TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae A new species, genus and tribe of Sapotaceae, endemic to Madagascar Laurent Gautier,1 Yamama Naciri,1 Arne A. Anderberg,2 Jenny E.E. Smedmark,3 Richard Randrianaivo4 & Ulf Swenson2 1 Conservatoire et Jardin botaniques de la Ville de Genève and Laboratoire de botanique systématique et biodiversité, University of Geneva, Case Postale 60, 1292 Chambésy/GE, Switzerland 2 Department of Botany, Swedish Museum of Natural History, P.O. Box 50007, 10405 Stockholm, Sweden 3 University Museum of Bergen, University of Bergen, P.O. Box 7800, 5020 Bergen, Norway 4 Missouri Botanical Garden, Madagascar Research and Conservation Program, B.P. 3391, Antananarivo 101, Madagascar Author for correspondence: Laurent Gautier, laurent.gautier@ville-ge.ch Abstract Phylogenetic relationships of the two Malagasy Sapotaceae endemic genera Capurodendron and Tsebona have been unclear until now. Recent collections from Madagascar, as well as a better representation of the tribe Isonandreae, altogether 95 terminals, were used to estimate a phylogeny of subfamily Sapotoideae. We analysed sequences of nrDNA (ITS) and cpDNA (trnH-psbA) with Bayesian inference and parsimony jackknifing. As in previous analyses, Sapoteae and Sideroxyleae are recovered monophyletic. In addition, Isonandreae, distributed in the Indo-Pacific, is for the first time resolved as monophyletic and sister to Sapoteae. All Malagasy accessions of Capurodendron, Tsebona, and a new species are grouped in another well-supported clade. This clade is accommodated in a new tribe Tseboneae characterized by caducous stipules, 5-merous flowers with quincuncial sepals, contorted aestivation of corolla lobes, absence of corolla appendages, one or three stamens opposite each corolla lobe, villous staminodes, seeds with an adaxial scar and plano-convex cotyledons, lacking endosperm. The new species is described in the new genus Bemangidia (B. lowryi) because it has a unique leaf venation for the tribe and combines different morphological features from Capurodendron and Tsebona. All three genera are well-supported monophyletic groups. Bemangidia lowryi is threatened with extinction due to extensive ongoing forest destruction and is assigned a preliminary conservation status of Critically Endangered. Keywords Bayesian analysis; Bemangidia; Capurodendron; Madagascar; Sapotaceae; Tsebona; Tseboneae Received: 25 Oct. 2012; revision received: 15 Apr. 2013; accepted: 27 Aug. 2013. DOI: http://dx.doi.org/10.12705/625.17 INTRODUCTION Madagascar is well-known for its high diversity of living organisms and especially for its unrivalled high level of endemism. Vascular plants are no exception to this pattern with a recently estimated richness of 10,889 native species in 1617 genera, with 82% endemism at the species level and 20% at the genus level (Callmander & al., 2011). Being only 400 km away from mainland Africa, it seems obvious that the closest relatives of Malagasy vascular plants are to be found in that continent. Accordingly, it is not surprising that Engler (1882) included Madagascar in his “African territory”. A recent estimate of similarity between Madagascar and Sub-Saharan African vascular floras (Gautier & al., 2012) revealed a very low value at the species level (0.029) but an eight-fold higher value at the genus level (0.246). However, as already pointed out by Engler (1882), Indo-Pacific elements also contribute to the flora of Madagascar, which have been reviewed by Schatz (1996). Like many other woody tropical families, Sapotaceae reach an unusually high score of species endemism in Madagascar. Although many taxa remain to be described, 81 of the 84 currently accepted species are endemic (96%; Gautier, 2003), which is close to the 98% endemism of the family in New Caledonia (Swenson, 2011). Depending on generic concepts, generic endemism is between 27% and 33%. Three genera are 972 endemic to Madagascar: Capurodendron Aubrév., Faucherea Lecomte, and Tsebona Capuron. The non-endemic genera are either pantropical (Manilkara Adans., Sideroxylon L.), palaeotropical (Donella Pierre ex Baill., Gambeya Pierre, Mimusops L.), or are shared with neighbouring western Indian Ocean islands (Labourdonnaisia Bojer, Labramia A. DC.). In the present state of our knowledge, there is no Sapotaceae genus shared between Madagascar and the Indo-Pacific area that does not occur elsewhere. Classification of Sapotaceae has long been a matter of dispute as exemplified by the fact that one year after Aubréville’s (1964) generic monograph of the family, which considered 122 genera in 15 tribes and four subfamilies, Baehni (1965) retained only 63 genera in six tribes and three subfamilies. The latest generic treatment by Pennington (1991) did not use subfamilies but recognized the five tribes Chrysophylleae, Isonandreae, Mimusopeae, Omphalocarpeae, and Sideroxyleae, with a total of only 53 genera, which can be interpreted as an extreme attempt to reconcile points of views through lumping. Analyses of molecular data dramatically improved our understanding of Sapotaceae phylogeny and have repeatedly demonstrated that many of Pennington’s (1991) larger genera are not monophyletic and tend to draw us back to Aubréville’s (1964) finer generic divisions. For example, Planchonella Pierre, Pleioluma (Baill.) Baehni, Sersalisia R. Br., and Van-royena Aubrév. are all Version of Record (identical to print version). TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae Pouteria Aubl. segregates that have been reinstated for groups in Australasia (Swenson & al., 2007, 2013). In Africa, Aningeria Aubrév. & Pellegr., Donella, Gambeya, and Malacantha Pierre are Chrysophylloideae satellite genera that need to be resurrected, but their circumscriptions are still unclear (Swenson & al., 2008b). Analyses of molecular data have also dramatically improved our understanding of Sapotaceae phylogeny at levels higher than the genus. Following the works of Anderberg & Swenson (2003), Swenson & Anderberg (2005), and Smedmark & al. (2006), it is now established that apart from the basal genera Sarcosperma Hook. f. and Eberhardtia Lecomte, all other extant Sapotaceae originated from two main evolutionary lineages to which the rank of subfamily is applied, viz. Chrysophylloideae and Sapotoideae. Unfortunately, these two clades do not conform to the classical (e.g., Aubréville, 1964) dichotomy based on calyx structure. Although Chrysophylloideae all have a single whorl of sepals, this character state is also present in part of Sapotoideae, the other members of the latter subfamily having two whorls of three or four calyx lobes. Recent phylogenies of Sapotoideae (Anderberg & Swenson, 2003; Swenson & Anderberg, 2005; Smedmark & al., 2006) have already revealed two monophyletic groups that are formally accepted at tribal rank: Sapoteae (including the Malagasy endemic Faucherea and its near-endemic relatives Labourdonnaisia and Labramia) and Sideroxyleae (with a few amendments compared with Pennington, 1991) whereas relationships among the other genera in the subfamily, including those formerly included in Isonandreae, remained uncertain. Specifically, regarding the remaining two endemic Malagasy genera, the position of Capurodendron was