth (Huelsenbeck et al., 2002
).
Parsimony analyses were performed using the heuristic search mode in PAUP* 4.0b10 (Swofford, 2002
) with 100 random addition sequence replicates and TBR branch swapping. To limit the computational effort, we restricted the maximum number of saved trees to 50 000. All character states were treated as unordered and equally weighted. Gaps were treated as missing data and additionally binary-coded according to Baldwin et al. (1995)
. Branch support was evaluated as bootstrap percentages (BP) from 1000 bootstrap replicates (Felsenstein, 1985
) and by Bremer support (BS) (Bremer, 1994
) using the software AutoDecay 3.0 (Eriksson and Wikström, 1995
).
RESULTS
Descriptive data on the chloroplast markers
The atpB-rbcL IGS, the trnL-trnF IGS, and the trnK-rps16 IGS were sequenced for all taxa investigated, except for some whose trnL-trnF sequences were taken from GenBank (Appendix). In a few samples, amplification of one of the markers failed repeatedly (see Appendix). An overview about alignment length, number of informative characters, number of binary-coded indels of each of the three markers, and the combined data set is given in Table 1. In the partition homogeneity test, there was no significant incongruence among the three data sets (P = 0.07). Therefore, we combined the three fragments into a single data set with a total length of 2412 positions.
|
, is supported by 100% a posteriori probability (PP), a bootstrap percentage (BP) of 96, and a Bremer support value (BS) of 14. Bayesian and parsimony analyses detected largely the same major clades within the ingroup (clades A–E in Figs. 1, 2).
|
|
Clade B comprises various genera of the Trichocereeae with hairy pericarpels. The monophyly of this clade is rather weakly supported in the parsimony analysis (BS = 1, BP = 54%). In the Bayesian analysis, the genus Gymnocalycium (clade C) is nested within clade B; therefore it is referred to as B* (PP = 96%) in Fig. 2.
Branch support of clade C, comprising nine species of Gymnocalycium, is very high (PP = 100%, BP = 100%, BS = 26), and branches are remarkably long (Fig. 2).
The monophyly of clade D, comprising six species of Rebutia, is supported by PP = 100, BP = 89%, and BS = 3. There are two monophyletic groups within clade D: (1) Rebutia pseudodeminuta, R. fiebrigii, and R. deminuta (PP = 100%, BP = 100%, BS = 7); and (2) Rebutia pygmaea and R. steinmannii (PP = 99%, BP = 100%, BS = 5). Rebutia einsteinii belongs to clade D but is not included in either of the two groups within this clade.
Clade E (PP = 84%, BP = 78%, BS = 3) contains Browningia Britton & Rose, two species of Rebutia, as well as the genera Sulcorebutia, Weingartia, and Cintia Kní
e & 
íha. The last three form a well-supported monophyletic group (PP = 100%, BP = 100%, BS = 9), but the relationships within this clade are poorly resolved. The positions of the two species of Browningia are different, though weakly supported, in the two approaches (Figs. 1, 2).
The phylogenetic relationships among clades A–E and the taxa not assigned to any of these clades (Cereus Mill., Stetsonia Britton & Rose) remain unclear. In the parsimony analysis Cereus hildmannianus and Stetsonia coryne have an unsupported sister group relationship to clade D (Fig. 1), whereas in the Bayesian analysis Stetsonia is sister to clade B*, and Cereus has a basal position in the ingroup.
DISCUSSION
Phylogenetic signal of the molecular markers
We chose the maternally inherited chloroplast markers to investigate the phylogeny of the study group because present knowledge does not imply any substantial influence of hybridization on the evolution within Cactaceae (Gibson and Nobel, 1986
). The number of informative nucleotide substitutions and insertion/deletion events was largest in the atpB-rbcL IGS. The trnK-rps16 IGS was employed for molecular phylogenetic analyses for the first time in this study, and although its length is comparable to the atpB-rbcL IGS, this marker has fewer base substitutions and fewer informative indels (see Table 1). Because of several very large indels, about two thirds of the marker was of limited use.
Homoplasy of morphological characters and polyphyly of the tribes
The ingroup (Figs. 1, 2) corresponds to the BCT clade of Nyffeler (2002)
, and its monophyly is highly supported. Furthermore, the ingroup genera share a large deletion in the chloroplast trnT-trnL IGS, present also in Neoraimondia herzogiana, which does not belong to the BCT clade (Applequist and Wallace, 2002
). None of the tribes Browningieae, Cereeae, and Trichocereeae is monophyletic, which is in accordance with the results of Nyffeler (2002)
.
The presence/absence of hairs or bristles on the pericarpels is homoplastic with respect to the whole tree and is therefore meaningless for the delimitation of tribes. However, this inconstancy disappears on a smaller scale in the molecular phylogeny, and hence the surface of the pericarpel in combination with presence/absence of cephalia and other morphological traits proves to be very useful to circumscribe several smaller groups (clades A–E in Fig. 1). Large columnar cacti and small globose species occur in all major clades, but this trait also becomes informative at lower taxonomic levels. This is the case for the globose genera Melocactus Link & Otto and Discocactus Pfeiff., which are the closest relatives to each other within a clade of otherwise columnar cacti. This also applies to the globose species of former Lobivia, now Echinopsis (E. ancistrophora, E. cinnabarina, E. pentlandii, and E. tiegeliana), which form a monophyletic group (Figs. 1, 2) within that genus.
