Pl Syst Evol 270: 183–207 (2008) DOI 10.1007/s00606-007-0604-4 Printed in The Netherlands Plant Systematics and Evolution Biosystematics of Chamaecrista sect. Absus subsect. Baseophyllum (Leguminosae-Caesalpinioideae) based on allozyme and morphometric analyses A. S. Conceição,1,2 L. P. Queiroz,1 S. M. Lambert,1 A. C. S. Pereira,1 E. L. Borba3 1 Departamento de Ciências Biológicas, Universidade Estadual de Feira de Santana, Feira de Santana, Bahia, Brazil Current address: Colegiado de Ciências, Universidade do Estado da Bahia, Campus VIII, Paulo Afonso Bahia, Brazil 3 Departamento de Botânica, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil 2 Received 16 February 2007; Accepted 4 September 2007; Published online 4 December 2007
Springer-Verlag 2007 Abstract. We carried out genetic and morphometric analyses in 33 populations belonging to all nine described and one putative taxa of Chamaecrista subsect. Baseophyllum. Genetic variability was low in all taxa. Morphological and genetic data are consistent with the status of C. depauperata and C. coriacea. However, divergence between C. cytisoides var. cytisoides and all the other six conspecific varieties is higher than the divergence between it and the other two species. Our data support the recognition of eight species in the subsection Baseophyllum. Chamaecrista coriacea and C. depauperata are supported in their original circumscription. However, the varieties ascribed to C. cytisoides are best treated as six different species. We propose five new combinations in order to raise varieties of C. cytisoides to specific rank: C. blanchetii, C. brachystachya, C. confertiformis, C. decora, C. unijuga. We also propose synonimization of one variety, C. cytisoides var. micrantha, under C. brachystachya. Keywords: allozymes; Caesalpinioideae; campos rupestres; Chamaecrista; Fabaceae; flower morphology; genetic diversity; morphometrics Chamaecrista Moench is one of the largest genera of the subfamily Caesalpinioideae. It includes about 330 species widely distributed in the tropical areas of America, Africa and Asia (Lewis 2005). The genus is divided in six sections according to Irwin and Barneby (1982) classification, diagnosed mainly by combining characters of the habit, inflorescence and indument. Section Absus is characterized by mostly shrubby habit with terminal racemes. Most of its subsections present glandular setose trichomes and lack extrafloral nectaries. Subsection Baseophyllum differs from all other subsections by the presence of extrafloral nectaries on the petiole or leaf rachis, lack of glandular trichomes, leaves with one to four pairs of leaflets, leaflets 3–9 palmately veined, and inflorescences racemosesubcorymbose (Irwin and Barneby 1978). Subsection Baseophyllum is mainly distributed in the Espinhaço mountain range in Bahia and Minas Gerais states, Brazil, usually in Correspondence: A. S. Conceição, Departamento de Ciências Biológicas, Universidade Estadual de Feira de Santana, Feira de Santana, Bahia, Brazil; current address: Colegiado de Ciências, Universidade do Estado da Bahia, Campus VIII, Paulo Afonso, Bahia, Brazil e-mail: adilva_souza@yahoo.com.br 184 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) ‘‘campo rupestre’’ vegetation (‘rocky fields’) but sometimes occurring in ‘‘cerrado’’ (savanna-like vegetation). It also occurs disjunctly in coastal ‘‘restinga’’ vegetation (low forests or scrubs on salty and sandy soils) of Eastern Brazil (Pernambuco to Espı́rito Santo states) and at a spot of ‘‘caatinga’’ vegetation (seasonally dry thorny forest) in Pernambuco state (Irwin and Barneby 1978, Conceição 2006). Subsection Baseophyllum has a tortuous taxonomic history mainly due to different interpretations of the species limits in the Chamaecrista cytisoides complex. Bentham (1870) recognized four independent species and two varieties (as Cassia): Cassia blanchetii, Cassia brachystachya (vars. brachystachya and unijuga), Cassia coriacea and Cassia cytisoides. One year later the same author did not recognize the variety unijuga (Bentham 1871). Irwin and Barneby (1978) treated all these taxa as varieties of Cassia cytisoides, except for Cassia coriacea. Later, they transferred these taxa to varieties of Chamaecrista cytisoides (Irwin and Barneby 1982). According to the circumscription of Irwin and Barneby (1982), the subsection Baseophyllum includes only two species: Chamaecrista coriacea (Benth.) H.S.Irwin & Barneby and C. cytisoides (Collad.) H.S.Irwin & Barneby, the later including seven varieties: C. cytisoides var. blanchetii (Benth.) H.S.Irwin & Barneby, C. cytisoides var. brachystachya (Benth.) H.S.Irwin & Barneby, C. cytisoides var. confertiformis (H.S.Irwin & Barneby) H.S.Irwin & Barneby, C. cytisoides var. cytisoides, C. cytisoides var. decora (H.S.Irwin & Barneby) H.S.Irwin & Barneby, C. cytisoides var. micrantha (H.S.Irwin & Barneby) H.S.Irwin & Barneby, and C. cytisoides var. unijuga (Benth.) H.S.Irwin & Barneby. According to Irwin and Barneby (1982), this wide circumscription of C. cytisoides is supported by the conservative floral morphology amongst the putative varieties and the presence of intermediate morphs between the different taxa, suggesting a continuous variation pattern in vegetative traits. Some varieties of the C. cytisoides complex occur in sympatry and sometimes are syntopic, although they flower in different periods (according to data presented in Irwin and Barneby 1978). Conceição (2000) recognized a ‘‘prostrata form’’ of C. cytisoides but did not ascribe a formal taxonomic rank since it was not possible to establish its relationships with other taxa of the complex. Later, Conceição et al. (2001) described C. depauperata Conc., L.P.Queiroz & G.P.Lewis, a species similar to the ‘‘prostrata’’ morph of C. cytisoides but with flowers smaller than those found in C. cytisoides and C. coriacea. A summary of the current taxonomic situation of the subsection Baseophyllum (sensu Irwin and Barneby 1982) and diagnostic characters of their taxa is presented in Table 1 and Figs. 1, 2. The purposes of this work are to explore the number of taxonomical entities, their taxonomic status and to provide diagnostic characters for all taxa currently ascribed to Chamaecrista sect. Absus subsect. Baseophyllym. In order to achieve these objectives we assessed the levels of genetic and morphological variation and sub-structuring within and between the species and populations of the taxa in subsect. Baseophyllum using allozyme markers and morphometric analyses, so as to expand the knowledge of intra and interspecific relationships within this group. We also investigated the correlation between genetic and morphologic variability. The use of genetic diagnostic markers, genetic distance/ similarities, and multivariate and statistical analyses of morphometric data can be helpful in evaluating specific and infraspecific limits, mainly when the traditional taxonomic procedures are not satisfactorily conclusive (Crawford 1989; van der Bank et al. 2001; Henderson 2006). A similar approach has been successfully used in studies of other plant groups native to the Brazilian ‘‘campos rupestres’’ (Borba et al. 2000, 2001, 2002; Lambert et al. 2006a, b; Pereira et al. 2007). Materials and methods Populations sampled. We collected 600 individuals from 30 natural populations distributed in the nine taxa of Chamaecrista subsection Baseophyllum (Table 2, Figs. 1, 2) at 22 localities in Espı́rito Santo, Minas Gerais, Bahia, Sergipe and Pernambuco states: Taxon (Irwin and Barneby 1982) Characters Leaflet shape Leaflet color Flower diameter (cm) 1–2 Suborbicular to oblanceolate Orbicular to suborbicular Green 1.5–1.8 campo rupestre N-MG C. coriacea Green 0.5–1.0 campo rupestre C-BA C. depauperata Absent or 1 short Orbicular to suborbicular Green 1.5–1.8 campo rupestre, cerrado, restinga C. blanchetii var. Shrub to brachystachya treelet Absent or 1–2 short (-3) Reniform to widely obovate Green 1.5–2.0 campo rupestre, caatinga, restinga var. confertiformis var. cytisoides Short 2 1.2–2.0 campo rupestre Short 3–4 1.5–2.0 cerrado S-MG C. confertiformis C. cytisoides Present 1 Oblong to Green oblanceolate Ovate to Green oblanceolate Obliquely obovate Purple or Green Obovate to Green oblanceolate Reniform to Green orbicular Suborbicular Green C-BA, N-MG, ct-ES N-MG, C-PE, ct-BA and SE C-BA 1.5–2.0 campo rupestre N-MG C. decora 1.5–1.7 campo rupestre 1.3–1.7 restinga C. brachystachya C. unijuga 1.5–2.0 campo rupestre C-BA, N-MG ct-PE, AL and SE C-BA C. depauperata C. cytisoides var. blanchetii var. decora var. micrantha var. unijuga ‘‘f. prostrata’’ Petiole Distribution Proposed classification (this work) Leaflet pairs C. coriacea Habit Habitat Prostate Present sub-shrub Prostate Present sub-shrub Shrub to treelet Shrub to treelet Shrub to treelet Shrub to treelet Shrub to treelet Shrub to treelet Prostate sub-shrub 1 Absent or 2 short Present 1 Absent or 1 short C. brachystachya* C. blanchetii A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Table 1. Diagnostic characters and distribution data of the taxa of Chamaecrista section Absus subsection Baseophyllum