unclear, and the relationships of Tsebona have never been investigated. Capurodendron comprises 23 species of trees and shrubs in dry and wet lowland forest types (Aubréville, 1974), and is the third-largest endemic genus in Madagascar (Callmander & al., 2011). This genus is characterized by caducous stipules, 5-merous flowers with quincuncial sepals and contorted corolla lobes, connivent woolly staminodes that form a chamber above the ovary, and a seed scar that is adaxial or tends to be basi-ventral. It was placed by Aubréville (1964) in the Pouterieae, in the Chrysophylleae by Pennington (1991), but transferred to Sapotoideae with uncertain affinity by Swenson & Anderberg (2005). These latter authors hypothesized that it could be related to the Asian genera Burckella Pierre and Diploknema Pierre in tribe Isonandreae sensu Pennington (1991). Unfortunately, only one species of Capurodendron and a restricted number of representatives of tribe Isonandreae were included in previous phylogenetic studies, which prevented a better resolution of the whole group. Tsebona, with the so far only known species T. macrantha Capuron, occurs as a tall tree in the lowland evergreen forests of north-eastern Madagascar. It is characterized by the presence of stipules, a calyx of five sepals that are partly induplicate valvate, five contorted corolla lobes, 15 stamens organized in five bundles of three opposite each corolla lobe, and a seed scar covering 75% of the seed surface. Presence of multiple stamens per corolla lobe has led generic monographers of the family to include Tsebona in tribe Omphalocarpeae (Aubréville, 1964; Pennington, 1991). However, Pennington concluded that the flower morphology of Tsebona has nothing in common with other members of Omphalocarpeae and suggested that it is an isolated genus, possibly better placed in Mimusopeae or Isonandreae on account of floral and pollen structure. Swenson & Anderberg (2005) demonstrated the polyphyly of Omphalocarpeae and united it with Chrysophylleae (Chrysophylloideae). However, Tsebona was not included in their study and the placement of the genus remains unknown. During an expedition to SE Madagascar in 2006, a flowering specimen of Sapotaceae was collected in Bemangidy Forest, 55 km NNE of Fort Dauphin (Tolagnaro) (Fig. 1A–B). The plant matched an earlier sterile but unidentified collection from the same locality hosted in the Paris herbarium. Field images revealed a general appearance of Tsebona including thick terminal twigs, large leaves on long petioles, a 5-merous flower, and a particular type of quincuncial calyx structure similar to the one described by Capuron for Tsebona (1962, his figure 1). Dissection of the flower revealed further characters in common with Tsebona such as hairy staminodes and a gynoecium of five fused uniovulate carpels. However, a brochidodromous leaf venation with close, weak parallel secondaries (Fig. 1D; not eucamptodromous with prominent secondaries) and one stamen opposite each corolla lobe (not three—see Fig. 1E), are characters that distinguish it from T. macrantha. Furthermore, in the endeavour of gathering fresh material for molecular analyses, we recently collected a fruiting specimen in the type locality, (Fig. 1C, F) which displayed a seed scar (Fig. 1G) that covers less than 30% of the seed surface (not 75%). Flower structure of this species also shares several features with Capurodendron including connivent hairy staminodes that form a chamber above the ovary, but the calyx aestivation is different. Without doubt, this material represents a new endemic species of Sapotaceae for Madagascar, but its relationship to Capurodendron, Tsebona, or other genera is unclear. In this paper we explore phylogenetic relationships within subfamily Sapotoideae by using Bayesian inference and parsimony analyses based on nrDNA and cpDNA sequence data. There are four specific points we wish to address: (1) to explore the monophyly of Capurodendron, (2) to place Tsebona in a phylogenetic framework, (3) to test whether the novel species from Bemangidy Forest is closely related to Tsebona, and (4) to test the monophyly of tribe Isonandreae. MATERIALS AND METHODS Nomenclature. — The subfamily and partly the tribal classification follow that of Swenson & Anderberg (2005). In order to test the monophyly of Isonandreae, we refer to the classification by Pennington (1991). The checklist of Sapotaceae (Govaerts & al., 2001) includes published names and synonyms to species and genera. Taxon sampling. — A total of 95 terminals were selected for this study (Appendix 1), which include representatives of all genera of Isonandreae sensu Pennington (1991), a significant Version of Record (identical to print version). 973 TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae geographic sample of Sideroxyleae (Smedmark & Anderberg, 2007), as well as Sapoteae (Swenson & Anderberg, 2005). The general collecting programs of the Conservatoire et Jardin Botaniques de la Ville de Genève (CJB) in northern and western Madagascar enabled us to considerably improve the sampling of Capurodendron. Two dedicated expeditions conducted by CJB and the Swedish Museum of Natural History in 2010 and 2011 added silica gel-dried material of several species in the genus and allowed us to collect Tsebona macrantha as well as further specimens of the new species. Fresh specimens were dried in silica gel and complemented the few herbarium specimens from which we were able to extract DNA. In total, nineteen accessions of Capurodendron, two of Tsebona macrantha, and six of the new species were included in the study. Sequencable material of the type species of Capurodendron (C. rubrocostatum (Jum. & H. Perrier) Aubrév.) is unfortunately not yet available. Based upon inferences from our earlier studies, we rooted our tree with Neohemsleya T.D. Penn., a member of Chrysophylloideae (Smedmark & al., 2006; Swenson & al., 2008b). Molecular data. — Total genomic DNA was isolated using either the CTAB method of Doyle & Doyle (1987) or the NucleoSpin Plant II kit (Machery-Nagel, Düren, Germany). All precautions were taken to avoid contaminations, specifically for the herbarium specimens: the leaves were ground in Fig. 1. Bemangidia lowryi L. Gaut., a new species of a new genus from Madagascar. A, Bemangidy Forest, SE Madagascar; B, flowers of the type specimen; C, branch with fruit; D, secondary leaf venation (upper surface); E, cross-section of a pre-anthesis flower from the type collection; F, longitudinal section of a fruit; G, seed. — Photos: A, C, D, F, G by Ulf Swenson; B by Pete Lowry (MO); E by Laurent Gautier. 