Clade A: cephalia-bearing cacti with naked pericarpels
This strongly supported clade unites species from two tribes and six genera (Fig. 1), which are highly diverse with respect to growth habit and floral syndromes. The clade comprises tall columnar cacti such as Coleocephalocereus goebelianus; rather shrubby, climbing plants such as Arrojadoa rhodantha; and globose cacti such as Melocactus oreas or Discocactus zehntneri. Members of this clade have either brightly colored diurnal flowers as in the genus Melocactus, which is bird- or insect-pollinated, or have large white nocturnal flowers as in the genus Coleocephalocereus Backeb., which is probably bat-pollinated. Remarkably, these highly diverse genera have several traits in common: (1) They are mainly distributed in the caatinga vegetation in northeastern Brazil, except for Discocactus and Melocactus, which have a wider distribution range. (2) Pericarpels are naked except for few scales. (3) Cephalia are present except for in Cipocereus minensis, though they are differently developed in the various genera. In Melocactus and Discocactus, flowers emerge from terminal cephalia, which stop the vegetative growth at the stem tips; the temporary terminal cephalia of Arrojadoa Mattf. become circular after they are broken through by resumed vegetative growth, and Espostoopsis Buxb., Coleocephalocereus, and Micranthocereus Backeb. form lateral cephalia (Mauseth, 2006
). Cipocereus minensis has no cephalia, but other species of this genus produce conspicuous amounts of long trichomes at flower-bearing areoles. Moreover, Cipocereus F. Ritter was formerly included in the genus Pilosocereus Byles & G. D. Rowley, which shows all transitions from wooly areoles to densely hairy flowering zones, called pseudocephalia because the internal portions of the shoots are not affected (Mauseth, 2006
).
Cephalia are also present in Espostoahttp://Britton & Rose (clade B), but this genus is characterized by hairy pericarpels. The evolution of cephalia, which in the case of lateral cephalia is interpreted as an adaptation to bat-pollination, may thus have occurred more than once. Backeberg's (1977) classification of the cephalia-bearing genera into two informal taxonomic units, "Cephalocerei" (Arrojadoa,Coleocephalocereus,Espostoa, Micranthocereus, and others) and "Cephalocacti" (Melocactus and Discocactus), are essentially corroborated by the molecular phylogeny, except for the position of Espostoa. He further noted that the "Cephalocacti" had presumably evolved from the "Cephalocerei." The sister group relationship of Melocactus and Discocactus, both globose and solitary cacti with terminal cephalia, is strongly supported in our analysis and was also proposed by Backeberg (1977)
and Endler and Buxbaum (1974)
but later declined by Barthlott and Hunt (1993)
and Anderson (2005).
Clade B: Trichocereeae with hairy pericarpels
This clade comprises columnar and globose Andean cacti with pericarpels bearing hairy or bristly areoles. In the Bayesian analysis, the genus Gymnocalycium with scaly, hairless pericarpels is nested within this group; hence, clade B is absent in Fig. 2. In the parsimony analysis, the monophyly of clade B has only low support. This analysis includes insertion/deletion events that cannot be evaluated in the Bayesian approach but are very informative in phylogenetic analyses (Kelchner, 2000
). Because the clade receives additional support from morphology (hairy pericarpels), clade B is discussed in detail next.
Within clade B, lateral cephalia are characteristic for Espostoa. The close relationship of Espostoa with Echinopsis and Haageocereus Backeb. was already indicated by Anderson (2005)
based on anatomical similarities. Furthermore, a close relationship between the genera of clade B is indicated by the occurrence of natural hybrids between Espostoa and Haageocereus, Cleistocactus Lem. and Matucana Britton & Rose (described as xHaagespostoa G. D. Rowley), xEspostocactus Mottram, and xEspocana P. V. Heath, as well as hybrids between the genera Cleistocactus, Matucana, Oreocereus (A. Berger) Riccob., Oroya Britton & Rose, and Samaipaticereus Cárdenas (Rowley, 1994
). The genus Espostoa s.l. [i.e., including Espostoa guentheri = Vatricania guentheri (Kupper) Backeb.] is probably polyphyletic because E. guentheri is more closely related to Cleistocactus and Echinopsis huotii than to the other Espostoa species.
Zygomorphic flowers occur in several genera (Cleistocactus, Denmoza Britton & Rose, Matucana, Oreocereus). Actinomorphic flowers are plesiomorphic within the South American cacti investigated here and generally within the Cactaceae (Griffith, 2004
). The previously mentioned genera with zygomorphic flowers do not constitute a monophyletic group, and thus the transition from actinomorphic to zygomorphic flowers occurred repeatedly or multiple reversions took place.
Echinopsis s.l
Echinopsis s.l. includes Trichocereus, Echinopsis s.s., Setiechinopsis, and Lobivia (Barthlott and Hunt, 1993
; Hunt 1999
; Anderson, 2005
; Hunt, 2006
) with ca. 50–100 total species (Barthlott and Hunt, 1993
). Lobivia, represented in our analysis by Echinopsis ancistrophora, E. cinnabarina, E. pentlandii, and E. tiegeliana, is the only well-supported group within Echinopsis s.l. Our taxon sampling of Echinopsis s.l. is not sufficient to test the monophyly of Lobivia. However, our results allow us to reject the hypotheses of Endler and Buxbaum (1974)
and Backeberg (1977) that Lobivia is more closely related to at least some species of Rebutia than to Echinopsis s.s and Trichocereus. These authors inferred a close relationship between Rebutia and Lobivia from the shared pollination syndrome (diurnal anthesis and short, mostly scentless, brightly colored flowers). In contrast, Echinopsis s.s is adapted to moth-pollination, possessing nocturnal, elongated, mostly scented white flowers. The same pollinator adaptations exist in distantly related cacti. Because these adaptations are highly homoplastic in Cactaceae, their value in phylogenetic analyses is limited.
Clade C: Gymnocalycium
Endler and Buxbaum (1974)
, Backeberg (1977)
, Augustin and Hentzschel (2002)
, and Ritter (1980)
inferred a close relationship between Gymnocalycium and Weingartia because both genera have hairless, scaly pericarpels. Molecular data clearly refute this hypothesis because Gymnocalycium is sister of clade B or part of B* (Figs. 1, 2), comprising various genera of the Trichocereeae with hairy or bristly pericarpels, and thus the similarities of Gymnocalycium and Weingartia are not based on a shared inheritance. Unfortunately, the high resolution within Gymnocalycium in our analysis is not marked by morphological characters. Gymnocalycium is subdivided into two subgenera based on fruit type and seed morphology (Till, 2001
), but these subgenera are not monophyletic in our analysis: one clade contains G. anisitsii, G. schickendantzii, and G. pflanzii, which belong to the subgenus Microsemineum Schütz. The other clade comprises species of the subgenera Microsemineum and Gymnocalycium. We did not find any obvious correlation of the phylogenetic relationships defined by the molecular characters and geographic distribution, because the Andean G. bruchii is the closest relative to G. rauschii found in Uruguay. On the contrary, the morphological classification correlates with the geographic distribution of the species (Till, 2001
). A possible explanation for the discrepancy of morphological and molecular characters is convergent evolution of species living in the same areas. It should be noted, however, that the sampling of the genus was rather small in our analysis, and we did not cover the entire range.
