Morphological and molecular evidence in the delimitation of Behria and Bessera, two genera of the Milla complex (Themidaceae)
Evidencia morfológica y molecular en la delimitación de Behria y Bessera, dos géneros del complejo Milla (Themidaceae)
Etelvina Gándara1,2, Victoria Sosa1, José Luis León de La Luz3
1 Biología Evolutiva, Instituto de Ecología, A.C., Apartado Postal 63, 91000 Xalapa, Veracruz, México 2 Corresponding author. E–mail: etelvina.gandara@posgrado.inecol.edu.mx
3 Centro de Investigaciones Biológicas del Noroeste S.C., Apartado Postal 128, 23000 La Paz, Baja California Sur, México
]]>Received: august 14, 2009.
Accepted: october 22, 2009.
Abstract
Among the taxa of the Milla complex (Themidaceae), a monocot group of perennial petaloid geophytes, there are two genera with controversial taxonomic status: Behria and Bessera. Tree–based and character–based analyses were conducted on 14 populations to determine if they should be considered different taxa. In addition, ecological separation was taken into account. As outgroups representative taxa of Dandya, Jaimehintonia, Milla and Petronymphe, i.e. the rest of the genera in the Milla clade, were used. Behria is a monotypic genus restricted to Baja California and Bessera includes two species from the Pacific slopes of Mexico and the Trans–Mexican Volcanic Belt. The chloroplast intergenic spacer psbK–psbI was sequenced and 37 morphological characters were coded. Tree–based analyses retrieved all populations of Behria and all populations of Bessera as monophyletic groups, both forming part of a more inclusive clade. Character–based analysis detected six diagnostic characters, all of which were floral. The conclusion of recognizing Behria and Bessera as independent genera was based on these results and also on their different distributional ranges.
Key words: chloroplast intergenic spacer psbK–psbI, Dandya, Jaimehintonia, petaloid geophytes, Petronymphe.
Resumen
Entre los taxones del complejo Milla (Themidaceae), un grupo de monocotiledóneas petaloides geófitas, existen dos géneros con estatus taxonómico incierto: Behria y Bessera. Con el objetivo de determinar si estos géneros pueden considerarse independientes, se realizaron análisis filogenéticos y de atributos diagnósticos con base en atributos morfológicos y moleculares de 14 poblaciones. Adicionalmente se consideraron individuos representativos del resto de los géneros del complejo, tales como Dandya, Jaimehintonia, Milla y Petronymphe. Behria es un género monotípico restringido al sur de la Península de Baja California. Bessera comprende dos especies de la vertiente pacífica de México y la Faja Volcánica Transmexicana. Se secuenció el espaciador intergénico psbK–psbI del cloroplasto y se codificaron 37 atributos morfológicos. Los análisis filogenéticos identificaron a las poblaciones de Behria y Bessera como grupos monofiléticos independientes, formando parte de un clado más inclusivo. Los resultados del análisis de atributos diagnósticos reconocieron seis atributos florales. Por lo tanto se concluye que estos taxones pueden ser identificados claramente por su morfología floral, además tienen diferentes patrones de distribución y se deben considerar como géneros independientes.
]]> Palabras clave: espaciador intergénico del cloroplasto psbK–psbI, Dandya, geófitas petaloides, Jaimehintonia, Milla y Petronymphe.
Themidaceae is a plant family of perennial petaloid geophytes found mainly in western North America (Fay and Chase, 1996). The genera currently included in Themidaceae were formerly recognized as tribe Brodiaeae in the Alliaceae (Dahlgren et al., 1985). In previous taxonomic studies they were divided into two complexes: the Milla complex, centered in Mexico, and the Brodiaea complex, centered in the western United States (Moore, 1953). More recently, Pires et al. (2001) and Pires and Sytsma (2002) found that the genera are grouped into four clades: (1) the Milla complex, (2) Brodiaea–Dichelostemma–Triteleiopsis, (3) Triteleia–Bloomeria–Muilla clevelandii, and (4) Androstephium–other members of Muilla.