974 Version of Record (identical to print version). TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae a Qiagen Tissue Lyser using tungsten beads previously washed in a 0.4 M HCl solution, the beads, the bench, the pipets and the tubes were UV irradiated before use, and filter tips were systematically utilized for all steps before amplification. The specimens were genotyped for the chloroplast trnH-psbA spacer and for the nuclear ribosomal internal transcribed spacer ITS. For each locus, regular PCRs were used on silica gel–dried specimens and nested PCRs were used on herbarium specimens. The trnH-psbA spacer was amplified using the primers designed by Hamilton (1999) and by Naciri & Manen (2010) for regular and nested PCRs (Table 1). Double-stranded DNA was amplified by polymerase chain reaction (PCR) in a total volume of 20 µl. The PCR mix contained 2 µl of the 10× PCR Fast Start Taq DNA Polymerase buffer containing 20 mM MgCl2 (Roche, Rotkreuz, Switzerland), 0.2 µl of 5% bovine serum albumin acetylated solution (BSA), 0.4 µl 10 mM of each dNTPs (Roche), 1 µl of each primer (10 µM), 0.15 µl 5 U/µl of Fast Start Taq DNA Polymerase (Roche) and 1 µl of genomic DNA. PCR reactions included an initial denaturation step of 5 min at 95°C, followed by 35 cycles with 45 s denaturation at 96°C, 1 min annealing at 52°C for trnH-psbA or at 59°C for ITS, 1 min extension at 72°C, and a final extension of 7 min at 72°C. In some cases, and exclusively for herbarium specimens, the number of cycles was increased to 40. When nested PCRs were necessary, the amplifications were obtained using touch down programs with an initial denaturation step of 4 min at 95°C followed by 1 min denaturation, 30 s annealing and 1 min extension. For ITS, the annealing temperature was initially fixed at 64°C and decreased by 1°C at each cycle to 55°C for the first 10 cycles and was then stabilized at 55 °C for the remaining 25 cycles. For trnH-psbA, the annealing temperature was initially fixed at 64°C and decreased by 1°C at each cycle to 61°C for the first 5 cycles, was then stabilized at 60 °C for 10 cycles, decreased again for 5 cycles to 56°C and was finally stabilized at 56°C for the 12 last cycles. DNA sequencing was performed using the BigDye Terminator v.3.1 (Applied Biosystems, Delaware, U.S.A.). Both strands were sequenced separately in a 10 µl reaction mix with 1 µl of the purified PCR product, 2 µl of the amplification primers at 1 mM, 4 µl of purified water, 1 µl of the 5× reaction buffer and 2 µl of BigDye Terminator. Sequence reactions included 24 cycles of 10 s denaturation at 96°C, 5 s annealing at 50°C, 4 min extension at 60°C. The sequence reactions were run on an ABI PRISM 377 DNA automated sequencer (PE Biosystems, Waltham, Massachusetts, U.S.A.). The forward and reverse sequences were assembled using Sequencher v.4.8 software (Gene Codes Corporation, Ann Harbor, Michigan, U.S.A.) and were aligned using the BioEdit Sequence Editor v.7.0.9.0 (Hall, 1999). Bayesian analysis. — Evolutionary models for the molecular data were selected based on the Akaike information criterion (AIC; Akaike, 1973), which was calculated with MrAIC (Nylander, 2004). For the trnH-psbA dataset, the general time reversible substitution model (GTR; Tavaré, 1986) with gamma distributed rate variation among sites (+ Γ) was selected. For ITS, the GTR + Γ model, with a portion of sites being invariable, was selected. Coded indels were included as a separate partition, which was analysed as standard data using the Markov k model (Lewis, 2001) + Γ. Bayesian inference analysis (Rannala & Yang, 1996; Yang & Rannala, 1997) of the combined dataset, under a mixed model incorporating the submodels selected for each individual dataset, was performed in MrBayes v.3.2 (Huelsenbeck & Ronquist, 2001; Ronquist & Huelsenbeck, 2003). The Markov chain was run for 50 million generations, sampling every 5000 generations, and was assumed to have reached convergence when plots of the overall likelihood, as well as individual parameters of the model, were fluctuating around stable values. At this point, the average standard deviation of split frequencies for two analyses run in parallel was less than 0.01 and the potential scale reduction factor was 1.00 for all parameters. Phylogenetic hypotheses sampled during the first 1,000,000 generations were discarded as “burn-in” and the remaining trees from the two parallel analyses were used to construct a majority-rule consensus tree and to calculate posterior probabilities of clades. The analysis was repeated three times to make sure that independent runs converged on similar topologies containing the same nodes with posterior probabilities (PPs) greater or equal to 0.95. Posterior probabilities below 0.8 are not reported. Parsimony analysis. — Jackknife analysis (Farris & al., 1996), as implemented in PAUP* v.4.0 (Swofford, 2002), was performed on the dataset to retrieve parsimony support values. The settings were: 1000 jackknife replicates with a single random addition sequence, TBR branch swapping, collapsing branches of zero length, steepest descent not in effect, and saving a maximum of 1000 trees in each replicate. The excluded fraction of characters in each replicate was set to 37%. Parsimony jackknife values (JK) below 50% are not reported; group frequencies of 50%–74% are considered weak, 75%–89% as moderate, and ≥ 90 as strongly supported. Table 1. Primers used for trnH-psbA and ITS amplifications. Primer name 5′ → 3′ Reference Usage in nested PCR trnH psbA 5′ ACT GCC TTG ATC CAC TTG GC 3′ 5′ CGA AGC TCC ATC TAC AAA TGG 3′ Hamilton (1999) 2nd amplification trnH2 psbA2 5′ GCA TGG TGG ATT CAC AAT CC 3′ 5′ CCT AGC TGC TAT CGA AGC TC 3′ Naciri & Manen (2010) 1st amplification ITS-5P ITS-8P 5′ GGA AGG AGA AGT CGT AAC AAG G 3′ 5′ CAC GCT TCT CCA GAC TAC A 3′ Möller & Cronk (1997) 1st amplification ITS-4 5′ TCC TCC GCT TAT TGA TAT GC 3′ White & al. (1990) 2nd amplification with 5P Version of Record (identical to print version). 975 TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae RESULTS Bayesian analysis and parsimony jackknifing of the molecular dataset partitions (ITS, trnH-psbA, indels) recovered similar tree topologies of Sapotoideae, although the former is better resolved (Fig. 2). Some nodes receive posterior probabilities between 0.8 and 0.98, and these may lack jackknife support. The tribes Isonandreae, Sapoteae and Sideroxyleae, as well as another clade here referred to as the “New Tribe” are recovered with strong Bayesian support while jackknife recovers these clades with moderate to strong support. The relationships between the tribes are uncertain, which is mainly due to some taxa being embedded between the tribes, viz. the sister pair Lecomtedoxa (Pierre ex Engl.) Dubard and Neolemonniera Heine are recovered between Sideroxyleae and the rest of Sapotoideae, and Inhambanella (Engl.) Dubard is placed between Isonandreae and Sapoteae. Northia Hook. f., a monotypic genus confined to the Seychelles, has an isolated position with moderate Bayesian support (PP 0.91), outside of the New Tribe, sister to Isonandreae and Sapoteae. Capurodendron is recovered with maximum support and two subclades within the genus can be distinguished. Molecularly, the genus is easily distinguishable from all other taxa, especially in the trnH-psbA intron, because at position 154, a deletion of 17 base pairs is present in all members. All accessions referred to C. bakeri (Scott-Elliot) Aubrév. form a clade, and two specimens of uncertain determination (LG 5780, LG 5788) fall out together but without significant support. The two accessions of Tsebona macrantha from north-eastern Madagascar form a well-supported monophyletic group, and all six accessions from Bemangidy Forest in southeast Madagascar form another. Although the relationships of Capurodendron, Tsebona, and the new species from Bemangidy Forest are still unclear, they