Clade D: Rebutia I
Barthlott and Hunt (1993)
, Hunt (1999)
, Anderson (2001)
, and Hunt (2006) merged Rebutia, Sulcorebutia, and Weingartia into Rebutia s.l. without discussing any morphological characters that would justify this decision. Rebutia in the sense of these authors is polyphyletic in our trees (Figs. 1, 2), which is in agreement with the classification of Backeberg (1977)
, who split Rebutia in the sense of Krainz (1967)
in three genera: Aylostera, Mediolobivia, and Rebutia. Rebutia, represented here by two species with hairless pericarpels (Rebutia II, comprising the generic type R. minuscula and R. padcayensis), constitutes together with Sulcorebutia and Weingartia a part of clade E. The remaining species, all characterized by hairy and bristly pericarpels (Rebutia I), constitute clade D. Within this clade, three subgroups can be identified: (1) R. pseudodeminuta, R. fiebrigii, and R. deminuta (corresponding to the genus Aylostera in Backeberg's classification); (2) R. pygmaea and R. steinmannii; and (3) R. einsteinii. Subgroups (2) and (3) together correspond to the genus Mediolobivia in Backeberg's classification. However, present data do not show whether Mediolobivia is paraphyletic or monophyletic (Figs. 1, 2). The patterns of geographic distribution of the species of Rebutia s.s. do not correspond either to the Rebutia I/Rebutia II split or to the subgroups within Rebutia I (Fig. 3). We propose that the Rebutia species with hairy and bristly pericarpels (Rebutia I) should be excluded from that genus.
|
, Parodia Speg., or Weingartia (Kníe and íha, 1995
).
The relationship between Sulcorebutia and Weingartia is not resolved, preventing confirmation of the monophyly of the genera. Endler and Buxbaum (1974)
, Backeberg (1977)
, and Ritter (1980)
assumed that Sulcorebutia is closely related to Rebutia s.s. and that Weingartia is associated with Gymnocalycium. Our results confirm the hypothesis of Augustin et al. (2000)
and Augustin and Hentzschel (2002)
, who advocated a close relationship between Sulcorebutia and Weingartia. Several morphological characters, e.g., the shape of the areoles and the branching of the funiculi, are usually cited to distinguish the two genera (Augustin et al., 2000
). However, transitions between the character states are gradual. The entire Sulcorebutia-Weingartia-Cintia complex is well characterized by hairless pericarpels with persistent auriculate scales as opposed to the deciduous, acute, triangular scales characteristic for Rebutia (Augustin et al., 2000
). The unsupported bipartition in the Bayesian analysis of the Sulcorebutia-Weingartia clade (clades E1 and E2 in Fig. 2) roughly reflects the geographic distribution of the species (Fig. 4). The species of clade E1 are distributed from the central to the northern part of the range covered by the entire Sulcorebutia-Weingartia-Cintia complex. Except for S. arenacea, all species of the clade E2 are distributed from the central to the southern part of that range.
|
as a synonym of Rebutia (alias Weingartia) neocumingii, but in the German edition of his work, translated and supplemented by Eggli (Anderson, 2005
), it was treated simply as Weingartia sp. Our data suggest that Weingartia buiningiana is close to or even belongs in Weingartia neocumingii.
Browningia
The genus Browningia s.l. includes the former genera Azureocereus Akers & H. Johnson, Castellanosia Cárdenas, Browningia, Gymnanthocereus Backeb., and Gymnocereus Backeb. (Barthlott and Hunt, 1993
; Hunt, 1999
, 2006
; Anderson, 2005
). In the phylogenetic trees (Figs. 1, 2), the type species of the genus, B. candelaris, is not sister to B. hertlingiana of the former Azureocereus. Instead, it is sister to Rebutia II. Our results and those of Nyffeler (2002)
, although based on a different taxon sampling, suggest that the genus Browningia s.l. is not monophyletic.
Biogeography
Uebelmannia pectinifera, as well as the species of clade A, are distributed in the mountainous region of northeastern Brazil (Bahia and Minas Gerais), whereas the vast majority of taxa in the clades B to E are mainly distributed in high altitudes at the eastern slopes of the central Andes. The basal positions of Uebelmannia Buining and clade A point to a northeastern Brazilian origin of the BCT clade. This suggests that South American cacti would have remigrated to the Andes, because the closest relatives of the BCT clade also have an Andean distribution (Nyffeler, 2002
) and the subfamily Cactoideae presumably originated in the central Andes 30–20 million years ago (Hershkowitz and Zimmer, 1997
; Nyffeler, 2002
; Edwards et al., 2005
). However, taking into account that the basal position of the northeastern Brazilian species is weakly supported and several independent colonization events of South American cacti in the central Andes are not parsimonious, we favor the hypothesis that the BCT clade originated in the Andes. Subsequently, the early ancestors of Uebelmannia and the species of clade A might have migrated from the Andes to northeastern Brazil. The present separation of centers of diversity in the Andean and in the northeast Brazilian part of the distribution area of the BCT clade may be connected to the Miocene uplift of the Andes causing a series of foreland basins along the eastern foothills (Horton and DeCelles, 1997
). The global sea level rise in the Middle to Late Miocene led to one or more marine incursions between 14–7 million years ago, possibly interconnecting the Amazonian, Bolivian, and Paraguayan Chaco basins (Räsänen et al., 1995
; Webb, 1995
; Hernández et al., 2005
; Hulka et al., 2006
) (Fig. 5). Whether or not this band of flooded basins created an intracontinental seaway, these widespread wetlands, shallow marine environments, and humid to semiarid floodplains separated the Andean and northeastern Brazilian distribution areas for several million years on a large geographical scale. Falling sea levels and a cooler and drier climate during the Quaternary favored the extension of the neotropical, seasonally dry forest (Pennington et al., 2000
). Uebelmannia and cacti of clade A are constituents of the caatinga vegetation, which is part of the neotropical, seasonally dry forests (see Fig. 5). These forests were replaced by the cerrado vegetation after the last glacial period (Pennington et al., 2000
). The cerrado is an environment almost unconquerable for cacti because of the regular fires. Only a few species of Discocactus that have morphological adaptations to fire occur in the cerrado (Hunt, 2006
). The other species of clade A are restricted to the refugial areas of the caatingas. Pennington et al. (2004)
performed an evolutionary rate analysis of several genera of the neotropical, seasonally dry forest to support the hypothesis that Quaternary climate changes triggered the speciation in these plant genera. Although his estimates of diversification times were of pre-Pleistocene age (8–20 Ma), they coincide with the time of the largest expansion of the foreland basins mentioned earlier. Nevertheless, the regression of the neotropical, seasonally dry forests resulting from Quaternary climate changes might be responsible for maintaining the gap between both diversification centers of South American cacti.
