Botanical Journal of the Linnean Society, 2018, 188, 355–376. With 2 figures. Phylogenetic study of Plectranthus, Coleus and allies (Lamiaceae): taxonomy, distribution and medicinal use ALAN PATON1* , MONTFORT MWANYAMBO2,3 and ALASTAIR CULHAM2 Science Directorate, Royal Botanic Gardens Kew, TW9 3AB, UK School of Biological Sciences, University of Reading, Whiteknights, Reading, RG6 6AS, UK 3 National Herbarium and Botanic Gardens of Malawi, P. O. Box 528 Zomba, Malawi 2 Received 20 April 2018; revised 12 July 2018; accepted for publication 13 August 2018 Lamiaceae subtribe Plectranthinae, a palaeotropical group of just over 450 species with mainly zygomorphic flowers and stamens that are contiguous at the point of insertion at the base of the lower corolla lip, include the medicinally and horticulturally important genus Plectranthus. Plectranthus currently includes the formerly recognized Coleus and Solenostemon. A phylogenetic analysis of the group is presented based on rps16, trnL-F and trnS-G regions of the plastid genome. Plectranthus as currently recognized is paraphyletic; a clade containing the type of Coleus and including Solenostemon, Pycnostachys and Anisochilus is sister to the rest of the group. Three endemic and monotypic Madagascan genera, Dauphinea, Madlabium, Perrierastrum and the Madagascan Capitanopsis belong to a single clade and are recognized under Capitanopsis; the new combinations are made here. Plectranthus s.s. is sister to a clade comprising Thorncroftia and Tetradenia. Tetradenia, unlike any other members of Plectranthinae, has actinomorphic corollas and is usually dioecious. A group of other species previously recognized as Plectranthus form a clade separate from Plectranthus s.s. and is recognized as Equilabium gen. nov. Estimates of clade age suggest that the genera begin to diversify from the mid to late Miocene. Plectranthinae are found in dry woodlands, montane grasslands and evergreen forest margins. Shifts between habitats occur in most clades, although significantly fewer than if the changes were random. The distribution of the clades in the major habitats is examined. Migration in Plectranthinae was from Africa to Madagascar and Asia, and there is no evidence of migration back to Africa. The phylogenetic pattern of medicinal use in Plectranthinae is weak, and issues surrounding this are discussed. ADDITIONAL KEYWORDS: Africa – Anisochilus – Capitanopsis – endemic – Equilabium – habitat shifts – Madagascar – medicinal uses – Pycnostachys – Solenostemon – Tetradenia – Thorncroftia. INTRODUCTION Subtribe Plectranthinae (tribe Ocimeae, Lamiaceae) are a palaeotropical group of 11 genera and just over 450 species (Harley et al., 2004) and are the largest of seven subtribes that belong to the pantropical tribe Ocimeae (Zhong et al., 2010; Pastore et al., 2011). The largest genus, Plectranthus L’Hér., is a widely used medicinal and horticultural genus including > 320 species (Lukhoba, Simmonds, Paton, 2006; Rice et al., 2011). Plectranthus incorporates the currently synonymized genera Coleus Lour. and Solenostemon Thonn. (Harley et al. 2004), names that are frequently still used in medicine and horticulture (e.g. Vanaja & *Corresponding author. E-mail: a.paton@kew.org Annadurai, 2013; Shepherd & Maybry, 2016; Cubey, 2017). Ocimeae and Plectranthinae have both been shown to be monophyletic based on a molecular phylogenetics using the trnL intron, trnL-trnF intergenic spacer and rps16 intron of plastid DNA (Paton et al., 2004). Paton et al. (2004) found Plectranthinae to be sister to subtribe Ociminae including Ocimum L., but the relationships to the other subtribes were unresolved. Although Plectranthinae cannot be diagnosed unambiguously by a single unifying morphological character, all members apart from Tetradenia Benth. have all their stamens inserted at the base of the lower (anterior) lip of the corolla, a character that is not found elsewhere in Ocimeae (Paton et al., 2004). © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 355 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 1 356 A. PATON ET AL. of this paper are to provide an updated molecular phylogenetic analysis of Plectranthinae, to identify major clades, to examine congruence with existing generic limits, to use the phylogenetic analysis to examine geographical and ecological distribution of the clades and to build upon previous studies to further explore the distribution of medicinal use of the group (Lukhoba et al., 2006) to facilitate further targeted medicinal and economic research. As a result of the analyses presented here, several changes to current generic delimitation are proposed and the nomenclatural changes necessary to recognize these generic level changes are made. However, the bulk of name changes affecting the names of species currently placed in Plectranthus s.l. will be published elsewhere, providing a conspectus of all names published in Plectranthus and Coleus. MATERIAL AND METHODS SAMPLING A survey of herbarium collections was undertaken to guide sampling of geographical and morphological diversity in Plectranthinae and clades in Plectranthus recognized by Paton et al. (2004, 2009, 2013) and Lukhoba et al. (2006). Recent floristic accounts and the World Checklist of Selected Plant Families were also consulted (Codd, 1985; Forster, 1992, 1994, 2011; Hedge et al., 1998; Suddee et al., 2005; Paton et al., 2009, 2013; Govaerts et al., 2016 and references therein). All genera of Plectranthinae recognized by Harley et al. (2004) were included in the analysis except Madlabium Hedge as DNA extractions from available herbarium material of this genus failed to amplify. Two species of Callicarpa