constitute three well-supported monophyletic groups. DISCUSSION The present phylogeny of Sapotoideae must be considered well-resolved and well-supported although only based on the nuclear marker ITS and the chloroplast marker trnH-psbA (Fig. 2). A previous phylogenetic study of the subfamily used five cpDNA markers with moderate success (Smedmark & al., 2006). Our study is congruent with recent phylogenies of the subfamily (Anderberg & Swenson, 2003; Swenson & Anderberg, 2005; Smedmark & al., 2006) by confirming that Sapoteae and Sideroxyleae are monophyletic, that relationships among tribes are not satisfactorily resolved, and that the position of the genera assigned to Glueminae sensu Pennington (1991), together with Northia, partly floats around in the phylogeny. Smedmark & Anderberg (2007) demonstrated that Sideroxylon oxyacanthum Baill. is not part of Sideroxylon and is better placed in Spiniluma (Baill.) Aubrév. (Aubréville, 1963), a genus occupying a basally branching position within subfamily Chrysophylloideae, and which is probably closely related to the Asian genus Xantolis Raf. We confirm this conclusion. 976 New in the present analysis is the discovery of a strictly endemic Malagasy clade that comprises three lineages, viz. Capurodendron, Tsebona, and the new species from Bemangidy Forest. This clade has been impossible to detect before because of insufficient sampling. Swenson & Anderberg (2005) used one species of Capurodendron and placed the genus in Sapotoideae, but with uncertain affinity. Smedmark & al. (2006) used two samples of Capurodendron that grouped with strong support but without strong indication of relationships. Isonandreae with seven genera as circumscribed by Pennington (1991) is for the first time recovered as monophyletic and receives strong posterior probability (PP 0.98; Fig. 2). The tribe is distinguished by a calyx that usually has two whorls of two or three sepals, undivided corolla lobes, and the absence of staminodes (Table 2). In contrast, its sister group Sapoteae usually has divided corolla lobes and staminodes present although exceptions exist. However, with the limited sample of 19 accessions in a group in which more than 250 species are currently recognized, it is obvious that future phylogenetic research is necessary to clarify natural limits of genera within Isonandreae. For instance, three accessions of Palaquium Blanco are found in a grade to the sister pair Aulandra longifolia H.J. Lam and Diploknema oligomera H.J. Lam, the latter genus being polyphyletic with the other species associated with Payena. Moreover, species of Madhuca Ham. ex J.F. Gmel. are found in two different well-supported subclades that are distantly related within the tribe. Also, the entire tribe is restricted to Asia with members of Burckella and Palaquium extending into the Pacific as far as Fiji (Smith, 1981). Needless to say, the rapid deforestation of Asia imposes tremendous exploitation pressure on the group and intense phylogenetic research as well as conservation assessments are urgently needed. Northia Hook. f., a monotypic genus confined to the Seychelles, has a calyx of two whorls with three sepals each, more or less entire corolla lobes, and vestigial staminodes (Table 2). Aubréville (1964) associated it with genera such as Manilkara and Mimusops that now are accommodated in Sapoteae (Swenson & Anderberg, 2005). Pennington (1991) placed Northia in his subtribe Manilkarinae, but stressed that it was the only member of the group lacking endosperm in its seeds. Our present analyses recover Northia as sister to the pair Isonandreae-Sapoteae with moderate posterior probability (PP 0.91) but without jackknife support. This position should be considered preliminary. Pennington (1991) classified Inhambanella and the two closely associated genera Lecomtedoxa and Neolemonniera in Glueminae, a subtribe of Mimusopeae with one member (Eberhardtia) now placed in Sarcospermatoideae (Smedmark & al., 2006). The nominal genus Gluema Aubrév. & Pellegr. is still unavailable for molecular analysis, but it is unlikely a member of the identified Malagasy clade since it lacks stipules, has divided corolla lobes, and staminodes are inserted inside and opposite the stamens. Gluema is furthermore restricted to West Africa. Inhambanella is presently recovered in between Isonandreae and Sapoteae, but this position is supported by very weak jackknife (JK 52; Fig. 2). Interestingly, the calyx of Inhambanella generally consists of a single whorl of five Version of Record (identical to print version). TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae Lecomtedoxa klaineana (new species) Fig. 2. Bayesian majority-rule consensus tree (shown as a phylogram) of Sapotoideae (Sapotaceae) based on DNA molecular sequence data (ITS, trnH-psbA) and gaps coded as binary characters. Node support is shown as posterior probabilities (above) and jackknife values (below) along the branches. Support values below PP < 0.8 and JK < 50% are not reported. Tribal classification is indicated to the right. Wavy lines indicate shortened molecular branches.Version of Record (identical to print version). 977 TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae sepals (as in Isonandreae), but some flowers display a biseriate calyx of 2 + 2 or 3 + 3 sepals (as in Sapoteae; Table 2). Gluema shares with both Lecomtedoxa and Neolemonniera the unique feature of having a dehiscing fruit. The two latter genera form a well-supported group (PP 0.99; JK 0.78) although its sister relationship is still uncertain, but it is not nested within the Malagasy clade, Isonandreae, or Sapoteae. Formal recognition of this clade should wait until Gluema becomes available for molecular analysis. Tseboneae and Bemangidia. — The pertinent question is if the Malagasy clade can be recognized as a new tribe. A close relationship between Capurodendron and Tsebona has long been obscured by the 15 stamens organized in five bundles of three opposite each corolla lobe in the latter, resulting in its artificial placement in Omphalocarpeae. The number of stamens opposite each corolla lobe in the family can be very homoplastic and differs between sister species as demonstrated for the New Caledonian genus Pycnandra (Swenson & al., 2008a). However, Capurodendron and Tsebona share many important features like the presence of generally caducous stipules, a 5-merous quincuncial calyx, contorted corolla lobes without appendages, villous staminodes, an adaxial seed scar and an embryo with plano-convex cotyledons lacking endosperm. The new species conforms perfectly to these morphological traits (Table 3). The molecular analysis conducted here provides further support to group all these three lineages in a separate monophyletic group, endemic to Madagascar. Despite the uncertain relationships of Northia and Inhambanella, as well as the sister relationship of Lecomtedoxa and Neolemonniera, the identified group is sister to Isonandreae and Sapoteae, two monophyletic tribes distinguished on character combinations (Table 2). We therefore suggest, in conjunction with the other two tribes, that the identified clade should likewise be recognized as a tribe, viz. Tseboneae. Considering the earlier history of splitting and lumping genera in