|
. The present distribution patterns may be a result of range fragmentation because of Quaternary climate changes. In contrast, branches within the Sulcorebutia-Weingartia-Cintia clade are short (Figs. 1, 2), and the distribution range is restricted to a rather small area in the central Andes (Fig. 4). These facts suggest a recent radiation of the group that might be comparable to rapid speciation in other Andean taxa (Bell and Donoghue, 2005
; Hughes and Eastwood, 2006
). According to these authors, the extensive topographical and environmental heterogeneity in island-like habitats after the Andean uplift has promoted rapid radiation. It would be very desirable to date the clades using a molecular clock approach, but this has been infeasible because of a lack of fossils and very heterogeneous evolution rates of the molecular markers within Cactoideae.
Conclusions
This study, exploring the phylogenetic relationships of South American cacti, contributes to the knowledge regarding how endemism and species richness evolved in the Andes mountains, which are a hotspot of biodiversity on earth. We suggest that the Andean uplift influences the speciation of the Andean cacti itself, and because of the marine incursions into the continent caused by the orogenesis, the uplift is also responsible for the separation of the two centers of South American cactus diversification: northeastern Brazil and the central Andes. The molecular phylogenetic tree reflects the geographic disjunction between the cephalia-bearing cacti with naked pericarpels and Uebelmannia, which occur in northeastern Brazil, and all other ingroup species, which are centered in the central Andes. The data also suggest that the radiation of Gymnocalycium is older than the radiation of the Sulcorebutia-Weingartia-Cintia complex, which is similar in size and morphological diversity. The current morphological classification within Cactaceae is not supported by the results of our molecular phylogenetic analyses. Many morphological traits that are used for classification, e.g., pollination syndromes and the presence of cephalia, are highly homoplastic and not suited for the definition of monophyletic groups. The results of our molecular phylogenetic analyses do not support either rigorous and schematic lumping of genera, as was done in recent classifications of Cactaceae, or extensive splitting. Our study reveals that Rebutia species with hairy pericarpels are rather distantly related to Rebutia species with naked pericarpels, so in this case, a narrow generic concept as proposed by Backeberg (1977)
better approximates a natural system. In contrast, Sulcorebutia and Weingartia should be united into one genus, because neither molecular nor morphological data reveal a distinction between these genera.
APPENDIX
Taxon—GenBank accessions: atpB-rbcL intergenic spacer (IGS), trnL-trnF IGS, rps16-trnK IGS; source, if the locality is unknown, the distribution of the species is given; (Common synonym); Voucher number.
Arrojadoa rhodantha (Gürke) Britton & Rose—AM502345, AM502513, AM502514; distrib.: Brazil, Bahia, NW Caitité, cult. Haage; —; CA38. —a, AM502515, AM502516; cult. BGJ; —; CA83.
Austrocactus philippii (Regel) Buxb. & F. Ritter— —a, AM502517, AM502518; distrib.: Chile, Cordillera de Maule, cult. Knebel; —; CA109.
Browningia candelaris (Meyen) Britton & Rose—AM502346, AM502519, AM502520; distrib.: South-Peru to North-Chile, cult. Haage; —; CA53. B. hertlingiana (Backeb.) Buxb.—AM502347, AY0153621, AM502521; distrib.: Peru, Huancavelica, Mantaro Valley, cult. BGJ; —; CA84.
Cereus hildmannianus K. Schum.—AM502348, AM502522, AM502523; distrib.: Brazil, Uruguay, Paraguay, cult. Uhlig; —; CA113.
Cintia knizei iha—AM502349, AM502350, AM502351; Bolivia, Chuquisaca, west of Santa Elena, cult. SSK; —; CA01. AM502352, AM502353, AM502354; distrib.: Bolivia, Chuquisaca, Padcoyo, cult. SSK; —; CA13.
Cipocereus minensis (Werderm.) F. Ritter—AM502524, AM502525, AM502526; distrib.: Brazil, Minas Gerais, Barao Guaiqucui, cult. BGJ; —; CA115.
Cleistocactus strausii (Heese) Backeb.—AM502527, AM502528, AM502529; distrib.: Argentina & Bolivia, cult. Haage; —; CA46.
Coleocephalocereus fluminensis (Miq.) Backeb.—AM502533, AY015364b, AM502534; distrib.: Brazil, Minas Gerais, cult. Haage; —; CA40. C. goebelianus (Vaupel) Buining—AM502530, AM502531, AM502532; distrib.: Brazil, Bahia, cult. Köhres; —; CA70.
Copiapoa laui Diers—AM502535, AM502536, AM502537; distrib.: Chile, Esmeralda, cult. Köhres; —; CA73.
Denmoza rhodacantha (Salm-Dyck) Britton & Rose—AM502538, AM502539, AM502540; distrib.: Argentina, Mendoza, La Rioja, San Juan & Salta, cult. Haage; —; CA47.
Discocactus zehntneri Britton & Rose subsp. boomianus (Buining & Brederoo) N.P. Taylor & Zappi—AM502541, AM502542, AM502543; distrib.: Brazil, Bahia, cult. Haage; —; CA51.