The Milla clade is supported by two synapomorphic character states: an ovary stipe adnate to the perianth tube, and a membranous corm (Pires and Sytsma, 2002). It is comprised of six genera (Behria, Bessera, Dandya, Jaimehintonia, Milla, and Petronymphe). Behria is a monotypic genus endemic to Baja California (Greene, 1886). In contrast, Bessera encompasses two species from western and central Mexico (Ramírez–Delgadillo, 1992; Moore, 1953). Dandya has four species with a restricted geographic distribution in western and northern Mexico (Espejo–Serna and López–Ferrari, 1992; Lenz, 1971b; Villarreal–Quintanilla and Encina–Domínguez, 2005). Milla includes ten species (Lenz, 1971a; Moore, 1953; Howard, 1999) with M. biflora Cav., the most widely distributed member of the group, found from Arizona to Guatemala (Espejo–Serna and López–Ferrari, 2003; McNeal, 2003). Lastly, Jaimehintonia and Petronymphe are both monotypic. The first is restricted to a couple of small areas in Nuevo Leon (Turner, 1993) and the latter is known from a single locality in Guerrero (Moore, 1951).
Greene (1886) described Behria as a new genus when he was writing the monograph on Brodiaea. Later, Macbride (1918) transferred Behria tenuiflora to Bessera arguing that both Bessera elegans Schult. f. and Bessera (Behria) tenuiflora possess a red perianth and exerted stamens fused at the base. According to him, the only difference between the two taxa was the degree of fusion of floral segments. The dispute on the status of Behria continued, some authors treat Behria tenuiflora as part of Bessera (Moore, 1953; Ramírez–Delgadillo, 1992; Espejo–Serna and López–Ferrari, 1992; Pires et al., 2001), while others place this taxon in its own genus (Krause, 1930; Lenz, 1971b; Shreve and Wiggins, 1964; León de la Luz and Pérez–Navarro, 2004).
Whether Behria and Bessera are separate genera has not yet been demonstrated. A study of Themidaceae based on morphological and cpDNA data (trnL–trnF and rbcL) retrieved Bessera elegans and B. (Behria) tenuiflora as sister species, but B. tuitensis was not considered (Pires et al., 2001). Another study based solely on cpDNA (trnL–F, rpl16, and ndhF) retrieved B. elegans and B. tuitensis as sisters, but that study did not include B. (Behria) tenuiflora (Pires and Sytsma, 2002). To date, no study has included the two species in Bessera and Behria tenuiflora simultaneously.
Wiens and Penkrot (2002) proposed a novel method for delimiting species, based on morphological and molecular characters. Tree–based and character–based analyses are performed separately and distribution patterns are also considered. This method utilizes populations as terminals rather than individuals to avoid a biased treatment of the polymorphisms shared between populations as homoplasies rather than synapomorphies in the tree–based morphological analyses. The character–based analysis finds diagnostic character states representing differences among the putative species. This procedure was designed to delimit closely related taxa.
The method of Wiens and Penkrot (2002) is utilized in this study to clarify whether Behria and Bessera are independent taxa. Thus, tree–based and character–based analyses were performed with populations of these closely associated groups, together with representative species of genera in the Milla clade as the study units, employing morphological characters and chloroplast DNA sequences of psbK–psbI.