Sapotaceae, it can be discussed if the identified group should be recognized as one, two, or three genera, all being acceptable solutions without violating the primary principle of monophyly (Backlund & Bremer, 1998). One possibility is to accommodate all lineages in a single genus with three clearly distinct subgenera. However, comparing the morphology of Capurodendron, Tsebona, and the new species, it is clear that a quite heterogeneous assemblage would be formed with species that differ regarding leaf venation, sepal aestivation, number of stamens opposite each corolla lobe, and percentage of seed surface occupied by the scar (Table 3). Moreover, at the molecular level, the deletion of 17 base pairs in the trnH-psbA chloroplast intron clearly distinguishes all Capurodendron species from other accessions in this clade, which supports the notion to maintain it as a separate genus. A second option could be to consider the new species as part of Tsebona, but this is a suboptimal solution since there is no support for such a relationship. Regarding morphology, these taxa are distinguished by leaf venation type, number of stamens, relative size of staminodes, Table 2. Character comparison between Isonandreae, Sapoteae, Tseboneae, and some genera with uncertain phylogenetic position in Sapotoideae. Isonandreae Tseboneae Inhambanella biseriate uniseriate uniseriate / uniseriate / uniseriate rarely biseriate rarely biseriate biseriate 5 / 2+2; 3+3 3+3; 4+4 5 5, rarely 2+2; 3+3 5, rarely 2+2; 3+3 5 3+3 polymerous iso- or polymerous isomerous isomerous isomerous isomerous isomerous usually divided with 2 lateral segments entire divided with 2 lateral segments divided with 2 lateral segments divided with 2 lateral segments divided with 2 lateral reduced or obsolete segments contortedimbricate contortedimbricate contorted not documented not documented not documented contorted Staminodes absent present present present present present vestigial Stamens iso- to polymerous (several whorls, not in bundles) isomerous iso- to polymerous (one whorl, in bundles) isomerous isomerous isomerous isomerous Fruit indehiscent indehiscent indehiscent indehiscent dehiscent dehiscent indehiscent scar adaxial adaxial to basi-ventral (rarely basal) adaxial adaxial adaxial adaxial adaxial endosperm present or absent present or absent absent absent thin layer thin layer absent Indo-Pacific Pantropical Madagascar Africa Africa Africa Seychelles series uniseriate / biseriate lobes lobes Calyx Corolla divsion of lobes entire aestivation Seed Distribution 978 Sapoteae Version of Record (identical to print version). Lecomtedoxa Neolemonniera Northia TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae Bemangidia L. Gaut., gen. nov. – Type: B. lowryi L. Gaut. Leaf venation brochidodromous with many, closely packed, parallel and weak secondaries, difficult to distinguish from the parallel tertiaries; calyx lobes 5, quincuncial, outer two with strongly involute margins that cause impressed longitudinal grooves on the external surface of the three inner sepals; corolla lobes 5, contorted, overlapping to the right; stamens 5; staminodes 5, villous, connivent and forming a chamber above the ovary; fruit 1-seeded with seed scar covering 30% of seed surface; embryo with plano-convex cotyledons; endosperm absent. Bemangidia differs from Tsebona by brochidodromous leaf venation (instead of eucamptodromous), five stamens (instead of 3 × 5), the staminodes that are connivent and form a chamber above the ovary, a single-seeded fruit (instead of 1–5 seeds), and a seed scar that occupies 30% of the seed surface (instead of 75%). Bemangidia differs from Capurodendron similarly by leaf venation, but also in sepal aestivation which causes an impressed median groove on the inner sepals, similar to the pattern in Tsebona but absent in Capurodendron, and by larger flowers, ca. 30 mm long compared to a 2–12 mm long corolla in Capurodendron. The genus is named after Bemangidy forest that is located approximately 55 km NNE of Tolañaro (Fort Dauphin) in south-western Madagascar, the only locality from where this genus is known. number of seeds per fruit, and seed scar surface (Table 3). We are therefore reluctant to include the new species in Tsebona. The remaining option is to keep Capurodendron and Tsebona as separate genera and to describe a new genus to accommodate the new species from Bemangidy Forest. The key character for this genus within the new tribe would be a brochidodromous leaf venation with weak, parallel, and close secondaries, hard to distinguish from the tertiaries, which is not present in any other member of Tseboneae. It further differs from Capurodendron by the way the external sepals impress a groove on the outer surface of the inner ones, and by the size of the corolla which is 2.5 times larger than the corolla of the largest Capurodendron flower. Compared with Tsebona, it also differs by the number of stamens, the staminodes that are connivent forming a chamber above the ovary, the number of seeds in the fruit and the percentage of the seed surface occupied by the scar. We conclude that the new species is better accommodated in a genus of its own. TAXONOMY Tseboneae L. Gaut. & Naciri, stat. nov. ≡ Tseboninae Baehni in Boissiera 11: 104. 1965 – Type: Tsebona Capuron. Trees with white latex, alternate leaves, and caducous stipules; calyx lobes 5, quincuncial; corolla lobes 5, entire, contorted, overlapping to the right; stamens 5, or 15 organized in bundles of 3, opposite each corolla lobe; staminodes 5, villous; seed scar adaxial; embryo with plano-convex cotyledons; endosperm absent. Three genera endemic to Madagascar: Bemangidia with one or possibly two species, Capurodendron with 23 known species, and the monotypic genus Tsebona. Bemangidia lowryi L. Gaut., sp. nov. – Type: MADAGASCAR. Toliara, Région de l’Anosy, Bemangidy Forest, ca. 3 km W of Antsato, along RN 12a, 65 km N of Ft. Dauphin. Humid evergreen forest on steep slope, E of Ivohibe Peak. 7 Feb. 2006, fl., P.P. Lowry II, J. Rabenantoandro, Table 3. Character comparison between Bemangidia, Capurodendron and Tsebona. Leaf venation Capurodendron Tsebona brochidodromous eucamptodromous series uniseriate uniseriate uniseriate lobes 5 5 5 aestivation quincuncial, externals with involute margins impressing a groove on the inner ones quincuncial quincuncial, externals with involute margins impressing a groove on the inner ones number of lobes 5 5 5 Calyx Corolla Bemangidia eucamptodromous lobes entire entire entire appendages absent absent absent aestivation contorted contorted contorted size ~30 mm long 2 to 12 mm long ~30 mm long present, connivent and forming a chamber above the ovary present, connivent and forming a chamber above the ovary present, free Staminodes Stamens 5 5 15, in 5 bundles of 3 1 1 1–5 Fruit number of seeds scar position adaxial adaxial adaxial Seed scar surface [%] 30 < 30 75 endosperm absent absent absent Version of Record (identical to print version). 