L., two species of Prostanthera Labill. and one species each of Gmelina L., Vitex L., Congea Roxb. and Tectona L.f. were selected as outgroups to represent early diverging lineages in the family. Nine members of tribe Mentheae and one of Elsholtzieae were used to represent subfamily Nepetoideae excluding Ocimeae. Two species each of Ocimum and Orthosiphon Benth. (Ociminae) and one species each of Hyptis Jacq. (Hyptidinae) and Isodon (Schrad. ex Benth.) Spach were chosen to represent the remaining subtribes of Ocimeae. Living plant material was sourced from the wild and voucher specimens deposited at RNG (Mwanyambo, 2008). Herbarium material was sourced from K and RNG (acronyms following Thiers, 2018). The study sample includes 123 species including 97 of the 455 species of Plectranthinae (Table 1), building on the 31 species used by Paton et al. (2004) and including a greater range of narrowly endemic and broadly distributed species, medicinal species and representatives of groups of Plectranthus as identified in Lukhoba © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 In the monophyletic Plectranthinae, Paton et al. (2004) and Lukhoba et al. (2006) recognized two clades: the Coleus clade and the Plectranthus clade. The Coleus clade was strongly supported, whereas the Plectranthus clade had < 50% bootstrap support (Paton et al., 2004). Neither of these clades could be unambiguously diagnosed by morphological characters. The Plectranthus clade comprised the currently recognized genera Tetradenia, Thorncroftia N.E.Br., Aeollanthus Spreng., Capitanopsis S.Moore. and Dauphinea Hedge (Harley et al., 2004) and part of Plectranthus including the type species, P. fruticosus L’Hér. The Coleus clade comprised the remaining species of Plectranthus (including P. amboinicus (Lour.) Spreng., the type of Coleus), the currently recognized Anisochilus Benth. and Pycnostachys Hook. and some genera recently placed in synonymy: Leocus A.Chev., Neohyptis J.K.Morton, Englerastrum Briq., Isodictyophorus Briq. and Holostylon Robyns & Lebrun (Paton et al., 2009, 2013). Paton et al. (2004) recommended a further analysis with increased taxon sampling of Plectranthinae to clarify the best way of dividing the subtribe into monophyletic, communicable groups. Floristic treatments of Plectranthus in eastern and southern tropical Africa (Paton et al., 2009, 2013) suggested that Plectranthus spp. in the Coleus clade have a larger anterior and reduced posterior (upper) corolla lip and calyces with a pedicel attaching opposite the posterior lip of the calyx; whereas those in the Plectranthus clade have ± equal anterior and posterior corolla lobes and calyces with a centrally fixed pedicel. Most species of Plectranthinae are found in Africa, but the group also has species in tropical Asia and Australia; a few species are naturalized in the New World. Plectranthinae occupy a variety of habitats including evergreen forest margins, seasonally dry woodlands and montane grassland, some of the latter being seasonally flooded. Species are generally restricted to one of these habitat types, and can be broadly distributed or narrowly endemic (Forster, 1992, 1994, 2011; Codd, 1985; Hedge et al., 1998; Paton et al., 2009, 2013; Suddee, Paton & Parnell, 2005). Lukhoba et al. (2006) reviewed the ethnobotanical uses of the 62 Plectranthus spp. cited as having medicinal use in the literature. Medicinal use was mapped to the phylogenetic tree of Paton et al. (2004). This work suggested that 70% of medicinal Plectranthus species belonged to the Coleus Clade and that medicinal usage tended to be concentrated in particular clades across the phylogeny. In the study presented, we increase the number of sampled Plectranthus spp., including medicinal species not included in the previous analyses, and include records of medicinal use in genera of Plectranthinae not considered by Lukhoba et al. (2006). The aims PLECTRANTHUS, COLEUS AND ALLIES 357 Table 1. Summary of the taxonomic breadth of sampling and the species numbers sampled Number of species Clade (Paton et al., 2004, 2009, 2013) Number of species sampled Sample as a percentage Anisochilus Wall ex. Benth. Leocus A.Chev. Pycnostachys Hook. Plectranthus L’Hér. Dauphinea Hedge Capitanopsis S.Moore Madlabium Hedge Thorncroftia N.E.Br. Tetradenia Benth. Aeollanthus C.Mart. ex Spreng. Alvesia Welw. Total ingroup Other Ocimeae Mentheae/Elsholtzieae Early-diverging lineages Total sample 17 5 36 322 1 3 1 6 19 42 3 455 Coleus Coleus Coleus Coleus/Plectranthus Plectranthus Plectranthus Plectranthus Plectranthus Plectranthus Plectranthus Plectranthus 2 1 3 78 1 3 0 1 3 3 2 97 8 10 8 123 10 20 8 25 100 100 0 17 16 7 67 21 et al. (2006). A list of accessions used is provided in Appendix 1. An additional attempt was made to include all the species of a group of similar, presumed closely related species, to allow study of habitat changes at species level in a set of taxa that are known to occur across a range of habitat types. Following a preliminary analysis (Mwanyambo, 2008), a monophyletic group of mainly African species with non-saccate, sigmoid corolla tubes in the Plectranthus clade was chosen for this detailed analysis [representing clade 2 groups 6, B and E of Lukhoba et al. (2006); Plectranthus species 4–33 in Paton et al. (2009); species 10–31 in Paton et al. (2013)]. Twenty-five of 38 species in this group were included, DNA being unobtainable from several species. DNA EXTRACTION, PCR AND SEQUENCING Total genomic DNA was extracted from dried leaf material or from floral material where good quality leaf material was not available. Most extractions used a 2 × CTAB method following protocols in use at