Materials and methods
]]> Taxon sampling and outgroup selection. Populations were considered as terminals in the analyses. The ingroup was comprised of a total of 14 populations: nine populations of Bessera elegans, the single known population of B. tuitensis and four populations of Behria tenuiflora (table 1; figures 1, 2). As outgroups, a single individual each of Dandya thad–howardii, D. balsensis, Jaimehintonia gypsophila, Milla biflora, M. bryani, M. magnifica, M. mexicana, Milla sp., and Petronymphe decora were used, as suggested in the procedure of Wiens and Penkrot (2002). The total number of individuals was 76 (table 1). Petronymphe decora, the most distantly related taxon according to previous phylogenetic studies, was used as the functional outgroup (Pires and Sytsma, 2002). Fresh leaves were collected from each individual and dried in silica and the vouchers were deposited at XAL and HCIB.Morphological data set. Morphological characters were scored from living material and herbarium specimens and, complemented with information from the literature where necessary (Moore, 1951, 1953; Lenz, 1971b; López–Ferrari and Espejo–Serna, 1992). Vouchers and specimens are listed in Appendix 1. The morphological matrix included 37 characters (12 vegetative and 25 floral). Morphological characters were scored as polymorphic if they varied within populations. Characters and character states are listed in Appendix 2; the data matrix is shown in Appendix 3. (Table 2)
Molecular data set. Total genomic DNA was isolated from silica–gel–dried leaf tissue using the modified 2× CTAB method (Cota–Sánchez et al., 2006). The intergenic spacer between chloroplast genes psbK and psbI was amplified and sequenced using primers and protocols of Lahaye et al. (2008). All amplified products and total DNA were purified using the QIAquick PCR purification kit (Qiagen, California, U.S.A.) following the protocols provided by the manufacturer. Clean products were sequenced using the Taq BigDye Terminator Cycle Sequencing Kit (Perkin Elmer Applied Biosystems, Foster City, California, U.S.A.) and analyzed with an ABI 310 automated DNA sequencer (Perkin Elmer Applied Biosystems, Foster City, U.S.A.). Electropherograms were edited and sequences were assembled using Sequencher 4.1 (GeneCodes, Ann Arbor, Michigan). Sequences were manually aligned with Se–Al v. 2.0a11 (Rambaut, 2002). All sequences were deposited in GenBank (accession numbers are included in Appendix 1).
Tree–based analyses. Data matrices were constructed on WinClada (Nixon 2002). Three sets of phylogenetic analyses were performed with the 23 populations as terminals: 1) A morphological analysis, 2) A molecular analysis, including DNA sequences of psbK–psbI, and 3) A combined analysis.
Parsimony analyses were run in TNT (Goloboff et al., 2003) under equal weights; gaps were taken as missing data, with 200 iterations of parsimony ratchet (Nixon, 1999). Clade support was estimated by Jackknife, with 1,000 replicates with 30% deletion on a traditional search in TNT (Goloboff et al., 2003). Bremer support (Bremer, 1994) was calculated using the BS5 option on 10,000 trees held in memory in NONA (Goloboff, 1999). The potential incongruence of the molecular and morphological data sets was tested using the incongruence length difference (ILD) test of Farris et al. (1995) as implemented in WinClada (Nixon, 2002).
Character–based analysis. Diagnostic character states that represent seemingly fixed differences between the putative taxa have to be detected in the morphological data matrix. The character–based approach was implemented by comparing the frequencies of qualitative characters and the range of trait values for quantitative continuous and meristic characters across all populations to search for potentially diagnostic characters. Characters were considered diagnostic for a species or a set of populations if they were invariant for alternative character states or showed no overlap in trait values as indicated by Wiens and Penkrot (2002).
Results
Tree based–analyses. Morphological analysis.– Parsimony analysis retrieved five most parsimonious trees (MPT) (L = 86 steps, CI =0. 69, RI = 0.85), and one of them is shown in figure 3. MPT shows that the four populations of Behria tenuiflora were retrieved as a monophyletic group with moderate support (Jackknife, jk = 74%, Bremer support, brs = 1) and supported by two synapomorphic character states: a discoid base neck of the perianth tube (character 20) and lobes less than 1/4 of the length of the perianth tube (character 22). The nine Bessera populations were also retrieved as a monophyletic group with moderate support (jk = 73%, brs = 1). Two synapomorphic character states supported this group: campanulate flowers (character 18), and perianth tube enclosing part of ovary (character 19). A population of B. elegans from Colima showed four autapomorphies: purple tepals (character 23), purple staminal tube of the same color as filaments, and a membrane apex truncated (characters 32–34). Bessera tuitensis population showed two autapomorphic character states: pinkish tepals (character 23), and staminal ring 1–1.5 mm length (character 29). The Behria/Bessera clade received moderate support (jk = 87%, brs = 1) and is marked by a single synapomorphy: filaments longer than tepals (character 28) (figure 4). Consensus tree topology was identical to the combined consensus (figure 5). Molecular analysis.– The psbK–psbI data matrix included 494 base pairs (bp) with nine (1.91%) being parsimony informative, and 38 (7.69%) variable. Parsimony analysis retrieved a single MPT displayed in figure 4 (L = 37 steps, CI = 1, Ri = 1), in which only a few groups received support: the four populations of Behria tenuiflora (jk = 68 %), the two representative species of Dandya (jk = 96 %) and two Milla species: Milla sp. and M. bryani (jk = 67 %).