Echinopsis ancistrophora Speg. subsp. arachnacantha (Buining & F. Ritter) Rausch—AM502405, AM502406, AM502407; Bolivia, Santa Cruz, Vallegrande, cult. SSK; ("Lobivia arachnacantha"); CA04. E. atacamensis (Phil.) H. Friedrich & G.D. Rowley subsp. pasacana (F.A.C. Weber) G. Navarro—AM502408, AM502409, AM502410; Bolivia, Potosí, Salar de Uyuni, Incahuasi, cult. SSK; ("Trichocereus pasacana"); CA31. E. cinnabarina (Hook.) Labour.—AM502355, AM502356, AM502357; distrib.: Bolivia, Chuquisaca, Zudañez, cult. SSK; ("Lobivia cinnabarina"); CA20. E. huotii Labour.— —a, AM502403, AM502404; Bolivia, Santa Cruz, El Trigal, cult. SSK; —; CA02. E. mamillosa Gürke—AM502432, AM502433, AM502434; Bolivia, Tarija, near Río Pilaya, cult. SSK; —; CA14. E. mirabilis Speg.—AM502544, AM502545, AM502546; distrib.: Argentina, Santiago del Estero, cult. Haage; ("Setiechinopsis mirabilis"); CA74. E. pentlandii (Hook.) Salm-Dyck—AM502441, AM502442, AM502443; Bolivia, Oruro, Vila Vila, cult. SSK; ("Lobivia pentlandii"); CA21. E. schickendantzii F.A.C. Weber—AM502467, AM502468, AM502469; Argentina, Tucuman, Trancas, cult. SSK; ("Trichocereus schickendantzii"); CA11. E. tiegeliana (Wessner) D.R. Hunt—AM502444, AM502445, AM502446; Bolivia, Tarija, Angostura, cult. SSK; ("Lobivia tiegeliana"); CA22.
Eriosyce napina (Phil.) Katt.— —a, AY015384b, AM502547; distrib.: Chile, Atacama, Río Huasco Valley, cult. BGJ; —; CA85.
Espostoa guentheri (Kupper) Eggli—AM502548, AM502549, AM502550; distrib.: Bolivia, cult. BGJ; ("Vatricania guentheri"); CA112. E. lanata (Kunth) Britton & Rose—AM502551, AM502552, AM502553; distrib.: Peru, North, cult. Haage; —; CA43. E. ritteri Buining—AM502554, AM502555, AM502556; distrib.: Peru, Marañón Valley, cult. BGJ; —; CA111.
Espostoopsis dybowskii (Rol.-Goss.) Buxb.—AM502557, AM502558, AM502559; distrib.: Brazil, Bahia, Itumirin, cult. Haage; —; CA44.
Gymnocalycium andreae (Boed.) Backeb. & F.M. Knuth subsp. carolinense G. Neuhuber—AM502492, AM502493, AM502494; Argentina, San Luis, Carolina, cult. SSK; —; CA89. G. anisitsii(K.Schum.) Britton & Rose subsp. anisitsii—AM502400, AM502401, AM502402; Bolivia, Santa Cruz, Motacusito, cult. SSK; ("G. anisitsii subsp. holdii"); CA15. G. bruchii (Speg.) Hosseus—AM502489, AM502490, AM502491; Argentina, Córdoba, Cruz del Eje, cult. SSK; ("G. bruchii var. niveum"); CA03. G. carminanthum Borth & Koop—AM502495, AM502496, AM502497; Argentina, Catamarca, Ambato, cult. SSK; —; CA88. G. leptanthum (Speg.) Speg.—AM502486, AM502487, AM502488; Argentina, Córdoba, San Pedro del Norte, Sierra de Ambargasta, cult. SSK; —; CA16. G. paraguayense (K.Schum.) Hosseus—AM502498, AM502499, AM502500; Paraguay, Cordillera, Caacupé, cult. SSK; —; CA91. G. pflanzii (Vaupel) Werderm. subsp. argentinense H. Till & W. Till—AM502483, AM502484, AM502485; Argentina, Salta, Juramento, cult. SSK; —; CA19. G. pflanzii (Vaupel) Werderm. subsp. pflanzii—AM502501, AM502502, AM502503; Paraguay, Boquerón, northwest of Mariscal Estigarribia, cult. SSK; —; CA90. G. rauschii H. Till & W. Till—AM502504, AM502505, AM502506; Uruguay, Tacuarembó, Ansina, cult. SSK; —; CA17. G. schickendantzii(F.A.C. Weber) Britton & Rose— —a, AM502481, AM502482; Argentina, Salta, Balboa, cult. SSK; —; CA18.
Haageocereus pacalaensis Backeb.—AM502560, AM502561, AM502562; distrib.: Peru, between Trujillo und Chimbote, cult. Haage; —; CA42.
Matucana aurantiaca(Vaupel) Buxb.—AM502563, AM502564, AM502565; distrib.: Peru, Río Fortaleza, cult. Haage; —; CA49.
Melocactus oreas Miq.—AM502566, AM502567, AM502568; distrib.: Brazil, Bahia, Río de Contas, cult. Haage; —; CA41.
Micranthocereus auriazureus Buining & Brederoo—AM502569, AM502570, AM502571; distrib.: Brazil, Minas Gerais, Grao Mogol, cult. Haage; —; CA39.
Neoraimondia arequipensis (Meyen) Backeb. subsp. roseiflora (Werderm. & Backeb.) Ostolaza— —a, AM502572, AM502573; distrib.: Peru, near Chosica and in the Pisco Valley; cult. Haage; —; CA52. N. herzogiana (Backeb.) Buxb— —a, AM502574, AM502575; distrib.: Bolivia, cult. BGJ; —; CA108.
Neowerdermannia vorwerkii Fri—AM502447, AM502448, AM502449; Bolivia, La Paz, Tambillo, cult. SSK; —; CA05.
Oreocereus celsianus (Lem. ex Salm-Dyck) Riccob.—AM502576, AY015373b, AM502577; distrib.: Bolivia, South-Peru, North-Chile, cult. Haage; —; CA48.
Oroya peruviana (K.Schum.) Britton & Rose—AM502578, AM502579, AM502580; distrib.: Peru, north of Lima, cult. Haage; —; CA50.