Biological Sciences, University of Reading (Mwanyambo, 2008) or the Jodrell Laboratory, Kew, both based on Doyle & Doyle (1987). DNA extracted at the Jodrell Laboratory was further purified through a caesium chloride gradient. Supplementary extractions were conducted at the Botanical Garden laboratories, University of Oslo, using a DNeasy Plant Mini Kit (Qiagen, Manchester, UK) following the manufacturer’s instructions. The extracted DNA was assessed for quality by visual inspection of an ethidium bromide (0.35 µg/ml) stained TAE pH 8.0 agarose gel and was then stored in water or TE buffer at −20 °C. Double-stranded DNA was amplified by the polymerase chain reaction (PCR) on an AB GeneAmp PCR System 2700 or 9700 thermocycler (Perkin Elmer, USA), in a 25 µl reaction volume. Final concentrations of: 1 × NH4 reaction buffer (Bioline, UK), 3 mM MgCl2, 200 μM each dNTP, 0.1–0.2 µM each primer and 4U TaqPolymerase; 0.1 mg/ml bovine serum albumin (BSA) or 0.9M betaine (B0300, Sigma-Aldrich, UK) were added where necessary. Three plastid DNA markers: rps16 and trnL-F (Paton et al. 2004) and trnS-G (Shaw et al., 2005) were amplified as summarized in Table S1. Amplification and sequencing of other markers was attempted; including plastid trnSGCU-trnGUUC-trnGUUC, trnCGCAycf6-psbM, ycf6-psbM-trnDGUC, rps4-trnTUGU-trnLUAA, trnDGUC-trnTGGU, rpoB-trnCGCA, psbM-trnDGUC, trnTUGUtrnL UAA and trnH GUG-psbA and nuclear G3pdh and ITS. Due to lack of or patchy amplification of products (Mwanyambo, 2008), these regions were not surveyed widely or used in this phylogenetic analysis. To improve amplification of difficult templates, BSA and betaine were used extensively. To further purify some difficult templates the following kits were employed: MF-Millipore membrane filter, VSWP02500 for drop dialysis; SureClean (Bioline, UK) and Micropure-EZ Enzyme Removers (Millipore, Bedford, MA, USA). BioTaq worked well for the majority of samples, but other enzymes used were Restorase DNA polymerase (Sigma) and Phusion DNA polymerase (New England Biolabs, Ipswich, MA, USA) for challenging samples. The details of primers used and thermocycling conditions are given in Table S1. All PCR products were purified for subsequent cycle sequencing using a Qiagen © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 Genus (Harley et al., 2004) 358 A. PATON ET AL. SEQUENCE ASSEMBLY AND ALIGNMENT Trace data were assembled, checked for trace quality and edited in Seqman II (DNAstar Inc.) and the resulting consensus sequence files were exported to Megalign (DNA Star Inc., USA) for initial alignment. FASTA alignments were later exported to BioEdit 7.2.5 (Hall, 2013). Aligned sequence lengths are: rps16, total length 1102 (1–691, 698–1102 used in analysis); trnL-F, total length 1009 (all used in analysis), trnS-G, total length 1088 (1–477, 561–1088 used in analysis) and sequences are deposited in GenBank/EBI/DDBJ (Appendix 1). Aligned files were exported in NEXUS for analysis using MrBayes (v3.2). The beginning and end of each alignment where < 80% of the DNA sequences were available were excluded from analysis using the EXCLUDE command. Further short regions with microsatellite-like or other characteristics that prevented unambiguous alignments were also excluded. PHYLOGENETIC ANALYSIS Each DNA region was explored using MrModeltest v.2.3 (Nylander, 2004) and all best fitted the GTR + I + G model based on the Akaike information criterion (Akaike, 1974). For combined data analysis, congruence between plastid DNA markers was previously tested and reported (Mwanyambo, 2008). Bayesian Inference analyses were performed in MrBayes v.3.2.6 (Ronquist et al., 2012). Gaps were treated as missing data. All the analyses were conducted with two separate runs each of four chains for 10 million iterations. Burn-in was established using Tracer v1.6 (Rambaut et al., 2014) and trees were sampled every 10 000 th generation based on tests for autocorrelation of treelength using the excel ‘corr’ function. The first 1 000 000 trees were discarded on this basis. Combinable component consensus trees generated in Bayes trees (Pagel, Meade & Barker, 2004) were used in subsequent investigations because these show the best supported clades, including those with low (< 50%) support. CHARACTER CODING OF NON-MOLECULAR DATA For phylogenetic character mapping and ancestralstate reconstruction, species of Plectranthinae were coded for habitat type, geographical distribution and medicinal use. The habitat of each species was coded from herbarium sheets, in-depth revisions (Codd, 1985; Forster, 1992, 1994, 2011; Hedge et al., 1998; Suddee et al., 2005; Paton et al., 2009, 2013) and field observations, and this was related to one of four broad categories: evergreen forest margins; seasonally dry woodland such as African Brachystegia woodland or Asian dry dipterocarp woodland; montane grassland that often burns in dry seasons and seasonally flooded grassland or marsh. A few species are found in rocky areas, but these also usually occur in one of the main habitats and were scored under the relevant habitat type or scored as polymorphic for habitat if found in more than one. Occurrence in four major geographical regions (sub-Saharan Africa, Madagascar, Tropical Asia and Australia) was recorded. The medicinal uses of Plectranthus reported in Lukhoba et al. (2006) were also mapped onto the phylogenetic tree, as individual classes of use following Cook (1995) and combined into an ‘any medicinal use’ category. Post-2006 literature was scanned for any more recently recorded uses of species of Plectranthinae included in the sample, and these are listed in Appendix 2. Species of Plectranthinae were coded