Combined analysis.– The combined analysis of psbK–psbI DNA sequences and morphology retrieved five MPT (L = 128 steps, CI = 0.75, RI = 0.84). The strict consensus is shown in figure 5. The same groups were retrieved as in the morphological analysis, though the clades received higher jackknife and Bremer support. The four populations of Behria were retrieved as monophyletic with strong support (jk = 92%, brs = 6). Bessera populations were also supported (jk = 80%, brs = 6). Behria appears as the sister group to Bessera and this clade received support (jk = 86%, brs = 6). Likewise there was another supported clade formed by several Milla species (jk = 82%, brs = 6) which was sister to Jaimehintonia (jk = 67%, brs = 6). Finally, Dandya species were in a highly supported clade (jk = 100%, brs = 6).
]]> Character based analysis. The morphological character–based approach supports the recognition of two different related taxa. Bessera can be distinguished from Behria by five diagnostic characters: 1) polytepalous flowers (vs gamotepalous) (character 17); 2) base neck of the perianth tube elongated (vs discoid) (character 20); 3) three lobes of flower different in shape and size (vs all equal shape) (character 21); 4) stamen filaments connate and 5) forming a membranous tube 10–25 mm length (vs filament base connate and forming a membranous cup to 0.6 mm length) (character 29). In addition Behria has tubular perianth (character 18); 3) which totally encloses the ovary (character 19).
Discussion
Tree based analyses. our results show that the single MPT retrieved from the molecular analysis is mostly unresolved. We utilized the chloroplast intergenic spacer psbK–psbI because it is a chloroplast region that was included as a candidate locus for DNA barcoding for its evolutionary rates and has a high ability to differentiate independently evolving entities corresponding to taxonomic species (Pennisi, 2007; Lahaye et al., 2008). However it was not variable enough to allow resolution in our MPT. Further loci must be added to improve resolution. Chloroplast DNA loci such as trnL–F, rpl16 and ndhF have been sequenced in Themidaceae but they did not have enough variation to resolve the relationships within this family (Pires and Sytsma, 2002). A large number of markers are known to be variable in angio–sperms, like trnS–trnG, rpl32–trnL and trnQ–rps16proposed by Shaw et al. (2005; 2007), and these could be tested for resolving relationships within Behria and Bessera, and in even in the Milla clade.
In contrast, the combined and morphological analyses recovered the same clades, but the combined analysis received higher Jackknife and Bremer support. Morphological and combined phylogenetic analyses identified the four populations of Behria as a monophyletic group. The nine populations of Bessera elegans and the single population of B. tuitensis were also identified by these analyses as a monophyletic group. These two clades form part of a more inclusive group. Although populations of Bessera and Behria were recovered by the morphological and combined analyses as sister groups, this grouping is supported by a single synapomorphic character state: filaments longer than tepals.
Character based analysis. Five diagnostic characters were detected in the character analysis and all of them are floral attributes. Vegetative morphology is similar in Behria and Bessera and also in the genera of the entire family. Bessera species have campanulate flowers, with a perianth tube partially enclosing the ovary. in contrast Behria has tubular flowers and the perianth tube completely encloses the ovary. Thus, the floral characters pointed out by previous authors (Schultes, 1829; Greene 1886; Moore, 1953; Lenz 1971b; León de la Luz and Pérez–Navarro, 2004) for differentiating Bessera from Behria (campanulate flowers vs. tubular flowers) were detected by us as diagnostic.