according to Irwin and Barneby (1982) classification with addition of C. depauperata by Conceição et al. (2001) and the unpublished form ‘‘prostrata’’ (Conceição 2000). It is also presented the proposed classification based on the results of this work. (see discussion in the text). Distribution: C center; N Northern; S Southern; ct coastal. Acronyms for Brazilian States: AL Alagoas; BA Bahia; ES Espı́rito Santo; MG Minas Gerais; PE Pernambuco; SE Sergipe * Except populations CBR1, CBR4 and CBR5 (see Table 2 and discussion in the text) 185 186 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) C. coriacea (one population), C. depauperata (two populations), C. cytisoides var. blanchetii (seven populations), C. cytisoides var. brachystachya (five populations), C. cytisoides var. confertiformis (five populations), C. cytisoides var. cytisoides (one population), C. cytisoides var. decora (three populations), C. cytisoides var. micrantha (four populations), and C. cytisoides var. unijuga (two populations). We also collected three populations of a putative taxon for the subsection, called here C. cytisoides ‘‘f. prostrata’’. The distance among sample individuals within a population was of approximately 20 meters. The individuals were not maintained in the same environmental conditions due to the difficulty of growing them in a greenhouse, as all species are large long-lived shrubs or small trees. Vouchers for each population are deposited in the herbarium of the Universidade Estadual de Feira de Santana (HUEFS). Electrophoretic procedures. Small sections of leaf tissue were crushed in 0.5 mL of grinding buffer (Lambert et al. 2006a). Extracts were absorbed in 1.0 · 0.3 cm Whatman number 3 paper wicks, which were loaded into 8.5% starch gels (Sigma hydrolyzed potato starch). For the electrodes and gels four buffer systems were used: 1) electrode: lithium hydroxide 0.05 mol L–1 boric acid 0.0935 mol L–1 EDTA 0.0059 mol L–1 pH 8.0; gel electrode solution diluted 1:10; modified from Ridgway et al. (1970); 2) electrode: citric acid 0.04M adjusted to pH 6.1 with N-(3-aminopropyl)-morpholine; gel: electrode solution diluted 1:20; modified from Clayton and Tretiak (1972); 3) electrode: Tris 0.17 mol/L, citric acid 0.04 mol/L, pH 8.6; gel: tris 0.02 mol/L, citric acid 0.0013 mol/L, pH 8.6; 4) electrode: boric acid 0.3 mol/L, NaOH 0.06 mol/L, pH 8.0; gel: Tris 0.01 mol/ L, pH 8.5; modified from Shaw and Prasad (1970). Standard horizontal eletrophoresis was performed until the inner marker (bromophenol blue) reached 9 cm from the application site using the following running conditions: systems 1, 2 and 3 – 25 mA; system 4 – 15 mA. Nine enzymatic systems gave enough resolution for reading and were used. System 1 was used for malate dehydrogenase (MDH; EC 1.1.1.37), phosphoglucomutase (PGM; EC 2.7.5.1), esterase (EST; EC 3.1.1.1), shikimate dehydrogenase (SKDH; EC 1.1.1.25), and glucose-6-phosphate dehydrogenase (G6PDH; EC 1.1.1.49); system 2 was used for isocitrate dehydrogenase (IDH; EC 1.1.1.42); system 3 was used for leucine aminopeptidase (LAP; AC 3.4.11.1); and system 4 was used for phosphoglucose isomerase (PGI; EC 5.3.1.9) and acid phosphatase (ACPH; EC 3.1.3.2). The staining procedures were similar to but slightly adjusted from Brune et al. (1998; ACPH, EST, LAP, SKDH, G6PDH), Corrias et al. (1991; IDH, PGI) and Soltis et al. (1983; PGM, MDH). Modifications were mainly in the amounts of the components used; the exact recipes can be obtained on request. Enzymatic systems showing more than one locus were numbered in ascending order from the locus with lowest mobility. The alleles were numbered according their mobility relative to the allele with highest mobility of a standard individual present in all gels and designated as 100. Analyses of allozyme data. Banding patterns were interpreted by direct observation of the homozygotes and heterozygotes stained in the gels. Genetic variability for each population was estimated by the following parameters: proportion of polymorphic loci (P; 0.95 criterion), mean number of alleles per locus (A), observed (Ho) and expected (He) mean heterozygosity per locus. Deviations from the expected mean heterozygosity under HardyWeinberg (HW) equilibrium were tested using v2 with a correction for small samples according to Levene (1949). Partitioning of genetic diversity among conspecific populations was estimated by F statistics (FIS, the inbreeding coefficient measures the reduction in heterozygosity due to nonrandom mating within a population; FST, measures the differentiation among populations; Wright 1978). Matrices of genetic distances (unbiased genetic distance; Nei 1978) and genetic identities (unbiased genetic identity; Nei 1978) were calculated for populations and species. Cluster analysis was performed with the genetic distance matrix of the populations using UPGMA (unweighted pair-group method using arithmetical averages) as grouping algorithm (Sneath and Sokal 1973). We also accomplished analyses with other clustering algorithm (complete linkage, Ward, weighted pair-group) to test the consistence of the groups. All analyses were carried out using the BIOSYS 1.0 software package (Swofford and Selander 1989), except for the cluster analysis that was performed using STATISTICA for Windows 6.0 version (StatSoft 2000). Morphometric analyses. The individuals sampled for the allozyme analysis were also used in analyses of morphological variability. We measured 54 quantitative morphological characters (except for variable #2, which is binary), with 38 floral and 16 vegetative characters (Table 3). All measurements of quantitative characters A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) 187 Fig. 1. Distribution map of Chamaecrista sect. Absus subsect. Baseophyllum based on field collection and herbarium material, and location of the populations studied: A, Buı́que (CBR3); B, Santo Amaro das Brotas (CU1); C, Pirambu (CU2); D, Itabaiana (CBR5); E, Indiaroba (CB7); F, Esplanada (CBR1); G, Salvador (CBR4); H, Saúde (CM1); I, Jacobina (CB2); J, Lençóis (CC4); K, Mucugê (CB4, CB6, CC1, CC2, CC3, CC5, CM3, D2); L, Abaı́ra (CB3, CFP2, CM4, D1); M, Ibicoara (CFP1); N, Barra da Estiva (CFP3); O, Pedra Azul (CB5); P, Grão Mogol (CD1, CM2); Q, Diamantina (CBR2, CD2, CD3); R, Conceição do Mato Dentro (C1); S, Guarapari (CB1); T, Sta. Bárbara M. Verde (CCY). See Table 2 for the names of the populations. Acronyms for Brazilian States: AL Alagoas; BA Bahia; ES Espı́rito Santo; MG Minas Gerais; PE Pernambuco; SE Sergipe were taken with the aid of a vernier caliper. Based on morphology and position of the nectaries, we considered that the leaflets of bifoliolate leaves is homologous to the second pair of leaflets of tetrafoliolate leaves. We carried out the following analyses: 1) all populations, using only vegetative morphological characters; 2) all populations, 188 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Fig. 2. A Chamaecrista depauperata (pop. D2); B C. coriacea (C1); C C. cytisoides var. cytisoides (CCY1); D C. cytisoides var. blanchetii (CB2); E C. cytisoides var. brachystachya (CBR5, population from coast); F C. cytisoides var. brachystachya (CBR3, population from interior); G C. cytisoides var. confertiformis (CC1); H C. cytisoides var. decora (CD1); I C. cytisoides var. micrantha (CM4); J C. cytisoides var. unijuga (CU2); K flower of C. cytisoides var. blanchetii (CB2); L flower of C. depauperata (D2). See Table 2 for the names of the populations A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) 189 Table 2. Populations of Chamaecrista coriacea, C. depauperata, C. cytisoides var. blanchetii, C. cytisoides var. brachystachya, C. cytisoides var. confertiformis, C. cytisoides var. cytisoides, C. cytisoides var. decora, C. cytisoides var. micrantha, C. cytisoides ‘‘f. prostrata’’ and C. cytisoides var. unijuga, occurring in northeastern and southeastern Brazil, used in the study. Vouchers deposited at the herbarium of the Universidade Estadual de Feira de Santana (HUEFS). Acronyms for Brazilian States: AL Alagoas; BA Bahia; ES Espı́rito Santo; MG Minas Gerais; PE Pernambuco; SE Sergipe Population Municipality C. coriacea C1 Conceição do Mato Costa Sena-Guritiba Dentro-MG Cima C. depauperata D1 Abaı́ra-BA Catolés-Virassaia D2 Mucugê-BA Paty-Gerais do Rio Preto C. cytisoides var. blanchetii CB1 Guarapari-ES P. Estad. Paulo César Vinha CB2 Jacobina-BA Serra do Tombador CB3 Abaı́ra-BA Catolés-Bicota CB4 Mucugê-BA Projeto Sempre Viva CB5 Pedra Azul-MG 47 km ao Sul Pedra Azul CB6 Mucugê-BA Gobira CB7 Indiaroba-SE 2 km da entrada de Pontal C. cytisoides var. brachystachya CBR1 Esplanada-BA Faz. Morro da Graça, Bú CBR2 Diamantina-MG 7 km E de Diamantina CBR3 Buı́que-PE Serra de Jerusalém CBR4 Salvador-BA APA Stella Mares CBR5 Itabaiana-SE Serra de Itabaiana C. cytisoides var. confertiformis CC1 Mucugê-BA Piabinha CC2 Mucugê-BA Campo Alegre CC3 Mucugê-BA Paty-Gerais do Rio Preto CC4 Lençóis-BA Foz do rio Capivara CC5 Mucugê-BA Morro Fervido C. cytisoides var. cytisoides CCY1 Santa Bárbara Três Cruzes-Serra M. Verde-MG Negra C. cytisoides var. decora CD1 Grão Mogol-MG Vau-Itacambiruçu