for distribution by continent, habitat and medicinal use category (Treebase http://purl.org/phylo/ treebase/phylows/study/TB2:S22332). Continent and habitat were multistate characters, and medicinal use categories were binary. The states were optimized over all trees to give a character state frequency per node on the consensus tree. A reduced taxon set of outgroups was used for optimization analyses, rooted on Orthosiphon and Ocimum. The pattern of distribution of the characters on the consensus tree was explored using randomization tests in Mesquite v3.04 (Maddison & Maddison, 2015) using the Reshuffle Character option following the protocol in Bytebier et al. (2011). Each coded character was subjected to reshuffling 100 000 times to generate a frequency graph of treelengths. Characters for which the actual steps on the consensus tree fell outside the 95th percentile of the randomized distribution were considered significantly different from random, either by being clustered (> 95 percentile of shortest tree lengths) or over-dispersed (> 95 percentile of longest treelengths). Tests were conducted for distribution, habitat and medicinal use characters. © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 QIAquick PCR Purification Kit (Qiagen, Manchester, UK) following the manufacturer’s instructions with slight modifications. PE buffer was added and left to stand for 5-15 min during the washing step to increase yield. DNA cycle sequencing was performed on the cleaned products in 10 µl reaction volumes. Reaction components were BigDye Terminator v3.1 (4 µl: Thermo Fisher Scientific Life Sciences, Waltham, MA, USA), 1 µM primers (as used for PCR, 1.6 µl), Nanopure water (2.4 µl) and DNA template (3 µl). Thermocycling parameters were 25 iterations of: 10 s at 96 °C; 5 s at 50 °C; 4 min at 60 °C. The products were run on an ABI Prism 3100 Genetic Analyser [Life Technologies, Carlsbad, CA, USA (with 50 cm 16 capillary array)]. Due to sequencing difficulties, the outgroups and three Plectranthus spp. [P. lactiflorus (Vatke) Agnew, P. ecklonii Benth. and P. amboinicus] were not sequenced for the trnS-G region. PLECTRANTHUS, COLEUS AND ALLIES 359 clade comprising the Madagascan endemic genera Capitanopsis, Dauphinea and Plectranthus bipinnatus A.J.Paton (previously recognized as the monotypic Perrierastrum Guillaumin) (1.00)] and Clade IV [a group of African Plectranthus with one Asian member P. mollis (Aiton) Spreng. (1.00)]. Clade IV was more extensively sampled and internal branches were generally shorter and less well supported than in other clades. The backbone of the tree supporting the relationships between clades I and II, clade III and clade IV was not strongly supported (< 0.50) and the MrBayes and the dated BEAST analyses differed in the topological ordering of these clades. In the Coleus clade, two clades with a PP of 1.00 were recovered (Fig. 1B). Coleus clade A contains species of Pycnostachys and Anisochilus in addition to species of the currently recognized Plectranthus, and Coleus clade B contains the type species of Coleus, P. amboinicus. The dating analysis supported the Plectranthinae + Ociminae clade as having diversified from c. 24.65 Mya; Plectranthinae diversified from c. 21.6 Mya and the Plectranthus and Coleus clades diversified from c. 19.2 and 18.0 Mya, respectively. Alvesia & Aeollanthus and clades I–IV, diversified between 4.2 Mya (clade III – Capitanopsis) and 10.75 Mya (clade IV) in the Plectranthus clade, with the Coleus A and B clades diversifying from c. 16.9 and 14.4 Mya, respectively. Species level divergence times varied from 0.5 to 8.0 Mya in the more densely sampled clade IV. The mean dates of the crown nodes of all clades named above and the 95% highest posterior density (HPD) intervals are given in Table 2 and Supporting information, Figure S1. RESULTS OPTIMIZATION OF GEOGRAPHY, HABITAT AND TREE TOPOLOGY AND DATING MEDICINAL USE DATING In the Bayesian analysis using MrBayes and BEAST, Plectranthinae were retrieved as monophyletic with a posterior probability (PP) of 1.00; in that clade, two sister groups, the Plectranthus clade and Coleus clade [as recognized by Paton et al. (2004) and Lukhoba et al. (2006)] both also had a PP of 1.00 (Fig. 1A, B). The dated tree produced in the BEAST analysis is topologically the same as that produced by MrBayes for all nodes with a posterior probability > 50% (Supporting Information, Fig. S1). At the base of the Plectranthus clade (Fig. 1A) Alvesia Welw. formed the first branch with a PP of 1.00 and the remainder of the Plectranthus clade was also supported as monophyletic (0.82). Five clades in this group were strongly supported: Aeollanthus (1.00); clade I [a clade comprising Tetradenia and Thorncroftia (0.96)], supported as sister (1.00) to clade II [Plectranthus s.s., a clade of African and Madagascan species morphologically similar to the type species of Plectranthus, P. fruticosus (1.00)]; clade III [a Geography All migrations occurred from Africa to other continents. When distribution states were mapped onto the phylogenetic results, the majority of internal nodes were optimized as African (Supporting information, Fig. S2). Character optimization showed nine continental migrations, significantly fewer (P < 0.01) moves between continents than that expected from a randomized distribution (P = 0.99, > 17 – < 21 steps). Four clades are non-African: the Australian Plectranthus congestus R.Br. – P. parviflorus Willd. clade that contains all the Australian species, and the Asian P. glabratus (Benth.) Alston – P. parishii Prain clade both in Coleus clade A; the Madagascan endemic clade III, Capitanopsis and Madagascan species of