Based on the tree–morphology and on the character–based results we conclude that Behria and Bessera should be recognized as separate genera. Populations of these taxa were retrieved as monophyletic groups, forming part of a more inclusive clade. There is still controversy about "how to chop up a tree" for assigning taxonomic rank and it has been suggested that recognition of the non–monophyletic and nomenclaturally redundant monotypic genera be avoided (Brumitt, 2002). Changes in generic delimitation practice have been influenced by new sources of data, such as molecular characters. Technological advances that have allowed widespread incorporation of molecular data into taxonomic studies have also facilitated a return to a global approach to the study of plant genera (Humphreys and Linder, 2009). Decisions on the identification of independent genera which are sister taxa differ depending on the plant group and on the author. For example, Tang and Lu (2005) decided to recognize Zabelia (Caprifoliaceae) as independent from Abelia. Although molecular analyses found these taxa as sisters they based their decision on morphological and ecological evidence. Devos et al. (2006) decided that the monotypic genus Coeloglossum (orchidaceae) was separate from Dactylorhiza, even though these genera were retrieved as sister groups of a more inclusive clade. in another example Specht and Stevenson (2006), based on morphological and molecular evidence, considered Monocostus (Costaceae) as a genus independent of Dimerocostus, though they were retrieved as sister groups.
We also based our decision to recognize Behria and Bessera as separate genera on differences in their distribution ranges. As mentioned above, Behria is restricted to the Baja California Peninsula in the southern area of La Paz–Los Cabos while Bessera occurs on continental Mexico on the Pacific slopes and along the Transmexican Volcanic Belt. (Moore, 1953; Ramírez–Delgadillo, 1992; León de la Luz and Pérez–Navarro, 2004). The species delimitation method of Wiens and Penkrot (2002) favors recognizing taxa as independent if their populations are geographically separated. Moreover, Martínez–Gutiérrez and Sethi (1997) indicated that the La Paz–Los Cabos morphotectonic subprovince (Ferrusquía–Villafranca, 1993) was connected to the Pacific slope, in Jalisco, during the lower Miocene, between 16–13 my before present. An interesting hypothesis that could be addressed in the future would include dating the divergence of Behria and Bessera to determine if it coincides with the separation of these two areas. Both are part of North America, but the Baja California Peninsula is rifting with the Pacific plate.
In Amaryllidaceae (Meerow et al., 1999) and in Iridaceae (Taylor et al., 2009), two groups of monocots that are closely related to Themidaceae, high floral diversity has been attributed to floral modification induced by pollinator–mediated selection, driven by changes in one or only a few genes. It has been documented that Behria is pollinated by hummingbirds (Arriaga et al., 1990). The pollination vectors for Bessera are unknown, though we did observe hummingbirds, bees, bumble bees and butterflies visiting flowers in the populations where we collected plants of this genus. It is important to determine if pollination systems have played a role in floral diversification in this group as it has in other monocot groups.
Behria, Bessera and the rest of the Milla clade represent an interesting group for evolutionary biology studies, such as determining time of divergence and factors influencing speciation, such as shifts in soil type and pollinators. The majority of the taxa in the Milla clade are restricted to a single or to a few localities, and each taxon grows on a different kind of soil, and is also pollinated by different vectors.
]]>Acknowledgments
This study was supported by a UC–MEXUS grant. The first author is grateful to CONACyT (13229) and also thanks A. Espinosa and A. Rodríguez for their help and guidance during this study. We appreciate the help of D. Angulo, J. Álvarez, C. Henríquez and H. Driscoll with field work. We also thank Eduardo Ruiz for his assistance with the field and lab work and for providing some of the photographs reproduced here.