CD2 Diamantina-MG Beriberi CD3 Diamantina-MG Cristais C. cytisoides var. micrantha CM1 Saúde-BA Serra de Saúde CM2 Grão Mogol-MG Cancela CM3 Mucugê-BA 9 km de Igatu Location Voucher 18430 3100 S, 43370 2400 W A.S.Conceição 869 13500 S, 41220 W 12450 49.100 S, 41300 0500 W A.S.Conceição 775 A.S.Conceição 863 20350 7.800 S, 40250 17.300 W A.S.Conceição 826 11120 S, 40410 W 13190 23.900 S, 41500 58.700 W 12590 3000 S, 41200 3000 W 16200 20.800 S, 41050 23.300 W A.S.Conceição A.S.Conceição A.S.Conceição A.S.Conceição 772 664 759 831 13040 3800 S, 41220 3100 W E.L.Borba 1849 11290 40.400 S, 37260 18.200 W A.S.Conceição 693 12030 3800 S, 37410 3900 W A.S.Conceição 687 18100 5400 S, 43330 45.500 W 08340 5700 S, 37140 2400 W 12560 S, 38210 W 10450 6.900 S, 42430 38.300 W A.S.Conceição A.S.Conceição A.S.Conceição A.S.Conceição 540 761 832 819 12590 33.500 S, 41200 27.800 W A.S.Conceição 755 13140 S, 41230 W A.S.Conceição 834 12460 15.700 S, 41290 29.300 W A.S.Conceição 864 12330 S, 41240 W A.S.Conceição 833 12590 46.400 S, 41200 17.700 W A.S.Conceição 835 21580 0.300 S, 43490 12.100 W A.S.Conceição 870 16350 26.900 S, 42540 8.100 W A.S.Conceição 810 18080 53.200 S, 43360 53.100 W A.S.Conceição 828 18100 0.0800 S, 43350 57.300 W A.S.Conceição 827 10560 3700 S, 40240 2900 W A.S.Conceição 774 16250 7.100 S, 42430 38.300 W A.S.Conceição 818 12560 40.500 S, 41170 20.700 W A.S.Conceição 754 190 Table 2. A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) (Continued) Population Municipality CM4 C. cytisoides CFP1 CFP2 CFP3 Abaı́ra-BA ‘‘f. prostrata’’ Ibicoara-BA Abaı́ra-BA Barra da Estiva-BA C. cytisoides var. unijuga CU1 Santo Amaro das Brotas-SE CU2 Pirambu-SE Location Voucher Catolés-Bicota 13190 42.300 S, 41510 09.900 W A.S.Conceição 567 13.2 km de Ibicoara Catolés-Virassaia 7 km de Barra da Estiva 13260 37.800 S, 41170 1600 W 13500 S, 41220 W 13400 48.300 S, 41180 40.200 S A.S.Conceição 757 A.S.Conceição 672 A.S.Conceição 758 Ca. 2 km Rio Pomonga 8.2 km de Pirambu 10480 2400 S, 36570 5000 W A.S.Conceição 694 10400 5600 S, 36520 5100 W A.S.Conceição 820 using only vegetative morphological characters excluding variables #1 and #2 (both relative to the number of leaflets, which could force a separation of biand tetrafoliolate groups); 3) using only floral morphological characters (excluding CD2, CD3, CFP1, CB1, CC2, CC3, CC4, CC5, which were not found flowering); 4) only the populations which present two or three pairs of leaflets, using only vegetative morphological characters, and including for the first pair of leaflets the same variables used for the second pair (variable #3 to 9). We performed canonical variate analysis (CVA) and cluster analysis using the software package STATISTICA 6.0. CVA was performed with population as the categorical variable (individuals were grouped according to the population to which they belong). The standardized coefficients for canonical variables obtained from discriminant analysis were used to identify the characteristics that most significantly contributed to the resulting patterns observed. The cluster analysis was carried out using the Mahalanobis Generalized Distance and UPGMA as the clustering algorithm (Sneath and Sokal 1973). The morphological variability of populations was obtained using the median of the Mahalanobis distances of individuals to the centroid of the group (D2m) (Goldman et al. 2004; Lambert et al. 2006a, b). We used the median of the squared Mahalanobis distances instead of an average of these distances because of the non-normal distribution of the data. A multi-response permutation procedure (MRPP) analysis made using PCOrd 4.10 program (McCune and Mefford 1999) was used to calculate the chance-corrected withingroup agreement (AMRPP) among populations of every species, and AMRPP-values were compared with the indexes of genetic differentiation among populations (FST) (Borba et al. 2002; Lambert et al. 2006a, b). The average Euclidian distance among the individuals of each population (ED) resulting from de MRPP analysis was also used as a measure of morphological variability within populations (Borba et al. 2002; Lambert et al. 2006a, b). The two indexes of morphological variability are essentially different, as D2m is more affected by form and ED is more affected by size of the characters (Lambert et al. 2006a, b). A Spearman rank correlation analysis between the morphological (ED and D2m) and genetic (He) variability of populations was also carried out with STATISTICA 6.0. Results Variation within populations. Using nine enzymatic systems 15 loci were obtained with good resolution and were used in the study (Table 4). Three loci were monomorphic for all populations (ACPH, MDH and IDH-1). Five loci displayed low polymorphism, with only two or three alleles (PGI-2, EST-1, SKDH, G6PDH and LAP-2). The remaining loci present four to eight alleles per locus. PGM-1 was the most polymorphic, with eight alleles. Some alleles were exclusive to a single taxon: five to C. coriacea; one to C. cytisoides var. unijuga; five to C. cytisoides var. cytisoides; two to C. depauperata. A few alleles were exclusive to one population. The percentage of polymorphic loci (P; 0.95 criterion) ranged from 0.0 to 33.3%, the mean A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) number of alleles per locus was 1.1 to 1.4, and mean heterozygosity (He) ranged from 0.010 to 0.067 (Table 5). Populations CCY1 (C. cytisoides var. cytisoides), D1 (C. depauperata), CBR1 (C. cytisoides var. brachystachya), CD3 (C. cytisoides var. decora) and CU1 (C. cytisoides var. unijuga) displayed the highest genetic variability. The majority of the populations (72.7%) were in HW equilibrium in all loci. The other populations presented significant deviations in only one locus each. Five loci were not in HW equilibrium in at least one population: EST-1 (C1), G6PDH (CCY1), LAP-1 (CM2) and IDH-2 (in all populations of CC) with deficit of heterozygotes, and LAP-2 (CD3) with excess of heterozygotes. The low values for FIS reflect the HW equilibrium in most of the taxa, except for C. cytisoides var. confertiformis (CC) that presented very high FIS (Table 6). In both morphological analyses, the population with the highest variability was CBR1, which also presented one of the highest genetic variability (Table 5). The other populations presenting high morphological variability expressed as D2m were CM3, CBR5 and CU1, while the lowest variability was found in CC3 (Table 5). The most variable characters considering all taxa pooled were leaflet number (variable #1), number of main veins (#11), length of the leaflet (#3), width at 1/3 of the leaflet (#5), apex angle of the leaflet (#9), least distance from the margin to the main nervure, in the base of the leaflet (#7), length of the rachis (#12), and length of the petiole (#10). The least variable character was the distance of the nectary in relation to the second pair of leaflets (#16). Spearman rank correlations between morphological and genetic variability resulted in a statistically significant positive high correlation between He and D2m (r = 0.917, p £ 0.001) and between He and ED (r = 0.901, p £ 0.001). Genetic structuring. Chamaecrista cytisoides var. blanchetii and C. cytisoides var. brachystachya displayed very high values of FST, interpreted as a very high level of genetic structuring. The other taxa showed moderate to low average values of FST (Table 6). By excluding populations CB1 and CB5 of C. 191 cytisoides var. blanchetii and CBR2 of C. cytisoides var. brachystachya the average values of FST drop to 0.042 and 0.053, respectively. In C. cytisoides var. blanchetii this is due to the allele frequencies of IDH-2, and in C. cytisoides var. brachystachya to the loci PGM-1 and PGI-1. Morphological structuring. Chameacrista cytisoides var. brachystachya also presented the highest value of morphological structuring (AMRPP = 0.361), correlated with the highest values of genetic differentiation (FST). C. cytisoides var. confertiformis and C. cytisoides var. micrantha that presented high values of morphological differentiation had low values of genetic differentiation (Table 6). Phenetic relationships – genetic data. Populations of C. depauperata, C. cytisoides var. decora, C. cytisoides var. micrantha and C. cytisoides var. unijuga did not present any intrataxon genetic differentiation. The populations of C. cytisoides var. blanchetii, C. cytisoides var. confertiformis, and C. cytisoides ‘‘f. prostrata’’ also presented very high genetic identity among intra-taxon populations, ranging from 0.93 to 1.00. The lowest intra-taxon genetic identity (0.87) was found within C. cytisoides var. brachystachya, between the population CBR2 and the four remaining populations. The UPGMA dendrogram obtained from the cluster analysis of Nei’s (1978) unbiased genetic distances (Fig. 3) reveals the formation of two main groups. The first contains the population of C. cytisoides var. cytisoides (CCY) and the two populations of C. depauperata (D). The high divergence of C. cytisoides var. cytisoides and C. depauperata is due to the presence of exclusive alleles fixed in two