Tetradenia in clade I (Figs 1, S2). Of the clades identified in the previous section only Alvesia and Aeollanthus are restricted to tropical Africa, although one Aeollanthus spp. is naturalized in Brazil. Isolated © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 Phylogenetic and divergence time analysis for the combined data set was performed using BEAST v2.4.6 (Bouckaert et al., 2014), following file processing using BEAUti. A trial analysis visualized in Tracer v1.6 (Rambaut et al., 2014) showed there was no requirement to partition the data. Rate constancy was rejected for all partitions and therefore we used the relaxed clock model. The GTR+G+I, birth death model was implemented. All partitions fit the same optimal model measured by AIC and were combined before analysis. Two separate BEAST analyses were conducted and subsequently combined to ensure a run did not stall at a local optimum. Rooting follows the MrBayes analysis. There is no fossil evidence to constrain the dates of clades in Ocimeae. The sampling of the analysis is heavily skewed towards Plectranthinae and thus the priors used by Drew & Sytsma (2012) for dating Mentheae would be inappropriate. Therefore, the most recent common ancestor of the Ocimeae/Elsholtziae taken from Drew & Sytsma (2012) was used as a calibration point. This node was constrained at a mean of 56 Mya and SD of 5% with normal distribution. A Yule tree prior was used given that we were sampling individuals from a wide range of species. Tree building ran for 40 million generations sampled every 10 000th generation. Stationarity was established by the four millionth generation. Clock models were unlinked. Resulting trees were explored in TreeAnnotator v1.6.1 prior to visualization in the program FigTree v1.3.1. 360 A. PATON ET AL. terminal taxa with extra-African distributions not included in the clades just described are also mostly found in Africa except P. emirnensis (Baker) Hedge (clade II) in Madacascar, P. mollis (clade IV) and P. scutellarioides (L.) R.Br. (Coleus clade B) in Asia (Supporting information, Fig. S2). In Plectranthus clade I, Tetradenia is split between Africa and Madagascar, with no species in common between the regions. Madagascan Tetradenia spp. may form a monophyletic group, although the sample of this genus is small. The other genus in clade I, Thorncroftia, has six species all restricted to southern Africa. Plectranthus clade II (Plectranthus s.s.) is found in Africa and Madagascar, although the Madagascan members of this clade are only represented in the analysis by one species. Plectranthus clade IV is mainly African with only one species included in the analysis, P. mollis, which is found in Asia. One species, P. flaccidus Gürke, occurs in both Africa and Madagascar, but there are no endemic Madagascan species of this clade. Habitats Coleus clades A and B and clade IV of the Plectranthus clade all have representatives in seasonally dry © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 Figure 1. A, Compatible component consensus tree produced from MrBayes analysis showing the outgroups, Plectranthus clade (P) with Alvesia, Aeollanthus and Plectranthus clades I–IV and the position of the Coleus clade (Fig. 1B). Posterior probabilities are given below the branches: blue > 0.50 support, red < 0.50. B, Compatible component consensus tree produced from MrBayes analysis, Coleus clade comprising Coleus clades A and B. Posterior probabilities are given below the branches: blue > 0.50 support, red < 0.50. PLECTRANTHUS, COLEUS AND ALLIES 361 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 Figure 1. Continued woodland, evergreen forest margins and montane grassland, although there was a notable grouping of related species in similar habitats. Thirty-one changes of habitat were recorded in the Plectranthinae, which is significantly fewer than random [P < 0.01 (P = 0.99, > 39 – < 51 steps), Supporting information, © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 362 A. PATON ET AL. DISCUSSION Table 2. Crown node ages of recognized clades (Mya). Columns give the mean age and the minimum and maximum of the 95% HPD calculated in BEAST min mean max 14.4 16.8 13.8 3.7 3.7 6.1 20.7 21.6 19.2 7.3 6.5 9.7 26.6 26.7 23.8 11.4 9.6 13.1 1.0 4.8 3.0 7.5 5.3 10.6 1.8 4.2 7.0 7.5 14.2 13.1 10.7 10.75 18.0 16.9 14.4 14.2 22.5 20.9 18.3 Fig. S3]. Aeollanthus, Tetradenia and Thorncroftia can also be found in these three habitats when all species including those not in the analysis, are considered. Alvesia and the Madagascan clade III are only found in dry woodland, whereas clade II, Plectranthus s.s., is found mainly in evergreen forest margins or forested gorges (Supporting information, Fig. S3). The flooded grassland habitat is only recorded in Coleus clade A. Medicinal use Several instances of medicinal use not recorded in Lukhoba et al. (2006) are reported in Appendix 2. Thirty species represented in the phylogenetic tree have recorded medicinal uses. This broad category (Medicinal use) has 24 steps, just significantly fewer than random (P = 0.05, > 24, < 31, Supporting information, Fig. S4). None of the individual classes of medicinal use reported in Lukhoba et al. (2006) was distributed across the phylogenetic tree in a pattern significantly different from randomized data. Of the named clades, Coleus clade B had the most recorded medicinal use, with 13 of the 22 sampled species being used. Internal nodes are only optimized for medicinal use in the following groups: the P. lasianthus (Gürke) Vollesen – P. lactiflorus clade, the P. scutellarioides – P. shirensis (Gürke) A.J.Paton clade, the P. alpinus (Vatke) O.Ryding – P. diversus S.T.Blake