Parodia magnifica (F. Ritter) F.H. Brandt—AM502507, AY015377a, AM502508; Brazil, Rio Grande do Sul, cult. SSK; —; CA55. P. microsperma (F.A.C. Weber) Speg.—AM502509, AM502510, —a; Brazil, Río Grande do Sul, cult. SSK; —; CA58.
Rauhocereus riosaniensis Backeb.—AM502511, AY015372, AM502512; North-Peru, between Chamaya and Jaen, cult. Uhlig; —; CA71.
Rebutia deminuta (F.A.C.Weber) Britton & Rose—AM502476, AM502477, AM502478; Argentina, Salta, Que. de Escoipe, cult. SSK; —; CA07. R. einsteinii Fri—AM502464, AM502465, AM502466; Argentina, Jujuy, Que. de Morado, cult. SSK; —; CA26. R. fiebrigii (Gürke) Britton & Rose—AM502435, AM502436, AM502437; Bolivia, Tarija, Cuesta de Sama, cult. SSK; —; CA25. R. minuscula K. Schum.—AM502470, AM502471, AM502472; Argentina, Tucumán, San Pedro de Colalao, cult. SSK; —; CA24. R. padcayensis Rausch—AM502473, AM502474, AM502475; Argentina, Salta, Sta. Victoria, cult. SSK; ("R. margarethae"); CA06. R. pseudodeminuta Backeb.— —a, AM502479, AM502480; Argentina, Salta, Que. del Toro, cult. SSK; —; CA94. R. pygmaea (R.E.Fr.) Britton & Rose—AM502459, AM502460, AM502461; Argentina, Jujuy, Tafna, cult. SSK; ("R. haagei"); CA08. —a, AM502462, AM502463; Argentina, Jujuy, Yavi, cult. SSK; —; CA93. R. steinmannii (Solms) Britton & Rose—AM502411, AM502412, AM502413; Bolivia, Potosí, Chaqui, cult. SSK; —; CA92.
Samaipaticereus corroanus Cárdenas—AM502581, AM502582, AM502583; distrib.: Bolivia, Florida, cult. BGJ; —; CA82.
Stetsonia coryne (Salm-Dyck) Britton & Rose—AM502584, AY015366b, AM502585; distrib.: Argentina, Bolivia, Paraguay, cult. Haage; —; CA54.
Sulcorebutia arenacea (Cárdenas) F. Ritter—AM502420, AM502421, AM502422; Bolivia, Cochabamba, Cocapata, cult. SSK; ("S. menesesii"); CA98. S. canigueralii (Cárdenas) Buining & Donald—AM502358, AM502359, AM502360; Bolivia, Chuquisaca, north of Tarabuco, cult. SSK; ("S. tarabucoensis"); CA27. AM502361, AM502362, AM502363; Bolivia, Chuquisaca, Tarabuco, cult. SSK; ("S. tarabucoensis"); CA103; AM502364, AM502365, AM502366; Bolivia, Chuquisaca, Sucre, cult. SSK; —; CA104. S. cardenasiana R.Vásquez—AM502397, AM502398, AM502399; Bolivia, Santa Cruz, Vallegrande, cult. SSK; ("S. langeri"); CA102. S. crispata Rausch—AM502391, AM502392, AM502393; Bolivia, Santa Cruz, Pucara, cult. SSK; —; CA09. AM502394, AM502395, AM502396; Bolivia, Santa Cruz, Pucara, cult. SSK; —; CA106. S. mentosa F.Ritter—AM502414, AM502415, AM502416; Bolivia, Potosí, Torotoro, cult. SSK; ("S. torotorensis"); CA30. S. purpurea (Donald & A.B. Lau) Brederoo & Donald—AM502423, AM502424, AM502425; Bolivia, Cochabamba, Aiquile, Cuesta de Santiago, cult. SSK; ("S. santiaginensis"); CA10. S. steinbachii (Werderm.) Backeb.—AM502426, AM502427, AM502428; Bolivia, Cochabamba, Colomi, cult. SSK; —; CA107. AM502429, AM502430, AM502431; Bolivia, Cochabamba, Rancho Tojo Tojo, cult. SSK; ("S. totorensis"); CA28; AM502388, AM502389, AM502390; Bolivia, Santa Cruz, Comarapa, cult. SSK; ("S. krahnii"); CA29. S. tarijensis F. Ritter—AM502438, AM502439, AM502440; Bolivia, Tarija, Cuesta de Sama, cult. SSK; —; CA100. S.sp.—AM502367, AM502368, AM502369; Bolivia, Chuquisaca, Azurduy, cult. SSK; —; CA101. S.sp.—AM502370, AM502371, AM502372; Bolivia, Chuquisaca, south of El Puente; cult. SSK; —; CA99.
Uebelmannia pectinifera Buining subsp. flavispina (Buining & Brederoo) P.J. Braun & Esteves—AM502586, AY015365b, AM502587; distrib.: Brazil, Minas Gerais; cult. BGJ; —; CA87.
Weingartia buiningiana Ritter—AM502450, AM502451, AM502452; Bolivia, Chuquisaca, Río Pilcomayo, cult. SSK; —; CA96. W. cintiensis Cárdenas—AM502417, AM502418, AM502419; Bolivia, Potosí, Calapaya, cult. SSK; —; CA33. W. fidaiana (Backeb.) Werderm.—AM502456, AM502457, AM502458; Argentina, Jujuy, El Moreno, cult. SSK; ("W. neumanniana"); CA36. W. neocumingii Backeb. subsp. neocumingii var. hediniana (Backeb.) K. Augustin & Hentzschel—AM502373, AM502374, AM502375; Bolivia, Chuquisaca, Icla, cult. SSK; —; CA12. W. neocumingii Backeb. subsp. neocumingii var. longigibba (F. Ritter) K. Augustin & Hentzschel—AM502376, AM502377, AM502378; Bolivia, Chuquisaca, Río Chico, cult. SSK; —; CA97. W. neocumingii Backeb. var. neocumingii—AM502379, AM502380, AM502381; Bolivia, Chuquisaca, Sucre, cult. SSK; ("W. trolli"); CA34. W. neocumingii Backeb. subsp. pulquinensis (Cárdenas) Donald—AM502385, AM502386, AM502387; Bolivia, Santa Cruz, Pampa Grande, cult. SSK; —; CA35. W. westii (Hutchison) Donald—AM502382, AM502383, AM502384; Bolivia, Chuquisaca, Padcoyo, cult. SSK; —; CA32.