References
Arriaga L., Rodríguez–Estrella R. and Ortega–Rubio. A.1990. Endemic hummingbird and madrones of Baja: Are they mutually dependent? The Southwestern Naturalist 35:76–79. [ Links ]
Bremer K. 1994. Branch support and tree stability. Cladistics 10:295–304. [ Links ]
Brummitt R.K. 2002. How to chop up a tree. Taxon 51:31–41. [ Links ]
Cota–Sánchez J.H., Remarchuk K. and Ubayasena K. 2006. Ready–to–use DNA extracted with a CTAB method adapted for herbarium specimens and mucilaginous plant tissue. Plant Molecular Biology Reporter 24:161–167. [ Links ]
Dahlgren R.M.T., Clifford H.T. and Yeo P.F. 1985. The families of monocotyledons: structure, evolution and taxonomy. Springer Verlag. Germany. [ Links ]
Devos N., Raspé O., Jacquemart A.L. and Tyteca D. 2006. On the monophyly of Dactylorhiza Necker ex Nevski (Orchidaceae): is Coeloglossum viride (L.) Hartman a Dactylorhiza? Botanical Journal of the Linnean Society 152:261–269. [ Links ]
Engler A. 1887. Liliaceae. En: Engler A. Prantl K. Eds. Die Natürlichen Pflanzenfamilien nebst ihren Gattungen und wichtigeren Arten insbesondere den Nutzpflanzen II. pp.10–91. W. Engelmann. Leipzig, Alemania. [ Links ]
Espejo–Serna A. and López–Ferrari A.R. 1992. Las monocotiledóneas mexicanas. Una sinopsis florística. 1 Lista de referencia parte I: Agavaceae, Alismaceae, Alliaceae, Alstroemeriaceae y Amaryllidaceae. Consejo Nacional de la Flora de México A.C., y Universidad Autónoma Metropolitana Iztapalapa, México, D.F [ Links ]
Espejo–Serna A. and López–Ferrari A.R. 2003. Alliaceae. Flora de Veracruz. Fascículo 132. Instituto de Ecología, A.C. Xalapa and University of California. Riverside, California. [ Links ]
Farris J.S., Källersjö M., Kluge A.G. and Bult C. 1995. Testing significance of incongruence. Cladistics 10:315–319. [ Links ]
Fay M.F. and Chase M.W. 1996. Resurrection of Themidaceae for the Brodiaea alliance, and recircumscription of Alliaceae, Amaryllidadeae and Agapanthoideae. Taxon 45:441–451. [ Links ]
Ferrusquía–Villafranca I. 1993. Geology of Mexico: A synopsis. In: Ramamoorthy T.P., Bye R., Lot A. and Fa J. Eds. Biological diversity of Mexico: Origins and distribution, pp:3–107. Oxford University Press, Nueva York. [ Links ]
Goloboff P. 1999. NONA (NO NAME) ver. 2 Published by the author, Tucumán, Argentina. [ Links ]
]]>Goloboff P, Farris S. and Nixon K. 2000. TNT (Tree analysis using New Technology) (BETA) ver. 1.1 Published by the authors, Tucumán, Argentina. [ Links ]
Greene E.L. 1886. Studies in the botany of California and parts adjacent IV. Bulletin of the California Academy of Sciences 2:125–154. [ Links ]
Howard T.M. 1999. Three new Milla species from Mexico. Herbertia 54:232–237. [ Links ]
Humphreys A.M. and Linder H.P. 2009. Concept versus data in delimitation of plant genera. Taxon 12: 1–21. [ Links ]
Lahaye R., Savolainen V., Duthoit S., Maurin O. and van der Bank M. 2008. A test of psbK–psbI and atpF–atpH as potential plant DNA barcodes using the flora of the Kruger National Park (South Africa) as a model system. Available on Nature Precedings <http://hdl.handle.net/10101/npre.2008.1896.1> [ Links ]
]]>Lenz L.W. 1971a. Chromosome numbers in the genus Milla Cav. (Liliaceae). Aliso 7:321–323. [ Links ]
Lenz L.W. 1971b. Two new species of Dandya (Liliaceae) from Mexico and reexamination of Bessera and Behria. Aliso 7:313320. [ Links ]
León–de la Luz J.L. and Pérez–Navarro J.J. 2004. Neotipificación de Behria tenuiflora Greene (Alliaceae). Acta Botánica Mexicana 67:59–66. [ Links ]
López–Ferrari A.R. and A. Espejo–Serna A. 1992. Una nueva especie de Dandya (Alliaceae) de la cuenca del río Balsas, México. Acta Botánica Mexicana 18:11–15. [ Links ]
Macbride J.F. 1918. Further new or otherwise interesting Liliaceae. Contributions of the Gray Herbarium 56:1–20. [ Links ]
]]>McNeal D.W. 2003. Milla. In: Flora of North America Editorial Commites. Eds. Flora of North America vol. 26: Magnoliophy–ta: Lilidae: Liliales and Orchidiales, pp.346. Oxford University Press. Oxford. Available on line: <http://www.efloras.org/flora–taxon.aspx?flora_id=1&taxon_id=120721> [ Links ].