979 TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae F. Randriatafika, E. Lowry, E. Ramisy & B. Mara 6657 (holotype: P! [P00568788]; isotypes: G! [G00340094], MO [6141396], TEF). — Figure 3. Tree up to 15 m tall, and 0.55 m in diameter, bark dark brown, oak-like fissured; inner bark tender, pink, with copious white latex, wood very hard, light yellow. Ultimate twigs 8–10 mm thick, with rusty pubescence on recent growth. Leaves alternate, grouped at the end of the twigs; stipules linear, 8 × 1.5 mm, rusty-pubescent, early caducous; petiole 60–100 × 1.5–2.0 mm, channelled, with a caducous rusty pubescence persisting inside the channel. Blade elliptical, 180–220 (to 300 mm on sterile shoots) × 60–65 mm, base cuneate, apex obtuse and sometimes shortly and broadly acuminate, chartaceous to coriaceous, with a caducous rusty pubescence on lower surface, glabrous on upper surface; leaf venation brochidodromous with very dense and parallel secondaries, almost indistinguishable from Fig. 3. Line drawing of the type collection of Bemangidia lowryi L. Gaut. (Lowry & al. 6657; P). A, flowering branch; B, flower with corolla and 3 calyx lobes removed, showing gynoecium; C, flower with corolla removed; D, two cross-sections of ovary; E, staminode; F, corolla opened; G, fruit; H, seed. — A–F drawn from P.P. Lowry II & al. 6657; G–H drawn from L. Gautier & al. 5789. — Drawing: C. Chatelain. 980 Version of Record (identical to print version). TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae tertiaries, diverging from the primary vein at an angle of 70° to 90°, looping very near the margin. Primary vein raised on lower surface, impressed above, in continuation of the channelled petiole. Flowers two or three in axillary fascicles; pedicel 50 × 2 mm, rusty-pubescent. Calyx quincuncial of 5 sepals united at the base; sepals broadly lanceolate, 20–2 × 10–12 mm, rusty-pubescent outside; the two outer sepals also pubescent on the inner side, except for the middle third, both margins involute and causing longitudinal grooves on the overlapped sepals; the intermediate sepal with only one involute margin, and a third of the inside pubescent; the two most central sepals glabrous inside. Corolla pale greenish or cream; tube 7–8 mm long, with 5 broadly lanceolate lobes 18–20 × 10 mm, contorted in bud, overlapping to the right. Stamens 5, opposite the corolla lobes; filaments glabrous, attached for half of their length to corolla tube, free part 7–8 mm long; anthers 10 mm long, attached to the filaments at 1/3 from base, sagittate; thecae ciliate; connective extending above thecae in an acute 1 mm long appendage. Staminodes 5, alternipetalous, broadly lanceolate, 8 × 5 mm, margin ciliate, connivent and forming a chamber over the ovary. Gynoecium superior, hirsute, 6.5 mm in diameter and 3.3 mm in height, composed of 5 fused carpels with one ovule per locule; style 27 mm long, with 5 weak longitudinal grooves; stigmatic area round, 0.3 mm in diameter. Fruit a berry, obconic and asymmetric, 40–50 × 30–35 mm with copious white latex, borne on 40–50 mm long, 3 mm thick pedicel, which gradually widens to 6 mm near the fruit; calyx partly persistent; seed 1, ellipsoid, slightly laterally compressed, ca. 30 × 22 × 19 mm; testa shiny, light brown; seed scar adaxial, covering 65% of the length and 40% of the circumference; embryo with plano-convex cotyledons; endosperm absent. Etymology. – It is the pleasure for the author of the species to dedicate it to the first collector of the type specimen, Porter Prescott (Pete) Lowry II from Missouri Botanical Garden, as an acknowledgment for all he has done for the botany of Madagascar in an unremitting spirit of collaboration. Conservation assessment. – With an estimated Extent Of Occurrence (EOO) of less than 10 km², an Area Of Occupancy (AOO) of 9 km² and a single subpopulation, occurring within a provisional protected area (Tsitongambarika), Bemangidia lowryi is assigned a preliminary status of “Critically Endangered” (CR B1ab(iii); B2ab(iii)) following IUCN Red List Categories and Criteria (IUCN, 2001). Other specimens. – MADAGASCAR. Toliara, Région de l’Anosy, Bemangidy Forest, no date, sterile, Service forestier, s.n. (P, 2 sheets [P00568786, P00568787]); Ibid., alt. 440 m, forêt dense humide de crête, 10 Dec. 2011, sterile, Gautier, Swenson & Randrianaivo 5784 (G, S, TAN); Ibid., alt. 280 m, forêt dense humide de crête, 11 Dec. 2011, sterile, Gautier, Swenson & Randrianaivo 5786 (G, S, TAN); Ibid., alt. 430 m, forêt dense humide de crête, 11 Dec. 2011, fr., Gautier, Swenson & Randrianaivo 5789 (G, K, MO, P, S, TAN). Note. – One specimen (Gautier 5790) collected with flower buds in the same forest, but at higher altitude (670 m) on an exposed ridge on the eastern summit of Ivohibe Peak, is clearly a member of the same genus. The tree was smaller in size (7 m) and had smaller leaves. It was first thought to be merely an individual adapted to a more exposed site, but molecular analysis indicates that it is also genetically different from the type population. Until more fertile material is collected from this population we refrain from including it in the new species. ACKNOWLEDGMENTS The authors would like to express their thanks to three anonymous reviewers for useful comments on the manuscript, Régine Niba for the lab work, Cyrille Chatelain for the line drawing, Martin Callmander for the conservation assessment calculations, and to Patrick Perret for nomenclature advice. The Swedish Research Council supported fieldwork by a grant to Ulf Swenson, and Arne Anderberg through a grant for Angiosperm phylogeny. LITERATURE CITED Akaike, H. 1974. A new look at the statistical model identification. I. E. E. E. Trans. Automatic Control 19: 716–723. http://dx.doi.org/10.1109/TAC.1974.1100705 Anderberg, A.A. & Swenson, U. 2003. Evolutionary lineages in Sapotaceae (Ericales): A cladistic analysis based on ndhF sequence data. Int. J. Pl. Sci. 164: 763–773. http://dx.doi.org/10.1086/376818 Aubréville, A. 1963. Notes sur des Sapotacées. Adansonia 3: 19–42. Aubréville, A. 1964. Les Sapotacées: Taxonomie et phytogéographie. Adansonia Mém. 1: 1–157. Aubréville, A. 1974. Sapotaceae. Pp. 1–128 in Humbert, H. (ed.), Flore de Madagascar et des Comores, vol. 164. Paris: Museum National d’Histoire Naturelle. Backlund, A. & Bremer, K. 1998. To be or not to be—Principles of classification and monotypic plant families. Taxon 47: 391–400. http://dx.doi.org/10.2307/1223768 Baehni, C. 1965. Mémoires sur les Sapotacées. III. Inventaire des genres. Boissiera 11: 1–262. Callmander, M., Phillipson, P.B., Schatz, G.E., Andriambololonera, S., Rabarimanarivo, M., Rakotonirina, N., Raharimampionona, J., Chatelain, C., Gautier, L. & Lowry, P.P., II 2011. The endemic and non-endemic vascular flora of Madagascar updated. Pl. Ecol. Evol. 144: 121–125. http://dx.doi.org/10.5091/plecevo.2011.513 Capuron, R. 1962. Tsebona, genre nouveau de Sapotacées de Madagascar. Adansonia, sér. 2, 2: 122–128. Doyle, J.J. & Doyle, J.L. 1987. A rapid DNA isolation procedure for small quantities of fresh leave tissue. Phytochem. Bull. Bot. Soc. Amer. 19: 11–15. Engler, A. 1882. Versuch einer Entwicklungsgeschichte der extratropischen Florengebiete der südlichen Hemisphäre und der tropischen Gebiete. Leipzig: Engelmann. Farris, J.S., Albert, V.A., Källersjö, M., Lipscomb, D. & Kluge, A.G. 1996. Parsimony jackknifing outperforms neighbor-joining. Cladistics 11: 99–124. http://dx.doi.org/10.1111/j.1096-0031.1996.tb00196.x Gautier, L. 2003. Sapotaceae. Pp. 342–346 in: Goodman, S.M. & Benstead, J.P. (eds.), The natural history of Madagascar. Chicago: The University of Chicago Press. Gautier, L., Chatelain, C., Callmander, M.W. & Phillipson, P.B. 2012. Richness, similarity and specificity of Madagascar flora compared with Sub-Saharan Africa. Pl. Ecol. Evol. 145: 55–64. http://dx.doi.org/10.5091/plecevo.2012.591 Govaerts, R., Frodin, D.G. & Pennington, T.D. 2001. World checklist and bibliography of Sapotaceae. Kew: Royal Botanic Gardens. Hall, T. 1999. BioEdit: A user-friendly biological sequence alignment Version of Record (identical to print version). 