enzymatic systems (both in PGM-2 and LAP-1), besides the same fixed allele in PGI-1, similar for the two taxa (Table 4). The second and largest group may be subdivided in three subgroups: one formed by all of the populations of C. cytisoides var. decora (CD) and population (C) of C. coriacea; another composed of the two populations of C. unijuga and population CBR2; the third formed by the populations of the remaining taxa. In this third subgroup only C. cytisoides var. confertiformis formed group containing exclusively all of their 192 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Table 3. Characters used in the morphometric analyses of 33 populations of nine described and one putative taxa of Chamaecrista sect. Absus subsect. Baseophyllum occurring in northeastern and southeastern Brazil Characters Vegetative characters Leaves (second par of leaflets) 1- Number of leaflets 2- Occurrence of variation in the number of leaflets 3- Length of leaflet 4- Maximum width of leaflet 5- Width of leaflet at 1/3 of its length 6- Width of leaflet at 2/3 7- Least distance from the margin to the main vein, in the base of the leaflet 8- Maximum distance from the margin to the main vein, in the base of the leaflet 9- Apex angle of the leaflet 10- Length of the petiole 11- Number of main veins 12- Length of the rachis 13- Length of the nectary 14- Width of the nectary 15- Number of nectaries 16- Distances of the nectary to the second par of leaflets Floral characters Inflorescence 17- Length 18- Number of flowers per raceme Sepal 19- Length of the first sepal 20- Width of the first sepal 21- Length of the second sepal 22- Width of the second sepal 23- Length of the third sepal 24- Width of the third sepal 25- Length of the fourth sepal 26- Width of the fourth sepal 27- Length of the fifth sepal 28- Width of the fifth sepal Petal 29- Length of the first petal 30- Maximum width of the first petal 31- Width of the first petal at 1/3 of its length 32- Width of the first petal at 2/3 33- Maximum distance from the margin to the main vein, in the base of the first petal Table 3. (Continued) Characters 34- Least distance from the margin to the main nervure, in the base of the first petal 35- Length of the second petal 36- Maximum width of the second petal 37- Width of the second petal at 1/3 38- Width of the second petal at 2/3 39- Length of the third petal 40- Maximum width of the third petal 41- Width of the third petal at 1/3 42- Width of the third petal at 2/3 43- Length of the fourth petal 44- Maximum width of the fourth petal 45- Width of the fourth petal at 1/3 46- Width of the fourth petal at 2/3 47- Length of the fifth petal 48- Maximum width of the fifth petal 48- Width of the fifth petal at 1/3 50- Width of the fifth petal at 2/3 Anther 51- Length of the anthers Style 52- Length of the style Ovary 53- Length of the ovary 54- Number of ovules intra-taxon populations. Populations of C. cytisoides var. blanchetii, C. cytisoides var. brachystachya, C. cytisoides var. micrantha and C. cytisoides ‘‘f. prostrata’’ formed mixed groups. Four populations of C. cytisoides var. brachystachya are nested in the group containing most of the populations of C. cytisoides var. blanchetii; the remaining population is very divergent (CBR2) due to the presence of two exclusive fixed alleles in two loci (Table 4). Phenetic relationships – morphological data. The scatterplots of the scores of individuals on the first two CVA canonical axes of the morphological analyses are shown in Fig. 4. The first and second canonical axes explained respectively 56% and 16% of the variation in the analysis including only vegetative characters. On the first canonical axis there is a separation of three groups, one with A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) C. cytisoides var. cytisoides, one with the populations of C. cytisoides var. confertiformis, C. coriacea and C. cytisoides var. brachystachya (except for CBR5), and another with C. unijuga, C. cytisoides var. decora, C. depauperata, C. cytisoides ‘‘f. prostrata’’, C. cytisoides var. blanchetii and population CBR5 (Fig. 4A). The variables that more contributed to that separation in the first axis were the leaflets number (variable #1), length of the second pair of leaflets (#3) and number of main veins (#11). In the second axis C. coriacea was separated from the remaining taxa in the second group, and the third group was divided in two subgroups, one containing C. unijuga, C. cytisoides var. decora and C. depauperata and the other with the remaining taxa. The variables that contributed most to the separation on this second canonical axis were the length of the petiole (variable #10) and the least distance from the margin to the main nervure in the base of the leaflets (#7). This structure had no significant changes in the analysis in which the variables #1 and #2 (both relative to the number of leaflets) were excluded (data not shown). In the matrix of classification the percentage of correct classifications ranged from 38 to 100%. The incorrect classifications mostly occurred between populations of the same taxon, except for eight individuals of CBR1 which were classified in CU1 and CB7, eight individuals of CBR5 in CB7 and CB2, and one individual of CBR3 in CM4. The first and second canonical axes explained 42% and 16%, respectively, of the variation in the analysis using only floral characters. In the first canonical axis the two populations of C. depauperata are separated from the other taxa (Fig. 4B). The variables that contributed most to this separation were maximum width of the fourth petal (variable #44), length of the anthers (#51), width at 2/3 of the fourth petal (#46) and number of ovules (#54). In the second axis occurred a slight separation of C. cytisoides var. cytisoides from the remaining taxa. The same structure observed in the CVA was found in the UPGMA dendrogram obtained from the cluster analysis of Mahalanobis distance among the centroids of the populations. The 193 analysis using vegetative morphological characters resulted in the separation of C. cytisoides var. cytisoides, the only taxon presenting predominantly three pairs of leaflets, and two other groups composed by the populations presenting one or two pair of leaflets (Fig. 5A). Populations of the same taxon are mostly present only in one of these two last groups, except for C. cytisoides var. brachystachya, with three population in one group and two in the other. In the subgroup presenting predominantly two pairs of leaflets, the population CBR4 and C. coriacea displayed the greatest differentiation, and the remaining populations were divided into two subgroups, one including all the populations of C. cytisoides var. confertiformis and the other composed by all populations of C. cytisoides var. micrantha plus the populations CBR2 and CBR3. In the subgroup presenting predominantly one pair of leaflets, two subgroups were formed: one composed of all populations which do not have a petiole, intermingled (C. cytisoides var. brachystachya, C. cytisoides ‘‘f. prostrata’’ and C. cytisoides var. blanchetii), and another of all populations that possess a petiole (C. unijuga, C. cytisoides var. decora and C. depauperata; the populations of each taxon grouping together in three subgroups). The morphological cluster analysis carried out only with the taxa presenting two or three pairs of leaflets resulted in the same general structure of the analysis carried out with only the second pair of leaflets. In the analyses using other clustering algorithms, the same structuring was found. In the analyses which variables 1 and 2 were excluded, the same groups were maintained, except for the population CBR4 grouping with CCY1. The UPGMA dendrogram obtained from the cluster analysis of morphological distance using only floral characters resulted in the formation of two main groups (Fig. 5B): a group containing only the populations of C. depauperata, and another larger group composed by the populations of all remaining taxa. This larger group presents high degree of chaining, with the populations of all taxa intermingled; no subgroups composed primarily by all populations of the same taxon were formed. 