Martínez–Gutiérrez G. and Sethi P.S. 1997. Miocene–Pleistocene sediments within the San José del Cabo Basin, Baja California Sur, Mexico. In: Johnson M.E. and Ledesma–Vázquez J. Eds. Pliocene carbonates and related facies flanking the Gulf of California, Mexico, pp.141–166. Special paper 318 The Geological Society of America.USA. [ Links ]
Meerow A.W., Fay M.F., Guy C.L., Li Q.B., Zaman F.Q. and Chase M.W. 1999. Systematics of Amaryllidaceae based on cladistic analysis of plastid rbcL and trnL–F sequence data. American Journal of Botany 86:1325–1345. [ Links ]
Moore H.E. 1951. Petronymphe, a new genus of Amaryllidaceae. Gentes Herbarum 8:258–260. [ Links ]
Moore H.E. 1953. The genus Milla (Amaryllidaceae–Allieae) and its allies. Gentes Herbarum 8:262–294. [ Links ]
]]>Nixon K.C. 1999. The parsimony ratchet, a new method for rapid parsimony analysis. Cladistics 15:407–414. [ Links ]
Nixon, K.C. 1999–2002. WinClada ver. 1.0000 Published by the author, Ithaca, NY, USA. [ Links ]
Pennisi E. 2007. Wanted: a barcode for plants. Science 318:190191. [ Links ]
Pires J.C. and Sytsma K.J. 2002. A phylogenetic evaluation of a biosystematic framework: Brodiaea and related petaloid monocots (Themidaceae). American Journal of Botany 89:13421359. [ Links ]
Pires J.C., Fay M.F., Davis W.S., Hufford L., Rova J., Chase M.W. and Sytsma K.J. 2001. Molecular and morphological analyses of Themidaceae (Asparagales). Kew Bulletin 56:601–626. [ Links ]
]]>Rambaut A. 2002. Se–Al Sequence Alignment Editor, v2.0a11. University of Edinburgh, Edinburgh, Scotland, United Kingdom. Available on line: <http://tree.bio.ed.ac.uk/software/seal/> [ Links ]
Ramírez–Delgadillo R. 1992. Una nueva especie de Bessera (Liliaceae) del Occidente de Jalisco, México: Boletín del Instituto de Botánica de la Universidad de Guadalajara 1:131–136. [ Links ]
Shaw J., Lickey E.B., Beck J.T., Farmer S.B., Liuet S.W., Miller J., Siripun K.C., Winder C.T, Schilling E.E. and Small R.L. 2005. The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. American Journal of Botany 92:142–166. [ Links ]
Shaw J., Lickey E.B., Schilling E.E. and Small R.L. 2007. Comparison of whole chloroplast genome sequences to choose non–coding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. American Journal of Botany 94:275–288. [ Links ]
Shreve F. and Wiggins I.L. 1964. Vegetation and flora of the Sonoran Desert. Stanford University Press. Stanford, California. [ Links ]
]]>Shultes J.H. 1829. Bessera elegans Genus novum Hexandriae Monogyniae. Linnaea 4:121–127. [ Links ]
Specht C.D. and Stevenson D.W. 2006. A new phylogeny–based generic classification of Costaceae (Zingiberales). Taxon 55:153–163. [ Links ]
Tang Y.C. and Lu A.M. Paraphyletic group, PhyloCode and phylogenetic species – the current debate and a preliminary commentary more options. Acta Phytotaxonomica Sinica 43:403–419. [ Links ]
Taylor S.J., Arnold M. and Martin N.H. 2009. The genetic architecture of reproductive isolation in Louisiana irises: hybrid fitness in nature. Evolution 63:2581–2594. [ Links ]
Turner B.L. 1993. Jaimehintonia (Amaryllidaceae–Allieae), a new genus from northeastern Mexico. Novon 3:86–88. [ Links ]
]]>Villarreal–Quintanilla J.A. and Encina–Domínguez J.A. 2005. Plantas vasculares endémicas de Coahuila y algunas áreas adyacentes, México. Acta Botánica Mexicana 70:1–46. [ Links ]
Wiens J.J. and Penkrot T.A. 2002. Delimiting species using DNA and morphological variation and discordant species limits in spiny lizards (Sceloporus). Systematic Biology 51:69–91. [ Links ]
]]>