981 TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae editor and analysis program for Windows 95/98/NT. Nucl. Acids Symp. Ser. 41: 95–98. Hamilton, M.B. 1999. Four primer pairs for the amplification of chloroplastic intragenic regions with intraspecific variation. Molec. Ecol. 8: 521–523. Huelsenbeck, J.P. & Ronquist, F. 2001. MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755. http://dx.doi.org/10.1093/bioinformatics/17.8.754 IUCN 2001. IUCN Red List categories and criteria, version 3.1. Gland & Cambridge: IUCN Species Survival Commission. http://www .iucnredlist.org/static/categories_criteria_3_1 Lewis, P.O. 2001. A likelihood approach to estimating phylogeny from discrete morphological character data. Syst. Biol. 50: 913–925. http://dx.doi.org/10.1080/106351501753462876 Möller, M. & Cronk, Q.C.B. 1997. Origin and relationships of Saintpaulia (Gesneriaceae) based on ribosomal DNA internal transcribed spacer (ITS) sequences. Amer. J. Bot. 84: 956–965. http://dx.doi.org/10.2307/2446286 Naciri Y. & Manen J.-F. 2010. Potential DNA transfer from the chloroplast to the nucleus in Eryngium alpinum. Molec. Ecol. Resources 10: 728–731. http://dx.doi.org/10.1111/j.1755-0998.2009.02816.x Nylander, J.A.A. 2004. MrAIC.pl, version 1.4. http://www.abc.se /~nylander/. Pennington, T.D. 1991. The genera of Sapotaceae. Kew: Royal Botanic Gardens. Rannala, B. & Yang, Z. 1996. Probability distribution of molecular evolutionary trees: A new method of phylogenetic inference. J. Molec. Evol. 43: 304–311. http://dx.doi.org/10.1007/BF02338839 Ronquist, F. & Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572–1574. http://dx.doi.org/10.1093/bioinformatics/btg180 Schatz, G.E. 1996. Malagasy/Indo-australo-malesian phytogeographic connections. Pp. 73–84 in: Lourenço, W.R. (ed.), Biogéographie de Madagascar. Paris: ORSTOM. Smedmark, J.E.E. & Anderberg, A.A. 2007. Boreotropical migration explains hybridization between geographically distant lineages in the pantropical clade Sideroxyleae (Sapotaceae). Amer. J. Bot. 94: 1491–1505. http://dx.doi.org/10.3732/ajb.94.9.1491 Smedmark, J.E.E., Swenson, U. & Anderberg, A.A. 2006. Accounting for variation of substitution rates through time in Bayesian phylogeny reconstruction of Sapotoideae (Sapotaceae). Molec. Phylgen. Evol. 39: 706–721. http://dx.doi.org/10.1016/j.ympev.2006.01.018 Smith, A.C. 1981. Flora vitiensis nova, vol. 2. Hawaii: Pacific Tropical Botanical Garden. Swenson, U. 2011. Det totala antalet nybeskrivna växtarter – en svensk flora om året. Fauna & Flora 106: 18–26. Swenson, U. & Anderberg, A.A. 2005. Phylogeny, character evolution, and classification of Sapotaceae (Ericales). Cladistics 21: 101–130. http://dx.doi.org/10.1111/j.1096-0031.2005.00056.x Swenson, U., Bartish, I.V. & Munzinger, J. 2007. Phylogeny, diagnostic characters and generic limitation of Australasian Chrysophylloideae (Sapotaceae, Ericales): Evidence from ITS sequence data and morphology. Cladistics 23: 201–228. http://dx.doi.org/10.1111/j.1096-0031.2006.00141.x Swenson, U., Lowry, P.P., Munzinger, J., Rydin, C. & Bartish, I.V. 2008a. Phylogeny and generic limits in the Niemeyera complex of New Caledonian Sapotaceae: Evidence of multiple origins of the anisomerous flower. Molec. Phylogen. Evol. 49: 909–929. http://dx.doi.org/10.1016/j.ympev.2008.09.022 Swenson, U., Richardson, J.E. & Bartish, I.V. 2008b. Multi-gene phylogeny of the pantropical subfamily Chrysophylloideae (Sapotaceae): Evidence of generic polyphyly and extensive morphological homoplasy. Cladistics 24: 1006–1031. http://dx.doi.org/10.1111/j.1096-0031.2008.00235.x Swenson, U., Nylinder, S. & Munzinger, J. 2013. Towards a natural classification of Sapotaceae subfamily Chrysophylloideae in Oceania and Southeast Asia based on nuclear sequence data. Taxon 62: 746–770. http://dx.doi.org/10.12705/624.11 Swofford, D.L. 2002. PAUP*: Phylogenetic analysis using parsimony (*and other methods), version 4. Sunderland, Massachusetts: Sinauer. Tavaré, S. 1986. Some probabilistic and statistical problems in the analysis of DNA sequences. Pp. 57–86 in: Miura, R.M. (ed.), Some mathematical questions in biology—DNA sequence analysis. Providence, Rhode Island: American Mathematical Society. White, T.J., Bruns, T., Lee, S. & Taylor, J.W. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. Pp. 315–322 in: Innis, M.A., Gelfand, D.H., Sninsky, J.J. & White, T.J. (eds.), PCR protocols: A guide to methods and amplifications. San Diego: Academic Press. Yang, Z. & Rannala, B. 1997. Bayesian phylogenetic inference using DNA sequences: A Markov chain Monte Carlo method. Molec. Biol. Evol. 14: 717–724. http://dx.doi.org/10.1093/oxfordjournals.molbev.a025811 Appendix 1. Taxa, voucher information, and GenBank accession numbers (ITS, trnH-psbA) for specimens used in this phylogenetic study of Sapotoideae (Sapotaceae). The African species Neohemsleya usambarensis was used as outgroup. Aulandra longifolia H.J. Lam, Borneo, Sarawak, Christensen 1720 (K), HF542797, HF542871. Autranella congolensis (De Wild.) A. Chev., Congo, Bokdam 4401 (WAG), HF542798, AM179710. Burckella macropoda (K. Krause) H.J. Lam, Indonesia (cultivated), Chase 1359 (K), HF542799, AM179713. Capurodendron androyense Aubrév., Madagascar, Humbert 28855 (B), AM408107, AM179714. Capurodendron bakeri (Scott-Elliot) Aubrév., Madagascar, Gautier, Randrianaivo & Bernard 5491 (G), HF542800, HF542872; Madagascar, Gautier, Randrianaivo & Aridy 5553 (G), HF542801, HF542873; Madagascar, Gautier, Randrianaivo & Aridy 5554 (G), HF542802, HF542874. Capurodendron cf. bakeri (Scott-Elliot) Aubrév., Madagascar, Gautier, Swenson & Randrianaivo 5780 (G), HF542803, HF542875; Madagascar, Gautier, Swenson & Randrianaivo 5788 (G), HF542804, HF542876. Capurodendron delphinense Aubrév., Madagascar, Gautier & al. 5801 (G), HF542805, HF542877. Capurodendron gracilifolium Aubrév., Madagascar, Gautier, Tahinarivony & Bolliger 5736 (G), HF542806, –. Capurodendron greveanum Aubrév., Madagascar, Gautier & Ranirison 5395 (G), HF542807, HF542878. Capurodendron mandrarense Aubrév., Madagascar, Gautier, Swenson & Randrianaivo 5793 (G), HF542808, HF542879. Capurodendron microphyllum (Scott-Elliot) Aubrév., Madagascar, Gautier, Swenson & Randrianaivo 5794 (G), HF542809, HF542880. Capurodendron nodosum Aubrév., Madagascar, Ranirison & Nusbaumer 454 (G), HF542810, HF542881. Capurodendron pseudoterminalia Aubrév., Madagascar, Gautier, Randrianaivo & Aridy 5544 (G), HF542811, HF542882. Capurodendron sakalavum Aubrév., Madagascar, Gautier, Nusbaumer & Ranirison 4670 (G), HF542812, HF542883; Madagascar, Gautier, Tahinarivony & Bolliger 5669 (G), HF542815, HF542885; Madagascar, Gautier, Tahinarivony & Bolliger 5570 (G), HF542813, HF542884; Madagascar, Gautier, Tahinarivony & Bolliger 5675 (G), HF542814, HF542886. Capurodendron aff. sakalavum Aubrév., Madagascar, Gautier, Tahinarivony & Bolliger 5681 (G), HF542816, HF542887. Capurodendron sp. LG 5762, Madagascar, Gautier, Swenson & Randrianaivo 5762 (G), HF542817, HF542888. Diploknema butyracea (Roxb.) H.J. Lam, Nepal, Polunin, Sykes & Williams 3975 (UPS), HF542818, HF542889. Diploknema oligomera H.J. Lam, Indonesia (cultivated), Chase 1360 (K), HF542819, AM179717. Faucherea parvifolia Lecomte, Madagascar, Birkinshaw & al. 357 (P), HF542820, AM179719; Madagascar, Gautier, Swenson & Randrianaivo 5756 (G), HF542821, HF542890. Faucherea sp. LG 5507, Madagascar, Gautier & al. 5507 (G), HF542822, HF542891. Inhambanella henriquezii (Engl. & Warb.) Dubard, South Africa, de Winter & J. Vahrmeijer 8536 (S), HF542823, AM179720. Isonandra compta Dubard, Sri Lanka, Emanuelsson 3039 (S), HF542824, HF542892. Isonandra perakensis King & Gamble, Malaysia, Pennington & Wong 10227 (K), HF542825, HF542893. Isonandra “Philcox10227”, Sri Lanka, Philcox, Weerasooriya & Weerasekera 10227 (K), HF542826, HF542894. Labourdonnaisia calophylloides Bojer, Réunion, Capuron 28241 (K), HF542827, 982 Version of Record (identical to print version). TAXON 62 (5) • October 2013: 972–983 Gautier & al. • New species, genus and tribe of Sapotaceae Appendix 1. Continued. HF542895. Labourdonnaisia revoluta Bojer, Mauritius, Bernardi 14717 (P), HF542828, HF542896. Labramia costata (Hartog ex Baill.) Aubrév., Madagascar, Randriamanalinarivo 577 (UPS), HF542829, AM179721. Labramia mayottensis Labat, Pignal & Pascal, Mayotte Island, Labat & al. 3309 (P), HF542830, AM179722. Lecomtedoxa klaineana (Pierre ex Engl.) Pierre ex Dubard, Holland (cultivated), Veldhuizen 1509 (WAG), AM408109, AM179723. Madhuca crassipes (Pierre ex Becc.) H.J. Lam, Borneo, Jugah ak. Kudi 23757 (K), HF542831, HF542897. Madhuca hainanensis Chun & F.C. How, Hainan (cultivated, South China Bot. Gard.), Hao 530 (S), HF542832, AM179725. Madhuca longifolia (J. König ex L.) J.F. Macbr., Hainan (cultivated, South China Bot. Gard.), Hao 531 (S), HF542833, AM179726. Madhuca microphylla (Hook.) Alston, Sri Lanka, Fagerlind 4790 (S), HF542834, AM179727. Madhuca motleyana (de Vriese) J.F. Macbr., Malaysia, Pennington & Kocummen 10259 (K), HF542835, HF542898. Madhuca palembanica (Miq.) Forman, Indonesia, Triono, Saman & Victobery 11 (K), HF542836, HF542899. Madhuca utilis (Ridl.) H.J. Lam, Malaysia, Pennington & Asri 10209 (K), HF542837, HF542900. Manilkara bidentata (A. DC.) A. Chev., South America, NL 110492 and P01860330 (CAY), FJ037872, FJ039059. Manilkara butugi Chiov., Ethiopia, Friis & al. 4097 (B), HF542838, AM179728. Manilkara chicle (Pittier) Gilly, Guatemala, Castillo & al. 2083 (B), HF542839, AM179729. Manilkara concolor (Harv.) Gerstner, South Africa, Swenson & Karis 635 (S), HF542840, AM179730. Manilkara discolor (Sond.) J.H. Hemsl., South Africa, Swenson & Karis 632 (S), HF542841, AM179731. Manilkara hexandra (Roxb.) Dubard, Thailand, Chantaranothai 2340 (KKU), HF542842, AM179732. Manilkara kauki (L.) Dubard, Thailand, Chantaranothai 2341 (KKU), HF542843, AM179734. Manilkara mayarensis (Ekman ex Urb.) Cronquist, Cuba, Ekman 9971 (LD), HF542844, AM179735. Manilkara obovata (Sabine & G. Don) J.H. Hemsl., Burkina Faso, Küppers 937 (FR), HF542845, AM179737. Manilkara zapota (L.) P. Royen, Thailand (cultivated), Chantaranothai 2378 (KKU), HF542846, AM179738. Mimusops caffra E. Mey. ex A. DC., South Africa, Swenson & Karis 636 (S), HF542847, AM179739. Mimusops comorensis Engl., Comoro Islands, Pignal & Ginguette 1065 (P), HF542848, AM179740. Mimusops elengi L., Thailand, Chantaranothai 2305 (KKU), HF542849, AM179741. Mimusops obovata Sond., South Africa, Swenson & Karis 633 (S), HF542850, AM179742. Mimusops zeyheri Sond., South Africa, Dahlstrand 6386 (GB), HF542851, AM179743. Neohemsleya usambarensis T.D. Penn., Tanzania, Borhidi & al. 84905 (UPS), DQ246684, DQ344118. Neolemonniera clitandrifolia (A. Chev.) Heine, Ghana, Jongkind, Schmidt & Abbiw 1777 (MO), HF542852, AM179745. Northia seychellana Hook. f., Seychelles, Chong-Seng s.n. (S), HF542853, AM179747. Palaquium amboinense Burck, Sri Lanka, Iuijesundara s.n. (K), HF542854, HF542901. Palaquium formosanum Hayata, Taiwan, Chung & Anderberg 1421 (HAST), AM408110, AM179748. Palaquium microphyllum King & Gamble, Malaysia, Pennington, Kochummen & Wong 10222 (K), HF542855, HF542902. Payena acuminata (Blume) Pierre, Indonesia (cultivated), Chase 1368 (K), HF542856, AM179749. Payena lucida A. DC., Borneo, Ambri & al. AA1604 (L), HF542857, AM179750. Sideroxylon americanum (Mill.) T.D. Penn., Bahamas, Gillis 11576 (B), AM408060, AM179751. Sideroxylon betsimisarakum Lecomte, Madagascar, Schönenberger, Balthazar & Smedmark A-102 (UPS), AM408063, AM179752. Sideroxylon capiri (A. DC.) Pittier, Mexico, García 1848 (BM), AM408065, AM407735. Sideroxylon capuronii Aubrév., Madagscar, Capuron 20151-SF (P), AM408066, AM407736. Sideroxylon foetidissimum Jacq., Hispaniola, Lundin 638 (S), AM408071, AM407740. Sideroxylon grandiflorum A. DC., Mauritius, Friedman & al. 2653 (P), AM408075, AM407741. Sideroxylon horridum (Griseb.) T.D. Penn., Cuba, Gutiérrez & Nilsson 5 (S), AM408076, AM179755. Sideroxylon inerme L., South Africa (cultivated in Denmark), Nielsen s.n. (S), AM408078, AM179756. Sideroxylon lycioides L., U.S.A., Radford & al. 11453 (B), AM408081, AM179758. Sideroxylon majus (C.F. Gaertn.) Baehni, Réunion, Capuron 28185 (B), AM408082, AM179759. Sideroxylon mirmulans R. Br., Canary Islands, Gomera, Swenson & Fernandez 581 (S), AM408084, AM179761. Sideroxylon polynesicum (Hillebr.) Smedmark & Anderb., Hawaii, Degener 20770 (S), AM408059, AM179746. Sideroxylon portoricense Urb., Jamaica, Mathew 1 (BM), AM408093, AM407749. Sideroxylon wightianum Hook. & Arn., South China Botanical Garden (cultivated), Hao 532 (S), AM408106, AM179770. Spiniluma oxyacantha (Baill.) Aubrév., Yemen, Wood Y/75/388 (BM), AM408089, AM407745. Tieghemella heckelii (A. Chev.) Pierre ex Dubard, Ghana, Jongkind 3936 (WAG), HF542858, AM179771. Sapotaceae LG 5784, Madagascar, Bemangidy, Gautier, Swenson & Randrianaivo 5784 (G), HF542859, HF542903. Sapotaceae LG 5786, Madagascar, Bemangidy, Gautier, Swenson & Randrianaivo 5786 (G), HF542860, HF542904. Sapotaceae LG 5789, Madagascar, Bemangidy, Gautier, Swenson & Randrianaivo 5789 (G), HF542861, HF542905; Sapotaceae LG 5790, Madagascar, Bemangidy, Gautier 5790 (G), HF542862, HF542906. Sapotaceae PPL 6657, Madagascar, Bemangidy, Lowry II & al. 6657 (P), HF542863, HF542907. Sapotaceae SF s.n., Madagascar, Bemangidy, Service Forestier s.n. (P), HF542864, HF542908. Tsebona macrantha Capuron, Madagascar, Masoala, Gautier 5509 (G), HF542866, HF542909; Madagascar, Masoala, Randrianaivo 116 (G), HF542865, HF542916. Vitellaria paradoxa C.F. Gaertn., Benin, Neumann 1512 (FR), HF542867, AM179776. Vitellariopsis cuneata (Engl.) Aubrév., Tanzania, Thomas 3662 (WAG), HF542868, AM179772. Vitellariopsis kirkii (Baker) Dubard, Kenya, Robertson 4085 (WAG), HF542869, AM179774. Vitellariopsis marginata (N.E. Br.) Aubrév., South Africa, Chase 1122 (K), HF542870, AM179775. Xantolis siamensis (Fletcher) P. Royen, Thailand, Smitairi 1 (L), AY552154, DQ344151. Version of Record (identical to print version). 983