194 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Table 4. Allelic frequencies at 15 loci in populations of Chamaecrista coriacea (C), C. cytisoides var. cytisoides (CCY), C. depauperata (D), C. cytisoides var. decora (CD), C. cytisoides var. unijuga (CU), C. cytisoides ‘‘f. prostrata’’ (CFP), C. cytisoides var. micrantha (CM), C. cytisoides var. blanchetii (CB), C. cytisoides var. brachystachya (CBR) and C. cytisoides var. confertiformis (CC), occurring in northeastern and southeastern Brazil. N = sample size. See Table 2 for the name of the populations Locus C1 ACPH 100 (N) PGI-1 140 130 115 110 100 90 (N) PGI-2 100 95 (N) PGM-1 120 116 107 100 92 83 75 41 (N) PGM-2 106 100 93 81 (N) MDH 100 (N) EST-1 106 100 75 (N) EST-2 120 105 100 95 90 (N) EST-3 100 94 85 80 (N) SKDH 100 93 (N) G6PD 105 100 95 (N) IDH-1 100 (N) IDH-2 100 CCY1 D1 D2 CD1 CD2 CD3 CU1 CU2 CFP1 CFP2 CFP3 CM1 CM2 CM3 CM4 1.00 1.00 19 20 1.00 1.00 1.00 19 20 18 1.00 20 1.00 20 1.00 17 1.00 20 1.00 20 1.00 18 1.00 20 - 1.00 1.00 19 20 1.00 19 - - - 1.00 14 - - 11 1.00 10 1.00 11 1.00 18 1.00 1.00 19 20 1.00 1.00 1.00 19 20 17 1.00 18 1.00 11 1.00 14 1.00 14 1.00 15 - - - - 1.00 18 - - - - - 0.83 1.00 0.17 18 20 - 1.00 16 0.83 1.00 16 - 1.00 0.17 0.13 18 16 16 - 1.00 8 0.88 - 1.00 - 7 1.00 1.00 19 14 15 12 - 1.00 1.00 14 16 1.00 1.00 1.00 19 20 17 1.00 20 1.00 20 - 19 1.00 16 20 20 20 - - - - 0.82 0.83 0.18 0.18 19 20 0.85 0.88 0.15 0.12 17 16 1.00 12 0.2 0.70 0.10 20 1.00 1.00 15 17 1.00 1.00 1.00 1.00 0.84 0.16 0.07 - 1.00 8 0.05 1.00 17 1.00 6 0.11 1.00 18 1.00 13 1.00 12 1.00 13 1.00 16 1.00 19 1.00 20 1.00 13 1.00 18 1.00 15 1.00 17 - - - - - - 1.00 12 1.00 13 - - 0.03 0.97 14 19 1.00 19 - - - - - 1.00 1.00 - 1.00 20 1.00 20 1.00 11 20 20 20 8 0.05 13 10 1.00 13 1.00 20 1.00 20 1.00 20 1.00 20 20 20 20 20 - - - 0.16 7 1.00 6 0.03 1.00 7 1.00 - - - - - 0.98 20 0.84 19 1.00 20 1.00 10 1.00 0.05 0.8 0.15 20 0.98 0.16 0.03 16 20 20 20 20 20 20 20 20 20 - 0.96 0.04 14 0.98 0.02 20 0.98 0.02 20 0.95 0.05 19 0.98 0.02 20 0.98 0.02 20 0.96 0.04 13 0.94 0.06 16 0.94 0.06 17 0.94 0.06 18 0.97 0.03 19 0.97 0.98 0.97 0.03 0.02 0.03 18 20 15 - - - - - - - - - - - 0.96 0.04 14 0.96 0.04 11 0.96 0.04 13 0.94 0.06 16 0.95 0.05 20 0.95 0.05 20 0.93 0.07 20 0.97 0.03 17 0.96 0.04 13 0.97 0.03 19 0.98 0.02 20 1.00 1.00 1.00 15 15 14 1.00 11 1.00 9 1.00 16 1.00 18 1.00 20 1.00 20 1.00 20 1.00 19 1.00 14 1.00 11 1.00 20 1.00 1.00 - - - - - - - - - - 0.94 0.96 0.98 0.06 0.04 0.02 9 14 20 - 1.00 1.00 1.00 0.93 0.05 0.18 0.82 10 0.03 - 0.87 0.03 - 19 0.84 0.81 0.12 - 16 - 0.83 0.12 - - - - 1.00 9 8 - 20 17 0.86 0.07 - 1.00 20 - 8 1.00 - 0.06 - - 19 0.06 0.94 1.00 20 0.06 0.94 8 1.00 1.00 20 14 19 0.91 0.09 1.00 1.00 1.00 20 0.96 11 1.00 1.00 15 20 - - 6 - - - 1.00 - 1.00 0.81 0.19 18 20 0.80 1.00 0.20 5 19 - 1.00 1.00 8 1.00 20 1.00 - 6 1.00 11 1.00 - 5 1.00 - 8 8 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) 195 Table 4. (Continued) CB1 CB2 CB3 CB4 CB5 CB6 CB7 CBR1 CBR2 CBR3 CBR4 CBR5 CC1 CC2 CC3 CC4 CC5 1.00 20 1.00 20 1.00 18 1.00 20 1.00 20 1.00 15 1.00 20 1.00 20 1.00 20 1.00 20 1.00 16 1.00 20 1.00 20 1.00 11 1.00 17 1.00 12 1.00 19 - - - - - - - - - - - - - - - - - 1.00 18 1.00 20 1.00 18 1.00 1.00 - 1.00 - 20 20 18 14 20 18 20 13 20 1.00 18 1.00 20 1.00 19 1.00 14 1.00 20 1.00 18 1.00 14 1.00 20 1.00 16 1.00 16 1.00 19 1.00 12 0.96 0.04 13 - - - - - - - - - - - - - - 20 13 1.00 10 1.00 12 12 - - - - - 1.00 0.06 0.94 1.00 1.00 1.00 0.04 0.96 1.00 0.14 0.86 1.00 1.00 1.00 1.00 17 20 13 11 16 - - - - - - - 1.00 1.00 1.00 20 17 - 1.00 20 1.00 20 1.00 18 - - - 1.00 1.00 - 1.00 1.00 14 1.00 20 - - - 1.00 13 - - - - 1.00 20 1.00 13 1.00 20 - 0.97 0.04 15 0.98 0.02 20 0.96 0.04 13 0.97 0.03 18 0.98 0.02 20 - - - - 0.82 0.18 14 0.97 0.03 18 0.96 0.04 13 0.92 0.08 12 1.00 18 1.00 13 1.00 20 1.00 11 - - - 1.00 13 1.00 11 - - - 1.00 11 1.00 - 9 9 - 0.79 0.21 - 1.00 20 1.00 20 - - 6 1.00 - 6 0.03 1.00 0.85 0.12 1.00 8 1.00 - - 1.00 17 0.96 0.04 13 0.96 0.04 14 - - - 1.00 14 - - - 1.00 - 8 1.00 1.00 11 1.00 - 9 - 1.00 9 - 1.00 1.00 13 1.00 20 1.00 20 1.00 16 1.00 20 20 20 20 20 20 - - - - - 1.00 16 1.00 20 1.00 20 1.00 15 1.00 16 - 20 - - - - - 1.00 19 1.00 16 1.00 18 12 13 10 20 1.00 - 1.00 10 1.00 11 1.00 10 - 1.00 5 0.98 0.02 - 1.00 - 6 9 6 9 0.92 0.08 0.89 0.11 0.90 0.10 0.90 0.10 18 1.00 10 - 0.97 0.04 15 0.95 0.05 20 0.85 0.15 17 0.97 0.03 18 0.96 0.04 14 0.95 0.05 20 0.94 0.06 16 0.97 0.04 15 0.97 0.04 15 0.98 0.02 20 0.98 0.02 20 0.96 0.04 11 - - - - - - - - - - - - - 0.92 0.08 13 0.92 0.08 13 0.98 0.02 20 0.95 0.05 19 0.97 0.03 15 0.96 0.04 14 0.98 0.02 20 0.91 0.09 16 1.00 14 1.00 6 1.00 20 1.00 18 1.00 9 1.00 18 1.00 20 1.00 20 - - - - - 8 1.00 - 8 1.00 - 7 1.00 6 - 1.00 1.00 - 6 5 - 5 1.00 11 1.00 - 10 0.92 0.08 20 7 20 1.00 9 15 - 1.00 1.00 19 - 9 1.00 1.00 14 - 0.17 0.81 0.03 1.00 9 20 12 8 - 1.00 0.96 0.04 7 17 16 1.00 - - 1.00 - 6 - 1.00 18 1.00 7 1.00 - - 1.00 0.10 0.90 1.00 8 1.00 12 0.92 0.08 - 1.00 17 8 6 1.00 - 1.00 6 8 1.00 1.00 1.00 - 18 9 1.00 1.00 1.00 13 1.00 - 8 7 1.00 0.917 - 1.00 0.083 12 1.00 10 - 9 13 1.00 5 1.00 7 1.00 11 1.00 12 1.00 6 0.2 0.08 0.64 0.29 0.18 9 196 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Table 4. (Continued) Locus C1 CCY1 D1 D2 CD1 CD2 CD3 CU1 CU2 CFP1 CFP2 CFP3 CM1 CM2 CM3 CM4 93 90 (N) LAP-1 100 93 90 89 (N) LAP-2 117 100 94 (N) 17 15 15 - - 14 10 1.00 17 19 1.00 1.00 12 9 - - - - - 1.00 - 17 1.00 19 1.00 1.00 17 1.00 19 1.00 20 0.17 0.83 12 11 9 1.00 - 1.00 - 8 7 0.17 0.83 0.25 0.75 20 12 - - 1.00 16 1.00 18 0.07 0.93 20 1.00 14 1.00 10 1.00 12 - 0.19 0.81 0.19 0.81 - - 16 Discussion Genetic variability. The genetic variation found within populations for all taxa of the subsection Baseophyllum are low when compared with values found by Hamrick and Godt (1992) for plants of regional and endemic geographical distribution, respectively (P = 36.4, A = 1.55, and P = 26.3, A = 1.39, He = 0.118 He = 0.063). However, the variability levels found in this study were similar to the values found for other genera of Leguminosae (e.g. Baatout et al. 1991, Liston 1992, Chamberlain 1998, Casiva et al. 2002, Sotuyo et al. 2004). The low variability presented within populations for all taxa of the subsection Baseophyllum may be associated with the pollinator behavior and mating systems. These plants are selfcompatible and pollinated by social bees presenting a behavior which favors endogamy (Costa 2007, Silva-Pereira 2007). The populations of C. cytisoides var. micrantha and C. depauperata which had the highest diversity (CM3 and D1, respectively) are located in preserved areas, where human activity is restricted. Conversely, population CC2 of C. cytisoides var. confertiformis occurring in a site under anthropogenic action displayed the lowest genetic diversity of the taxon, in spite of its moderate morphological variability. The populations of C. cytisoides var. blanchetii with slightly higher genetic diversity are located in disjunct areas, CB1 in the restinga of Espı́rito Santo state, and CB5 in the savanna of Minas Gerais state, whilst the other populations occur in areas of 16 1.00 12 - - - - 1.00 20 1.00 20 1.00 19 1.00 14 1.00 9 0.05 0.95 20 1.00 1.00 10 1.00 19 0.95 1.00 10 1.00 16 - - - - 8 1.00 20 1.00 10 1.00 16 - 0.05 20 20 1.00 15 1.00 16 ‘‘campos rupestres’’ in Bahia state. These two disjunct populations presented an exclusive fixed allelle in the locus IDH-2, indicating a restricted gene flow among these three major areas of occurrence of the taxon. Chamaecrista cytisoides var. confertiformis was the only taxon presenting high values of FIS, which could be the result of endogamy or substructuring of the populations. This taxon displays a singular spatial distribution for subsection Baseophyllum, almost always occurring in valleys along the margins of rivers, forming dense linear populations. In addition to this spatial distribution, self-compatibility and the behavior of its pollinator favors endogamy and nonrandom mating (Costa 2007), probably leading to genetic sub-structuring of the populations. Morphologic variability. Studies of intraand inter-populational multivariate morphological variability are apparently absent in Chamaecrista and scarce in Leguminosae (Riggins et al. 1977, Casiva et al. 2002, Leht and Jaaska 2002, Boonkerd et al. 2005). The number of leaflets is not the only characteristics that forced the separation of the three groups in the multivariate analyses. The length of the rachis, apex angle of the leaflet, length of the leaflets, and least distance from the margin to the main vein in the base of the leaflet sustained the structure of three groups, as the same groups were maintained in the analysis excluding the variables 1 and 2. It is important notice since that these characters were not evidenced by Irwin and Barneby (1978) or by Bentham (1870, 1871) in A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) 197 Table 4. (Continued) CB1 CB2 CB3 CB4 CB5 CB6 CB7 CBR1 CBR2 CBR3 CBR4 CBR5 CC1 CC2 CC3 CC4 CC5 18 1.00 6 1.00 12 - - - 1.00 13 1.00 17 1.00 12 1.00 13 1.00 16 1.00 10 - - - - - 0.97 0.03 15 1.00 14 12 1.00 9 1.00 - 1.00 - 7 7 1.00 6 - - - - 1.00 20 1.00 18 1.00 9 1.00 18 1.00 20 1.00 9 1.00 20 1.00 11 1.00 11 1.00 16 - 1.00 9 - - - - - 1.00 20 - 1.00 12 1.00 16 the distinction of the taxa in the subsection Baseophyllum. Taxonomic implications. Chamaecrista coriacea was recognized (as Cassia) by Bentham (1840) as a peculiar species, characterized by one or two pairs of palmately veined leaflets, and its status has never been objected (Irwin and Barneby 1978, 1982). Both allozyme and morphological data supported its distinction at specific rank. Chamaecrista coriacea differs from the other taxa of the subsection Baseophyllum in at least one allozyme locus, with a genetic identity of 0.78 to C. cytisoides var. cytisoides. In the discriminant analysis C. coriacea differed from C. cytisoides var. cytisoides mainly is its leaves predominantly with two pairs of leaflets, smaller rachis and smaller leaflets. The recently described C. depauperata was the only taxon with clearly distinctive flowers, as