Yavia cryptocarpa R. Kiesling & Piltz—AM502453, AM502454, AM502455; Argentina, Jujuy, La Quiaca, cult. SSK; —; CA59.
FOOTNOTES
102 b Sequences were not produced during this study but taken from GenBank. 
1 This analysis was initiated by the research group Studiengemeinschaft Südamerikanische Kakteen e.V. (SSK) and partially conducted within the SSK Project 2005, which was financially sponsored by the Freistaat Thüringen, Germany. The members of this study group, H. Amerhauser (Eugendorf, Austria), H. Erritzoe (Taps, Denmark), K. Fickenscher (Marburg, Germany), W. Gertel (Ingelheim, Germany), H. Jucker (Teufen, Switzerland), K. Köhler (Waltershausen, Germany), R. Märtin (Jena, Germany), R. Mecklenburg (Langstedt, Germany), J. Mortensen (Kolding, Denmark), J. de Vries (Vierpolders, the Netherlands), and R. Wahl (Limburg, Germany), provided specimens and useful hints in many fruitful discussions. The authors thank the cactus breeders Köhres (Erzhausen, Germany), Piltz (Düren, Germany), and Karl-Heinz Knebel (Bad Ems, Germany) for providing specimens; M. Sandmann for her lab assistance; C. Löser for help with analyses; C. Fehringer for cultivating the cacti and for valuable background information; J. Müller and R. Nyffeler for valuable discussions; and anonymous reviewers for comments on the manuscript.
4 Author for correspondence (christiane.ritz{at}uni-jena.de
)
5 The first two authors contributed equally to the publication.
LITERATURE CITED
Anderson E. F.. 2001. The cactus family Timber Press, Portland, Oregon, USA.
Anderson E. F.. 2005. Das große Kakteenlexikon Eugen Ulmer, Stuttgart, Germany.
Applequist W. L. Wallace R. S.. 2002. Deletions in the plastid trnT-trnL intergenic spacer define clades within Cactaceae subfamily Cactoideae. Plant Systematics and Evolution 231: 153-162.[CrossRef][Web of Science]
Augustin K. Gertel W. Hentzschel G.. 2000. Sulcorebutia Eugen Ulmer, Stuttgart, Germany.
Augustin K. Hentzschel G.. 2002. Die Gattung Weingartia Werdermann. Gymnocalycium 15: 7453-472.
Backeberg C.. 1977. Das Kakteenlexikon, 4th ed Gustav Fischer Verlag, Jena, Germany.
Baldwin B. G. Sanderson M. J. Porter M. J. Wojciechowski M. F. Campbell C. S. Donoghue M. J.. 1995. The ITS region of nuclear ribosomal DNA: a valuable source of evidence on angiosperm phylogeny. Annals of the Missouri Botanical Garden 82: 247-277.[CrossRef][Web of Science]
Barthlott W. Hunt D. R.. 1993. Cactaceae. In K. Kubitzki, J. G. Rohwer, and V. Bittrich [eds.], The families and genera of vascular plants 161-197 Springer Verlag, Berlin, Germany.
Bell C. D. Donoghue M. J.. 2005. Phylogeny and biogeography of Valerianaceae (Dipsacales) with special reference to the South American valerians. Organisms, Diversity and Evolution 5: 147-159.[CrossRef]
Bremer K.. 1994. Branch support and tree stability. Cladistics 10: 295-304.[CrossRef][Web of Science]
Brinkmann K.-H.. 1976. Die Gatttung Sulcorebutia Steinhart KG, Titisee-Neustadt, Germany.
Buining A. F. Donald J. D.. 1965. The revision of the genus Rebutia K.Schumann. Cactus and Succulent Journal of Great Britain 27: 36-41.
Buining A. F. H. Donald J. D.. 1963. Die Gattung Rebutia K.Schumann. Sukkulentenkunde 7/8: 96-107.
Butterworth C. A. Cota-Sanchez J. H. Wallace R. S.. 2002. Molecular systematics of tribe Cacteae (Cactaceae: Cactoideae): a phylogeny based on rpl16 intron sequence variation. Systematic Botany 27: 257-270.[Web of Science][Medline]
Butterworth C. A. Wallace R. S.. 2004. Phylogenetic studies of Mammillaria (Cactaceae)—insights from chloroplast sequence variation and hypothesis testing using the parametric bootstrap. American Journal of Botany 91: 1086-1098.
Edwards E. J. Nyffeler R. Donoghue M. J.. 2005. Basal cactus phylogeny: implications of Pereskia (Cactaceae) paraphyly for the transition to the cactus life form. American Journal of Botany 92: 1177-1188.
Endler J. Buxbaum F.. 1974. Die Pflanzenfamilie der Kakteen, 3rd ed Albrecht Philler Verlag, Minden, Germany.
Eriksson T. Wikström N.. 1995. AutoDecay 3.0 Stockholm University, Stockholm, Sweden.
Farris J. S. Källersjö M. Kluge A. G. Bult C.. 1995. Testing significance of incongruence. Cladistics 10: 315-319.[CrossRef][Web of Science]
Felsenstein J.. 1985. Confidence-limits on phylogenies—an approach using the bootstrap. Evolution 39: 783-791.[CrossRef][Web of Science]
Gibson A. C. Nobel P. S.. 1986. The cactus primer Harvard University Press, London, UK.
Griffith M. P.. 2002. Phylogenetic relationships in the Opuntioideae (Cactaceae) based on nrITS sequences. International Organization for Succulent Plant Study Bulletin 10: 15-16.