clade, all in Coleus clade B and in clade A, the Pycnostachys reticulata (E.Mey.) Benth. – Pycnostachys urticifolia Hook. clade (Figs 1B, S4). The phylogenetic analysis presented here supports the monophyly of Plectranthinae, but Plectranthus, as currently recognized, is paraphyletic. The division of Plectranthinae into Plectranthus and Coleus clades as suggested by Paton et al. (2004) and Lukhoba et al. (2006) is supported, with support values much increased over these previous analyses. There are several morphological features that can be used to diagnose the clades identified in the results and thus splitting Plectranthus into smaller monophyletic clades is the preferred option. Tetradenia, is morphologically distinct from the rest of Plectranthinae (Harley et al., 2004; Paton et al., 2004; Phillipson & Steyn, 2008). It has actinomorphic corollas and anterior and posterior pairs of stamens separated by a clear gap at the point of insertion to the corolla, rather than having a strongly zygomorphic corolla and stamens contiguous at the base of the lower corolla lip as in all other Plectranthinae. These differences make morphological diagnoses of an enlarged Plectranthus comprising the whole of Plectranthinae, including Tetradenia, impossible. The genera recognized here are summarized in Table 3 and Figure 2. Although only plastid markers have been used, the fact that the strongly supported clades recognized at generic rank are either existing genera with morphological apomorphies and have been previously informally recognized on the basis of morphology [Plectranthus clade IV (Paton et al., 2009, 2013)] or have strong morphological similarity (Plectranthus clade III), suggests that the recognized groups themselves are robust, although the deeper relationships between them still needs to be fully resolved. The differences between the generic delimitation here and that proposed by Harley et al. (2004) and Paton et al. (2009, 2013) are that: Coleus is recognized at generic rank, corresponding to the Coleus clade with Pycnostachys, Leocus, Solenostemon and Anisochilus placed in synonymy; Plectranthus s.s. is restricted to species similar to the type of Plectranthus, P. fruticosus (Plectranthus clade II); a new genus, Equilabium, is created here to include species in clade IV; Capitanopsis as currently circumscribed is paraphyletic as P. bipinnatus is included in it, and therefore the Madagascan endemic and monotypic genera Dauphinea, Madlabium, Perrierastrum (the last previously considered as Plectranthus by Harley et al., 2004) are all placed in synonymy of the Madagascan endemic Capitanopsis, the earliest generic name (clade III). Although Madlabium is not sampled in our analysis, its corolla morphology with a truncate corolla throat with small upper lobes is similar to the other taxa in the clade. The necessary new combinations in © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 Ociminae Plectranthinae Plectranthus clade Alvesia Aeollanthus Plectranthus clade I. Tetradenia & Thorncroftia Tetradenia Plectranthus clade II. Plectranthus s.s. Plectranthus clade III. Capitanopsis Plectranthus clade IV Coleus clade Coleus clade A Coleus clade B PHYLOGENY AND TAXONOMY PLECTRANTHUS, COLEUS AND ALLIES 363 Table 3. Genera recognized and diagnostic features Recognized Genus and important synonyms Coleus Lour. Anisochilus, Holostylon, Isodictyophorus, Leocus, Pycnostachys, Solenostemon Alvesia Welw. Morphological diagnostic characters Coleus Calyx funnel shaped. Corolla with upper lip much shorter than lower; pedicel attachment opposite upper calyx lip Tetradenia Benth. Plectranthus clade: Alvesia Plectranthus clade: Aeollanthus Plectranthus clade I Thorncroftia N.E.Br Plectranthus clade I Plectranthus L’Hér. Germanea Lam. Plectranthus clade II Plectranthus s.s. Capitanopsis S.Moore Dauphinea, Perrierastrum, Madlabium Equilabium Mwanyambo, A.J.Paton & Culham Plectranthus clade III Calyx three-lobed- expanded and membranous in fruit Calyx basally circumscissile; calyx tube distally dorso-ventrally flattened Corolla actinomorphic; corolla tube straight, funnelshaped; anterior and posterior pairs of stamens separated by a clear gap at point of insertion on the corolla Corolla zygomorphic; corolla tube straight or curved downwards, cylindrical. Lateral lobes of the corolla deflexed towards the lower lip, rather than forming a four-lobed upper lip as in all other genera except Tetradenia. Calyx funnel-shaped, Corolla with upper and lower lips equal in length; lateral lobes of the calyx adjacent to the lower lobes of the calyx Calyx variously shaped, expanded and membranous to cylindrical. Corolla throat truncate; lateral and upper lobes short. Calyx funnel-shaped. Upper and lower corolla lips equal in length; corolla tube strongly sigmoid; lateral lobes of the calyx held midway between uppermost and lowermost lobes Aeollanthus Spreng. Plectranthus clade IV Capitanopsis and Equilabium are formally made at the end of this paper. Coleus has previously been recognized as a separate genus, and a good account of the taxonomic histories of Coleus and Plectranthus was provided by Codd (1975). Most treatments following Codd have merged Coleus into Plectranthus, although it was maintained as a genus in the Flora of China by Li & Hedge (1994). Coleus was diagnosed by having fused stamens (Bentham, 1832; Li and Hedge, 1994), but this character is homoplasious as shown by Paton et al. (2004), and the characters listed in Table 3 and used by Paton et al. (2009, 2013) to identify the Coleus clade provide a more stable basis for diagnosis of Coleus as a genus. Solenostemon was maintained as a separate genus by Codd (1975), but Solenostemon