evidenced in the morphological analyses of floral characters, mainly due to the width of fourth petal, length of the anthers and fewer ovules. In the analyses using vegetative morphological data, C. depauperata also differed of the other taxa by presenting petiolate leaves with two pairs of smaller leaflets. Furthermore, this species is very distinctive genetically in at least three allozyme loci (two of them being diagnostic and the other similar to C. cytisoides var. cytisoides). Chamaecrista depauperata and C. cytisoides var. cytisoides are quite different in both habit and habitat (prostrate sub-shrubs from ‘‘campo rupestre’’ vegetation vs. shrubs to 1.00 16 0.05 0.95 20 1.00 1.00 - 5 6 - 1.00 10 1.00 16 - - - - - 0.8 10 0.92 12 0.36 14 0.71 14 0.82 11 1.00 16 1.00 13 1.00 11 1.00 11 - - - - - - 1.00 16 1.00 13 1.00 10 1.00 11 1.00 9 1.00 15 treelets from ‘‘cerrado’’), and in number and form of the leaflets (Table 1). Besides, C. depauperata occurs exclusively in the state of Bahia in the southern portion of the Chapada Diamantina, while C. cytisoides var. cytisoides is endemic to southern portion of Minas Gerais state (Fig. 1). The high similarity in floral traits has probably lead Irwin and Barneby (1978) to treat some of the former species of the subsection Baseophyllum as varieties of C. cytisoides. This picture was possibly reinforced by the occurrence of hybridization in some taxa, such as between C. cytisoides var. confertiformis and C. cytisoides var. blanchetii and between C. cytisoides var. confertiformis and C. depauperata (Conceição et al. unpubl.). These hybrids are very rare in nature due to several isolation barriers between the sympatric species, such as different flowering period in C. cytisoides var. confertiformis and C. cytisoides var. blanchetii (Costa 2007), but may create confusion among taxonomists when they examine only herbarium collections. In spite of being reproductively viable, apparently these hybrids are not stable and do not compromise taxon circumscription in this group (Costa 2007). The values of genetic identities found among populations of the C. cytisoides complex were similar to the values usually found between populations of distinct species (Thorpe 1982, Crawford 1989, van der Bank et al. 2001). The average genetic identities between the population where the holotype of C. cytisoides var. cytiso- 198 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Table 5. Genetic variability at 15 allozymic loci and morphological variability (D2m and ED) based on the morphometric analyses of 54 characters in 33 populations of nine described and one putative taxa of Chamaecrista sect. Absus subsect. Baseophyllum, occurring in northeastern and southeastern Brazil. N = mean sample size per locus; genetic parameters - A = mean number of alleles per locus, P = percentage of polymorphic loci, Ho = observed and He = expected mean heterozygosity per locus (Nei, 1978; unbiased estimate); morphological parameters - D2m = median of the Mahalanobis Generalized Distance of the individuals to the centroid of the population, ED mean of the Euclidean distance between the individuals of the population. Standard deviations in parentheses. A locus was considered polymorphic if the frequency of the most common allele did not exceed 0.95. See Table 2 for the name of the populations Population N P Ho He (0.1) 33.3 0.092 (0.041) 0.010 (0.039) 8.98 - (0.1) (0.1) 26.7 13.3 0.075 (0.034) 0.040 (0.020) 0.067 (0.029) 0.038 (0.018) 5.88 6.40 12.26 10.62 (0.1) (0.1) (0.1) (0.1) (0.1) (0.1) (0.2) 13.3 0.0 6.7 13.3 13.3 6.7 13.3 0.036 0.011 0.026 0.033 0.042 0.015 0.030 (0.025) (0.006) (0.012) (0.021) (0.030) (0.011) (0.021) 0.032 0.011 0.025 0.031 0.037 0.014 0.028 (0.021) (0.006) (0.012) (0.019) (0.025) (0.011) (0.019) 9.34 12.19 5.91 15.88 5.94 6.73 17.47 18.11 19.37 14.37 21.37 15.55 13.52 24.28 (0.2) (0.1) (0.1) (0.1) (0.1) 26.7 0.0 0.0 6.7 20.0 0.066 0.011 0.010 0.010 0.033 (0.033) (0.006) (0.006) (0.007) (0.016) 0.058 0.011 0.010 0.010 0.032 (0.028) (0.006) (0.006) (0.007) (0.015) 38.13 7.17 17.76 14.34 20.70 32.47 18.02 16.24 19.55 21.44 (0.1) (0.1) (0.1) (0.1) (0.1) 13.3 13.3 13.3 13.3 20.0 0.021 0.020 0.017 0.022 0.032 (0.012) (0.016) (0.014) (0.014) (0.015) 0.043 0.029 0.048 0.049 0.052 (0.024) (0.017) (0.033) (0.030) (0.024) 5.33 4.25 3.35 5.75 8.47 12.81 16.38 13.88 12.46 16.01 (0.2) 20.0 0.053 (0.029) 0.066 (0.038) 7.12 - (0.2) (0.2) (0.2) 13.3 13.3 13.3 0.049 (0.028) 0.056 (0.031) 0.068 (0.048) 0.045 (0.025) 0.050 (0.027) 0.058 (0.033) 13.61 11.93 11.47 20.71 18.62 16.85 (0.1) (0.1) (0.2) (0.1) 6.7 13.3 13.3 6.7 0.012 0.013 0.031 0.014 0.012 0.019 0.034 0.013 (0.009) (0.010) (0.023) (0.008) 10.89 4.55 20.90 5.57 17.05 9.10 19.64 14.10 (0.1) (0.1) 20.0 13.3 0.034 (0.014) 0.037 (0.024) 0.034 (0.014) 0.033 (0.021) 5.38 8.52 15.38 19.90 A C. coriacea C1 15.6 (1.1) 1.3 C. depauperata D1 17.6 (0.7) 1.3 D2 17.0 (0.7) 1.3 C. cytisoides var. blanchetii CB1 16.7 (1.1) 1.2 CB2 16.3 (1.0) 1.2 CB3 14.4 (1.1) 1.3 CB4 12.3 (1.0) 1.2 CB5 13.6 (1.4) 1.2 CB6 11.0 (1.3) 1.1 CB7 16.1 (1.5) 1.3 C. cytisoides var. brachystachya CBR1 15.9 (1.2) 1.3 CBR2 13.5 (1.3) 1.2 CBR3 14.5 (1.6) 1.1 CBR4 14.7 (1.5) 1.1 CBR5 14.8 (1.1) 1.3 C. cytisoides var. confertiformis CC1 11.1 (1.3) 1.3 CC2 10.6 (1.1) 1.2 CC3 13.1 (1.4) 1.2 CC4 12.3 (1.2) 1.3 CC5 11.6 (1.2) 1.3 C. cytisoides var. cytisoides CCY1 18.6 (0.4) 1.3 C. cytisoides var. decora CD1 13.3 (1.4) 1.3 CD2 12.3 (1.7) 1.3 CD3 10.8 (1.6) 1.3 C. cytisoides var. micrantha CM1 14.2 (1.6) 1.1 CM2 12.3 (1.7) 1.2 CM3 13.4 (1.4) 1.3 CM4 15.3 (1.4) 1.2 C. cytisoides ‘‘f. prostrata’’ CFP1 13.3 (1.8) 1.3 CFP2 15.3 (1.8) 1.3 (0.0009) (0.0009) (0.021) (0.008) D2m ED A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Table 5. Population 199 (Continued) N CFP3 13.9 (1.8) C. cytisoides var. unijuga CU1 14.7 (1.3) CU2 13.7 (1.5) A P Ho He D2m ED 1.3 (0.1) 13.3 0.040 (0.023) 0.036 (0.020) 8.54 18.02 1.4 (0.2) 1.3 (0.2) 20.0 26.7 0.059 (0.029) 0.050 (0.027) 0.054 (0.026) 0.046 (0.023) 20.47 15.94 23.48 23.22 Table 6. F statistics (Wright 1978) at 15 loci and AMRPP from morphometric analysis in populations of seven taxa and one morphospecies of Chamaecrista sect. Absus subsect. Baseophyllum occurring in northeastern and southeastern Brazil Taxon No. of Pops FIS FST AMRPP C. C. C. C. C. C. C. C. 2 7 5 5 3 4 3 2 –0.141 –0.122 –0.105 0.476 –0.177 0.084 –0.111 –0.141 0.016 0.539 0.648 0.109 0.005 0.049 0.035 0.007 0.128 0.182 0.361 0.335 0.055 0.321 0.070 0.080 depauperata cytisoides var. blanchetii cytisoides var. brachystachya cytisoides var. confertiformis cytisoides var. decora cytisoides var. micrantha cytisoides ‘‘f. prostrata’’ cytisoides var. unijuga ides was collected and all other taxa in this complex are lower than the values found between conspecific populations, ranging from 0.64 to 0.78. Conversely, the genetic identity of that population with both C. depauperata and C. coriacea was higher. Chamaecrista cytisoides var. cytisoides was also the most divergent population in the morphological analyses based on vegetative characters, mainly due to its leaves with three pairs of leaflets. Both low values of genetic identity and high morphological differences of the type taxon for the complex with the other conspecific varieties reinforce the hypothesis of the existence of distinct species in the complex. These results point out that only the population of C. cytisoides var. cytisoides sampled, among the 33 populations studied, should remain in the species. The populations of C. cytisoides var. confertiformis, C. cytisoides var. decora and C. cytisoides var. unijuga formed three very distinctive groups in both genetic and morphological analyses. These three taxa present different geographic distribution, C. cytisoides var. confertiformis being endemic to interior Bahia, C. cytisoides var. decora endemic to inland Minas Gerais, and C. cytisoides var. unijuga endemic to coastal areas of Sergipe and Alagoas states. Therefore, we propose they should be treated as distinct species, which may be easily recognized by morphological differences in their leaves: shortly petiolate leaves with two pairs of oblong to oblong-lanceolate leaflets in C. confertiformis, petiolate leaves with one pair of obliquely obovate leaflets in C. decora, and petiolate leaves with one pair of orbicular to reniform leaflets in C. unijuga. The populations of C. cystisoides var. blanchetii, C. cytisoides var. brachystachya, C. cystisoides var. micrantha and C. cytisoides ‘‘f. prostrata’’ presented a reticulate pattern of grouping in both genetic and morphological analyses, but with different topology of phenetic relationships in each analysis. Almost all populations presented very high genetic identities, except for three disjunct populations of C. cystisoides var. blanchetii (CB1 and CB5) and C. cytisoides var. brachystachya (CBR2). The differences pointed out by Conceição (2000) in the populations named as C. cytisoides ‘‘f. prostrata’’ are not 200 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Fig. 3. Dendrogram showing the phenetic relationships among 33 populations of nine described and one putative taxa of