Griffith M. P.. 2004. What did the first cactus look like? An attempt to reconcile the morphological and molecular evidence. Taxon 53: 493-499.[CrossRef][Web of Science]
Harpke D. Peterson A.. 2006. Non-concerted ITS evolution in Mammillaria (Cactaceae). Molecular Phylogenetics and Evolution 41: 579-593.[CrossRef][Web of Science][Medline]
Hartmann S. Nason J. D. Bhattacharya D.. 2001. Extensive ribosomal DNA genic variation in the columnar cactus Lophocereus. Journal of Molecular Evolution 53: 124-134.[Web of Science][Medline]
Hernández R. M. Jordan T. E. Farjat A. D. Echavarría L. Idleman B. D. Reynolds J. H.. 2005. Age, distribution, tectonics, and eustatic controls of the Paranense and Caribbean marine transgressions in southern Bolivia and Argentina. Journal of South American Earth Science 19: 495-512.[CrossRef]
Hershkowitz M. A. Zimmer E. A.. 1997. On the evolutionary origins of the cacti. Taxon 46: 217-232.[CrossRef][Web of Science]
Horton B. K. DeCelles P. G.. 1997. The modern foreland basin system adjacent to the central Andes. Geology 25: 895-898.
Huelsenbeck J. P. Larget B. Miller R. E. Ronquist F.. 2002. Potential applications and pitfalls of Bayesian inference of phylogeny. Systematic Biology 51: 673-688.[CrossRef][Web of Science][Medline]
Huelsenbeck J. P. Ronquist F.. 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17: 754-755.
Hughes C. Eastwood R.. 2006. Island radiation on a continental scale. Exceptional rates of plant diversification after uplift of the Andes. Proceedings of the National Academy of Sciences, USA 103: 10334-10339.
Hulka C. Gräfe K.-U. Sames B. Uba C. E. Heubeck C.. 2006. Depositional settings of the Middle to Late Miocene Yecua formation of the Chaco foreland basin, southern Bolivia. Journal of South American Earth Sciences 21: 135-150.
Hunt D.. 1999. CITES Cactaceae checklist Royal Botanic Gardens Kew, Richmond, UK.
Hunt D.. 2006. The new cactus lexicon dh Books, Milborne Port, UK.
Kelchner S. A.. 2000. The evolution of non-coding chloroplast DNA and its application in plant systematics. Annals of the Missouri Botanical Garden 87: 482-498.[CrossRef][Web of Science]
Kníe K. íha J.. 1995. Cintia Kníe & íha—nov
kaktus st
ední Bolívie. Kaktusy 2: 35-39.
Krainz H.. 1967. Die Kakteen: eine Gesamtdarstellung der eingeführten Arten nebst Anzucht- und Pflege-Anweisungen Franckh'sche Verlagshandlung, Stuttgart, Germany.
Mauseth J. D.. 2006. Structure–function relationships in highly modified shoots of Cactaceae. Annals of Botany 98: 901-926.
Myers N. Mittermeier R. A. Mittermeier C. G. da Fonseca G. A. B. Kent J.. 2000. Biodiversity hotspots for conservation priorities. Nature 403: 853-858.[CrossRef][Medline]
Nores M.. 1999. An alternative hypothesis for the origin of Amazonian bird diversity. Journal of Biogeography 26: 475-485.[CrossRef][Web of Science]
Nyffeler R.. 2002. Phylogenetic relationships in the cactus family (Cactaceae) based on evidence from trnK/matK and trnL-trnF sequences. American Journal of Botany 89: 312-326.
Nylander J. A. A.. 2004. MrModeltest, version 1.0b Department of Systematic Zoology, EBC, Uppsala University, Uppsala, Sweden.
Pennington R. T. Lavin M. L. Prado D. E. Pendry C. A. Pell S. K. Butterworth C. A.. 2004. Historical climate change and speciation: neotropical seasonally dry forest plants show patterns of both Tertiary and Quarternary diversification. Philosophical Transactions of the Royal Society, B, Biological Sciences 359: 515-537.[CrossRef]
Pennington R. T. Prado D. E. Pendry C. A.. 2000. Neotropical seasonally dry forests and Quaternary vegetation changes. Journal of Biogeography 27: 261-273.[CrossRef][Web of Science]
Räsänen M. E. Linna A. M. Santos J. C. R. Negri F. R.. 1995. Late Miocene tidal deposits in the Amazonian foreland basin. Science 269: 386-390.
Ritter F.. 1980. Kakteen in Südamerika Friedrich Ritter Selbstverlag, Spangenberg, Germany.
Rowley G. D.. 1994. Spontaneous bigeneric hybrids in Cactaceae. Bradleya 12: 2-7.
Savolainen V. Manen J. F. Douzery E. Spichiger R.. 1994. Molecular phylogeny of families related to Celastrales based on rbcL 5' flanking sequences. Molecular Phylogenetics and Evolution 3: 27-37.[CrossRef][Medline]
Schumann K.. 1898. Gesamtbeschreibung der Cactaceae im Verhältnis zu ihrer systematischen Gliederung Verlag J. Neumann, Neudamm, Germany.
Swofford D. L.. 2002. PAUP*: phylogenetic analyses using parsimony (*and other methods), version 4.0b10 Sinauer, Sunderland, Massachusetts, USA.
Taberlet P. Gielly L. Pautou G. Bouvet J.. 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105-1109.[CrossRef][Web of Science][Medline]
Till H.. 2001. Neuordnung der Gattung Gymnocalycium. Gymnocalycium 14: 385-404.
Wallace R. S.. 1995a. A family-wide phylogeny, subfamiliar and tribal relationships, and suggestions for taxonomic realignments. IOS Bulletin 6: 13.
Wallace R. S.. 1995b. Molecular systematic study of the Cactaceae: using chloroplast DNA variation to elucidate cactus phylogeny. Bradleya 13: 1-12.
Wallace R. S. Dieckie S. L.. 2002. Systematic implication of chloroplast DNA sequence variation in subfam. Opuntioideae (Cactaceae). Succulent Plant Research 6: 9-24.[Medline]
Webb S. D.. 1995. Biological implications of the Middle Miocene Amazon seaway. Science 269: 361-362.
Young K. R. Ulloa Ulloa C. Luteyn J. L. Knapp S.. 2002. Plant evolution and endemism in Andean South America: an introduction. Botanical Review 68: 4-21.[CrossRef]
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.80 are given above branches. An asterisk (*) indicates that the amplification of the atpB-rbcL IGS failed; two asterisks (**) indicate that the amplification of the trnK-rps16 IGS failed