here represented by P. scutellarioides, P. shirensis, P.sigmoideus A.J.Paton and P. schizophyllus Baker is paraphyletic (Fig. 1B). Recognition of Solenostemon as a genus would also render Coleus paraphyletic and so it is not recognized at generic rank. Intermediates between Coleus and its synonyms previously recognized at generic rank by Harley et al. (2004) have been reported in the past: Pycnostachys (Paton et al., 2009, 2013); Anisochilus (Suddee et al., 2014) and Leocus (Pollard & Paton, 2009). In addition, none of the sections of Coleus or subgenera of Plectranthus recognized by Codd (1975) is monophyletic. Further work is required to identify morphological characters to diagnose Coleus clades A and B, or other clades within these. The merging of the Madagascan endemic and monotypic genera of clade III into Capitanopsis has not been suggested before. Taxonomic over-splitting of Madagascan clades into several genera has been reported by Buerki et al. (2013), investigating the phylogenetic clustering of Madagascan endemic genera. These authors also identified recent radiations and extinctions as factors contributing the recognition of endemic genera. Both these factors might have contributed to the over-splitting of this clade. The variation in calyx form from funnel-shaped, expanded and membranous in Capitanopsis to tubular and non-membranous in Dauphinea, Madlabium and Perrierastrum (Hedge et al., 1998) might reflect recent rapid radiation in the clade. The stem age of Capitanopsis clade III is 14.3 Mya, but the extant © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 Clade (Fig. 1) 364 A. PATON ET AL. species diversification occurs much later with the crown node at 4.2 Mya (Table 2), which might reflect a long period of stasis, but more probably extinction of earlier branches. DATING AND CHARACTER OPTIMIZATION The clades recognized here in the Plectranthus clade and suggested for generic recognition diversified in the late Miocene to Pliocene [10.75 Mya (Equilabium), 3.0 Mya (Tetradenia)]. However, the Coleus clade has an earlier crown date (18.0 Mya), although morphologically identifiable subclades have not been recognized in this clade suggesting greater degree of morphological continuity of form in the Coleus clade as opposed to the Plectranthus clade. Given the lack of available fossil evidence directly relevant to Plectranthinae, the dated phylogenetic tree should be regarded as preliminary and different sampling between the clades may result in different dating results being found. There are few eco-phylogenetic studies of African plants, particularly of those occurring in seasonally dry woodland, wooded grassland or savanna floras (Bakker et al., 2005; Holstein & Renner, 2011; Linder, 2014), even though these habitats are widespread. This is in part due to the difficulties associated with species-level © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 Figure 2. Summary diagram showing genera recognized in Plectranthinae with species numbers and dates of major clades. Species featured from top to bottom are: Alvesia rosmarinifolia (Photograph: M. Finckh); Aeollanthus buchnerianus (Photograph: B. Wursten); Tetradenia nervosa, Plectranthus saccatus (Photographs: both RBG Kew), Capitanopsis angustifolia (Photograph: P. Lowry), Plectranthus petiolaris (Photograph: N. Couch); Plectranthus barbatus (Photograph: RBG Kew). 280 12 61 46 13 8 35 455 97 21% 36 1 0 2 0 0 0 39 24 62% 152 11 9 43 13 8 35 271 52 19% Africa endemic Africa and Madagascar Madagascar endemic Tropical Asia endemic Widespread in Asia or Asia and Africa Temperate Asia endemic Australia Total Sample % sampled 3 0 0 0 0 0 0 3 2 67% 42 0 0 0 0 0 0 42 3 7% 20 0 9 0 0 0 0 29 4 14% 27 0 37 1 0 0 0 65 7 11% 0 0 6 0 0 0 0 6 5 83% Total number of species Plectranthus clade III Capitanopsis Plectranthus clade II Plectranthus s.s. Plectranthus clade I Tetradenia and Thorncroftia Aeollanthus Alvesia Coleus clade Distribution Shifts in habitat are significantly fewer than if habitat was randomly distributed on the phylogenetic tree. The seasonally flooded habitat is only found in Coleus clade A in the analysis (Supporting Information, Fig. S3), although a few unsampled Aeollanthus spp., including A. engleri Briq., P. orbicularis Gürke and P. pulcherissimus A.J.Paton of clade IV also occur in this habitat. With the exception of these few species, habitat shifts have been between seasonally dry woodland, evergreen forest margins and montane grassland. In the Coleus clade and Plectranthus clade IV there are still several shifts in habitats, predominantly from dry woodland to either montane grassland or forest. Similar habitat shifts were reported in Coccinia Wight & Arn. (Cucurbitaceae), moving between woodland, forests and arid habitats. (Holstein & Renner, 2011). As in Coccinia, the diversification of Plectranthinae clades date from the mid to late Miocene as the climate became cooler and drier habitats expanded and rainforests shrunk in range in Africa (Holstein & Renner, 2011; Hoetzel et al., 2013; Pokorny et al., 2015). The habitats occupied by Plectranthinae are contiguous. Dry woodlands are frequently found at midelevations, and mosaics of evergeen forest patches and Table 4. Total number of species found in each major clade from each of the geographical areas HABITATS © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 365 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 sampling. An attempt was made to achieve this level of sampling in Plectranthus clade IV (Equilabium), but the lack of well-preserved herbarium material and the difficulty of field collection in many different countries in a short time period made