Chamaecrista sect. Absus subsect. Baseophyllum occurring in northeastern and southeastern Brazil. Constructed using the matrix of genetic distances (Nei, 1978; unbiased estimate) based on 15 allozymic loci with UPGMA as clustering algorithm. See Table 2 for the names of the populations consistent with our analyses, and we suggest that these populations should not be treated as a distinct taxon. Populations of C. cytisoides ‘‘f. prostrata’’ are morphologically very similar to C. cytisoides var. blanchetii, and the only genetic distinction was the absence of the loci EST-1 and EST-3 in C. cytisoides ‘‘f. prostrata’’. A similar picture was found in Acianthera (Orchidaceae) species of campo rupestre vegetation by Borba et al. (2001). Populations CBR2 (from which the holotype was collected) and CBR3 of C. cytisoides var. brachystachya are very similar morphologically to C. cytisoides var. micrantha, in spite of the genetic differences of the former population. These two populations occur in the interior of Brazil, as also C. cytisoides var. micrantha, near the central (Pernambuco state, CBR3) and northern (center of Minas Gerais state, CBR2) extremes of distribution of the group (C. cytiso- c Fig. 4. Representation of the scores on the two first canonical axes of the CVA using 16 vegetative (A) and 38 floral (B) characters in 33 populations of nine described and one putative taxa of Chamaecrista sect. Absus subsect. Baseophyllum occurring in northeastern and southeastern Brazil. See Table 2 for the names of the populations A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) 201 202 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) c Fig. 5. Dendrogram showing the phenetic relationships among 33 populations of nine described and one putative taxa of Chamaecrista sect. Absus subsect. Baseophyllum occurring in northeastern and southeastern Brazil. Constructed using the matrix of Mahalanobis generalized distance based on 16 vegetative (A) and 38 floral (B) characters. See Table 2 for the names of the populations ides var. micrantha occupies an intermediate distribution, occurring in Bahia and northern Minas Gerais states). The other three populations of C. cytisoides var. brachystachya are very similar morphologically to C. cytisoides var. blanchetii, differing by their occurrence near the coast (versus interior of Bahia in C. cytisoides var. blanchetii) and by the presence of two pairs of leaflets. These three populations have been identified as C. cytisoides var. brachystachya due to the presence of two pair of leaflets and probably by second pair being similar in morphology to that found in individuals of the population CBR2. However, the first pair of leaflets is very similar to the single pair found in C. cytisoides var. blanchetii. Based on our present data, we suggest a broad circumscription of two polymorphic species in this group: (1) C. brachystachya, including the populations CBR2 and CBR3 (C. cytisoides var. brachystachya s.str. Crom.) and C. cystisoides var. micrantha populations, and (2) C. blanchetii, including the populations of C. cytisoides ‘‘f. prostrata’’ and populations CBR1, CBR4 and CBR5 formerly identified as C. cytisoides var. brachystachya. This decision is based mainly on morphology, but it is also supported by genetic data and by the occurrence of these two taxa in sympatry with no interbreeding (Costa 2007). In the multivariate morphological analyses the major characters which separated the populations of these two species were number of leaflets, width of leaflet at 1/3 of its length, angle of the apex of the leaflet, least distance from the margin to the main vein in the base of the leaflet, and length of the petiole. Taxonomic conclusions. The results of multivariate morphological and allozyme data in populations of all taxa of Chamaecrista sect. Absus subsect. Baseophyllum showed that several distinct species exist in the subsection. Genetics and morphology confirmed the circumscription and specific rank for C. coriacea and C. depauperata, corroborating the delimitation proposed by Bentham (1870, 1871) and Conceição et al. (2001), respectively. However, the present analyses did not support the delimitation proposed by Irwin and Barneby (1978, 1982) for the Chamaecrista cytisoides complex. Our data showed that the type taxon of the complex, C. cytisoides var. cytisoides, is genetic and morphologically very distant from all the other recognized varieties, and thus the complex may be re-arranged in six species based on C. cytisoides var. blanchetii, C. cytisoides var. brachystachya, C. cytisoides var. confertiformis, C. cytisoides var. cytisoides, C. cytisoides var. decora and C. cytisoides var. unijuga. On the other hand, C. cytisoides var. micrantha is proposed as synonym of C. cytisoides var. brachystachya. Chamaecrista blanchetii is a variable species, with sessile leaves or bearing a minute petiole, with one to two (rarely three) pairs of coriaceous leaflets, the proximal pair reniform to sub-orbicular, base reniform to obtuse, frequently embracing the stem, and the distal pair orbicular with base obtuse. Chamaecrista brachystachya is also a variable species, with petiolate leaves and two pairs of coriaceous to semi-succulent leaflets, the proximal pair orbicular to obovate with cordate to cuneate base, the distal pair oblong to suborbicular with cuneate base. The genetic differences presented by the population CBR2, as generally observed in allozyme differentiation, are probably a result of genetic drift or directional selection. The presence of two fixed exclusive alleles in this population can be interpreted as a result of its geographic isolation; gene flow between CBR2 and the remaining C. brachystachya populations is possibly very low or nonexistent, and genetic drift could be responsible A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) 203 204 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) for this differentiation. This process can quickly lead to speciation in small, geographically isolated populations (Levin 2000). Below we present a synopsis of the taxa belonging to Chamaecrista sect. Absus subsect Baseophyllum as recognized in this work. The taxonomic treatment of the group, including descriptions, illustrations and identification keys, is being published elsewhere. 1. Chamacrista blanchetii (Benth.) Conc., L.P. Queiroz & G.P. Lewis, comb. nov. Basionym: Cassia blanchetii Benth., J. Bot. (Hooker) 2: 78 (1840). Type: ‘‘Serra de Jacobina’’, s.d., Blanchet 2649 (holotype: K! [photo: HEFS!]; isotypes [numbered 2549] B, BM!, C, F, G!, K!, LE, MG, MO, NY!, P, SP!, [photo: HUEFS!]). : Cassia cytisoides var. blanchetii (Benth.) H.S.Irwin & Barneby, Mem. New York Bot. Gard. 30: 13 (1978). : Chamaecrista cytisoides var. blanchetii (Benth.) H.S.Irwin & Barneby, Mem. New York Bot. Gard. 35: 647 (1982). 2. Chamaecrista brachystachya (Benth.) Conc., L.P. Queiroz & G.P. Lewis, comb. nov. Basionym: Cassia brachystachya Benth., J. Bot. (Hooker) 2: 78 (1840). Type: ‘‘Tejuco, Herb, Acad. Petrop.’’, s.d., Riedel 566 (lectotype: K-Benth.!; isolectotype: LE!). : Cassia cytisoides var. brachystachya (Benth.) H.S. Irwin & Barneby, Mem. New York Bot. Gard. 30: 13 (1978). : Chamaecrista cytisoides var. brachystachya (Benth.) H.S. Irwin & Barneby, H.S.Irwin & Barneby, Mem. New York Bot. Gard. 35: 647 (1982). = Cassia cytisoides var. micrantha H.S. Irwin & Barneby, Mem. New York Bot. Gard. 30: 12 (1978). Type: Brazil, Bahia, Andarahy, Chapada Diamantina, ChiqueChique (Xique-Xique), April 1938, Bondar 2610 (holotype: SP!). : Chamaecrista cytisoides var. micrantha (H.S.Irwin & Barneby) H.S.Irwin & Barneby, Mem. New York Bot. Gard. 35: 647 (1982). 3. Chamaecrista confertiformis (H.S. Irwin & Barneby) Conc., L.P.Queiroz & G.P. Lewis, comb. et stat. nov. Basionym: Cassia cytisoides var. confertiformis H.S.Irwin & Barneby, Mem. New York Bot. Gard. 30: 12 (1978). Type: Brazil, Bahia, ± km S. of Mucugê, along rio Cumbuca, R.M. Harley 15936 (holotype: NY!; isotypes: K!, P!). : Chamaecrista cytisoides var. confertiformis (H.S. Irwin & Barneby) H.S. Irwin & Barneby, Mem. New York Bot. Gard. 35: 647 (1982). 4. Chamaecrista coriacea (Benth.) H.S. Irwin & Barneby, Mem. New York Bot. Gard. 35(2): 647 (1982). Type: ‘‘Inter Serra del frio et Cachoeira, Herb., Acad. Petrop.’’, 01 Jan. 1825, Riedel 563 (holotype: LE [photoK!]; isotypes: A, K! [photos: HUEFS!, IPA!]). 5. Chamaecrista cytisoides (Collad.) H.S Irwin & Barneby, Mem. New York Bot. Gard. 35: 647 (1982). Type: ‘‘Hab. in Brasilia’’ (holotype: G-DC!; probable isotype: P!) 6. Chamaecrista decora (H.S.Irwin & Barneby) Conc., L.P. Queiroz & G.P. Lewis, comb. et stat. nov. Basionym: Cassia cytisoides var. decora H.S. Irwin & Barneby, Mem. New York Bot. Gard. 30: 14 (1978). Brazil, Minas Gerais, Serra do Espinhaço, ±18 km E of Diamantina, 19 March 1970, Irwin et al. 27894 (holotype: UB; isotypes: C, F, GH, K!, NY!, RB!, UC, US). : Chamaecrista cytisoides var. decora (H.S.Irwin & Barneby) H.S.Irwin & Barneby, New York Bot. Gard. 35: 647 (1982). 7. Chamaecrista depauperata Conc., L.P. Queiroz & G.P. Lewis, Sitientibus Sér. Ci. Biol. A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) 1(2): 114 (2001). Type: Brazil, Bahia, Abaı́ra, distrito de Catolés, Campo do Virassaia, 30 Dec. 1993, W. Ganev 2722 (holotype: HUEFS!; isotypes: K!, SPF!). 8. Chamaecrista unijuga (Benth.) Conc., L.P.Queiroz & G.P.Lewis, comb. et stat. nov. Basionym: Cassia brachystachya var. unijuga Benth., Fl. Bras. (Martius) 15(2): 139. 1870. Type: ‘‘in maritimis prope Maceio prov. Alagoas’’, April 1848, Gardner 1282 (lectotype: K!; isolectotype: BM!). : Cassia cytisoides var. unijuga (Benth.) H.S. Irwin & Barneby, Mem. New York Bot. Gard. 30: 14 (1978). : Chamaecrista cytisoides var. unijuga (Benth.) H.S. Irwin & Barneby, New York Bot. Gard. 35: 647 (1982). We thank the Fundação de Amparo à Pesquisa do Estado da Bahia (FAPESB) for financial support (#245/04). ASC received a fellowship from Universidade do Estado da Bahia (PAC-UNEB). ELB and LPQ are supported by a grant (PQ2) from Conselho Nacional de Desenvolvimento Cientı́fico e Tecnológico (CNPq). References Avise JC (1994) Molecular markers, natural history and evolution Chapman & Hall, New York Baatout H, Marrakchi M, Combes D (1991) Genetic divergence and allozyme within and among populations of Hedysarum spinosissimum subsp. capitatum and subsp. spinosissimum (Papilionaceae). Taxon 40: 239–252 Bentham G (1840) Leguminosae. J Bot (Hooker ) 2: 78 Bentham G (1870) Cassia. In: Martius (ed) F. Brasiliensis 15: 82–176 Bentham G (1871) Revision of the genus Cassia. Trans Linn Soc London 27: 503–591 Boonkerd T, Pechsri S, Baum BR (2005) A phenetic study of Cassia sensu lato (Leguminosae-Caesalpinioideae: Cassieae: Cassiinae) in Thailand. Pl. Syst Evol 232: 153–165 Borba EL, Felix JM, Semir J, Solferini VN (2000) Pleurothallis fabiobarrosii, a new Brazilian spe- 205 cies: morphological and genetic data with notes on the taxonomy of Brazilian rupicolous Pleurothallis. Lindleyana 15: 2–9 Borba EL, Felix JM, Solferini VN, Semir J (2001) Fly-pollinated Pleurothallis (Orchidaceae) species have high genetic variability: evidence from isozyme markers. Amer J Bot 88: 419–428 Borba EL, Shepherd GJ, van den Berg C, Semir J (2002) Floral and vegetative morphometrics of five Pleurothalis (Orchidaceae) species: correlation with taxonomy, phylogeny, genetic variability and pollination systems. Ann Bot 90: 219–230 Brune W, Alfenas AC, Junghans TG (1998) Identificações especı́ficas de enzimas em géis. In: Alfenas AC (ed) Eletroforese de isoenzimas e proteinas afins: fundamentos e aplicações em plantas e microorganismos, Universidade Federal de ViçosaViçosa, pp 201–328 Casiva PV, Saidman BO, Vilardi JC, Cialdella AM (2002) First comparative phenetic studies of Argentinean species of Acacia (Fabaceae), using morphometric, isozymal, and RAPD approaches. Amer J Bot 89: 843–853 Chamberlain J (1998) Isozyme variation in Calliandra calothyrsus (Leguminosae): its implications for species delimitation and conservation. Amer J Bot 85: 37–47 Clayton JW, Tretiak DN (1972) Amine-citrate buffers for pH control in starch gel eletrophoresis. J Biol Board Canada 29: 1169–1172 Conceição AS (2000) O gênero Chamaecrista Moench (Leguminosae-Caesalpinioideae) em Catolés, Abaı́ra, Chapada Diamantina-BA-Brasil. M.Sc. Thesis, Universidade Federal de Pernambuco, Recife Conceição AS (2006) Filogenia do gênero Chamaecrista (Leguminosae-Caesalpinioideae) e taxonomia do grupo Baseophyllum. Ph.D. Thesis, Universidade Estadual de Feira de Santana, Feira de Santana Conceição AS, Queiroz LP, Lewis GP (2001) Novas espécies de Chamaecrista Moench (LeguminosaeCaesalpinioideae) da Chapada Diamantina, Bahia, Brasil. Sitientibus série Ciências Biológicas 1: 112–119 Corrias B, Rossi W, Arduino P, Cianchi R, Bullini L (1991) Orchis longicornu Poiret in Sardinina: genetic, morphological and chorological data. Webbia 45: 71–101 Costa CBN (2007) Mecanismos de isolamento reprodutivo em espécies simpátricas: biologia reprodutiva de Chamaecrista (LeguminosaeCaesalpinioideae) em Mucugê, Bahia. Ph.D. 206 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Thesis, Universidade Estadual de Feira de Santana, Feira de Santana Crawford DJ (1989) Enzyme electrophoresis and plant systematics. In: Soltis DE, Soltis PS (eds) Isozymes in plant biology, Dioscorides Press Portland, pp 146–164 Elisens WJ, Boyd RD, Wolfe AD (1992) Genetic and morphological divergence among varieties of Aphanostephus skirrhobasis (Asteraceae-Astereae) and related species with different chromosome numbers. Syst Bot 17: 380–394 Gilles BE (1984) A comparison between quantitative and biochemical variation in the wild barley Hordeum murinum. Evolution 38: 34–41 Goldman DH, van den Berg C, Griffith MP (2004) Morphometric circumscription of species and infraspecific taxa in Calopogon R. Br. (Orchidaceae). Pl Syst Evol 274: 37–60 Hamrick J L, Godt M J (1992) Allozyme diversity in plant species. In: Brown AHD, Clegg MT, Kahler AL, Weir BS (eds) Plant population genetics, breeding and genetic resources, Sinauer Sunderland, pp 43–63 Henderson A (2006) Traditional morphometrics in plant systematics and its role in palm systematics. Bot J Linn Soc 151: 103–111 Irwin HS, Barneby RC (1978) Monographic studies in Cassia (Leguminosae). III. Sections Absus and Grimaldia. Mem New York Bot Gard 30: 1–300 Irwin HS, Barneby RC (1982) The American Cassinae: a synoptical revision of Leguminosae tribe Cassieae subtribe Cassinae in the New World. Mem New York Bot Gard 35: 1–918 Lambert SM, Borba EL, Machado MC, Andrade SCS (2006a) Allozyme diversity and morphometrics of Melocactus paucispinus (Cactaceae) and evidence for hybridization with M. concinnus in the Chapada Diamantina, North-eastern Brazil. Ann Bot 97: 389–403 Lambert SM, Borba EL, Machado MC (2006b) Allozyme diversity and morphometrics of the endangered Melocactus glaucescens (Cactaceae), and investigation of the putative hybrid origin of Melocactus · albicephalus (Melocactus ernestii · M. glaucescens) in north-eastern Brazil. Pl Spec Biol 21: 93–108 Leht M, Jaaska V (2002) Cladistic and phenetic analysis of relationships in Vicia subgenus Vicia (Fabaceae) by morphology and isozymes. Pl Syst Evol 232: 237–260 Levene H (1949) On a matching problem arising in genetics. Ann Math Statist 20: 91–94 Levin DA (2000) The origin, expansion, and demise of plant species Oxford University Press, New York Lewis GP (2005) Tribe Cassieae. In: Lewis GP, Schrire B, MacKinder B, Lock M (eds) Legumes of the world, Royal Botanical Gardens, Kew, pp 111– 124 Liston A (1992) Isozyme systematics of Astragalus sect. Leptocarpi subsect. Californici (Fabaceae). Syst Bot 17: 367–379 McCune B, Mefford MJ (1999) PCOrd-Multivariate analysis of ecological data, version 4.10. MjM Software, Gleneder Beach Mitton JB (1978) Relationship between heterozygosity for enzyme loci and variation of morphological characters in natural populations. Nature 273: 661– 662 Nei M (1978) Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89: 583–590 Pereira ACS, Borba EL, Giulietti AM (2007) Genetic and morphological variability of the endangered Syngonanthus mucugensis Giul. (Eriocaulaceae), from the Chapada Diamantina, Brazil implications for conservation and taxonomy. Bot J Linn Soc 153: 401–416 Ridgway GJ, Sherburne SW, Lewis RD (1970) Polymorphism in the esterases of Atlantic herring. Transactions of the American Fisheries Society 99: 147–151 Riggins R, Pimentel RA, Walters DR (1977) Morphometrics of Lupinus nanus (Leguminosae) I. Variation in natural populations. Syst Bot 2: 317– 325 Shaw CR, Prasad R (1970) Starch gel eletrophoresis of enzymes-a compilation of recipes. Biochem Genet 4: 297–320 Silva-Pereira V (2007) Fluxo gênico e estrutura genética espacial em micro-escala em população de Chamaecrista blanchetii (Leguminosae) em campo rupestre na Chapada Diamantina, nordeste do Brasil. Ph.D. Thesis, Universidade Estadual de Feira de Santana, Feira de Santana Sneath PHA, Sokal RR (1973) Numerical taxonomy Freeman and Co, San Francisco Soltis DE, Haufler CH, Darrow DC, Gastony GJ (1983) Starch gel electrophoresis of ferns: a compilation of grinding buffers, gel and electrode buffers, and staining schedule. Amer Fern J 73: 9– 27 A. S. Conceição et al.: Biosystematics of Chamaecrista subsect. Baseophyllum (Leguminosae) Sotuyo S, Contreras JL, Delgado-Salinas A, Oyama K (2004) Genetic diversity and structure of the endemic Caesalpinia hintonii complex (Leguminosae: Caesalpinioideae) in Mexico. Pl Syst Evol 247: 131–143 StatSoft INS (2000) Statistica for windows (computer program manual) Salf Soft Inc, Tulsa Swofford DL, Selander RB (1989) BIOSYS-1: computer program for the analysis of allelic variation in population genetics and biochemical systematics Illinois Natural History Survey, Champaign 207 Thorpe JP (1982) The molecular clock hypothesis: biochemical evolution, genetic differentiation and systematics. Annual Rev Ecol Syst 13: 139–168 van der Bank H, van der Bank M, van Wyk B (2001) A review of the use of allozyme eletrophoresis in plant systematics. Biochem Syst & Ecol 29: 469– 483 Wright S (1978) Evolution and the genetics of populations: variability within and among natural populations. vol 4 University of Chicago Press, Chicago