this challenging and only 62% of species were sampled in this clade (Table 4). For many studies of large, broadly distributed groups, comprehensive species-level sampling is not possible and there is a need for better collection of material for DNA-based research (Gaudeul & Rouhan, 2013). Due to the lack of species-level sampling, conclusions based on character optimization are preliminary and need to be tested with more in-depth sampling. The morphological features outlined in Table 3 were used to place all species of Plectranthinae into one of the recognized clades. This information is used to examine whether patterns of character distribution observed in the analysis might be artefacts of incomplete sampling. The distribution of species between clades and geographical area is summarized in Table 4 and the character optimization studies are interpreted in this context. Table 4 gives the total number of species found in each major clade from each of the geographical areas, using morphological characters to place species not included in the analysis. It has not been possible to give species numbers for the Coleus clades A and B due to the difficulties of morphologically diagnosing these subclades as discussed above. Plectranthus clade IV Equilabium PLECTRANTHUS, COLEUS AND ALLIES 366 A. PATON ET AL. in several different genera, including Ocimum, Orthosiphon, Platostoma and Syncolostemon, less speciation in the forest habitat in Ocimineae is seen than in Plectranthinae. GEOGRAPHICAL DISTRIBUTION Continental migrations are significantly fewer than expected from randomized distribution and there are no migrations into Africa, although some extant species occur both in Africa and Asia and are discussed below. Three historical migrations of Plectranthinae to Asia from Africa are shown: P. mollis (clade IV); P. scutellarioides (Coleus clade B) and the P. glabratus – P. parishii clade (Coleus clade A) (Figs 1, S2). The Indian P. gardneri Twaites in clade II (not sampled) probably represents another migration to Asia. Only one of these migrations shows speciation in Asia: the P. glabratus – P. parishii clade containing Anisochilus in Coleus clade A. However, Asian speciation in the P. scutellarioides and P. shirensis clade in Coleus clade B is also likely, due to there being other species morphologically similar to P. scutellarioides in Asia and the Indian P. subincisus Benth. is suggested to be closely related to the Asian P. mollis in clade IV, representing another possible Asian speciation (Smitha & Sunojkumar, 2015). All Australian species, which are morphologically similar, arose from a single migration event (1.6 Mya) from Asia, but increased sampling is necessary to confirm the monophyly of the Australian species. There are three historical migration events to Madagascar: clade I (Tetradenia and Thorncroftia); clade II (Plectranthus s.s.) and clade III (Capitanopsis) (Supporting Information, Fig. S2). The migration events from Africa to Madagascar or to Asia in clades I and III occurred in dry woodland clades, whereas those in clade II were more likely through forest habitat species. Eleven species currently occur in both Africa and Madagascar, and represent both forest and dry woodland species. This pattern of migration from Africa to Madagascar mainly through dry habitats, but also with some through wetter forest, is also seen in Apocynaceae subfamily Asclepiadoideae (Meve & Liede, 2002), although, unlike this group, there are no migrations from Madagascar to Africa in Plectranthinae. A few species, all members of Coleus clade B, are found in Africa, Madagascar and tropical Asia, including P. barbatus Andr., P. rotundifolius Spreng., P. amboinicus, P. hadiensis (Forssk.) Sprenger P. montanus Benth. and P. caninus Roth. These species are all recorded as having medicinal uses and so the broad distribution might reflect trade and human transport. USES In Plectranthinae, the pattern of distribution of any particular medicinal use as categorized by Lukhoba © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 montane grassland occur in adjacent higher elevations, particularly in eastern Africa. With variations in aridity, the boundaries between these habitats are likely to move, creating a dynamic landscape (Oliveras & Malhi, 2016; Fer et al., 2017), influencing the spread and restriction of species distributions. Another parallel with Coccinia is that some forest species of Plectranthus clades II and IV and the Coleus clade have wide discontinuous distributions, perhaps reflecting forest expansion during the Pleistocene (Holstein & Renner, 2011). Such widespread species include P. kamerunensis Gürke and P. laxiflorus Benth. in E.H.F.Meyer (clade IV), P. alboviolaceus Gürke (clade II) and P. alpinus (Coleus clade). On the other hand, the disjunct distribution of P. leptophyllus (Baker) A.J.Paton (Coleus clade) found in the mountains in Uganda, coastal Kenya and Tanzania and in eastern Zimbabwe and adjacent Mozambique, or the occurrence of P. sylvestris Gürke and P. melleri Baker in Africa and Madagascar, probably reflect long-distance dispersal, as suggested for Coccinia schliebenii Harms (Holstein & Renner, 2011; Paton et al., 2009, 2013). Fire might influence the distribution of clades in Plectranthinae. Clade IV and the Coleus clade frequently inhabit dry woodland and montane grassland prone to burning, often having thick underground fire resistant rootstocks, a character rare in the other clades. In contrast, Aeollanthus, Capitanopsis (clade III) and Tetradenia and Thorncroftia (clade I) are most often associated with seasonally dry habitats and often have succulent or thick leaves and sometimes stems. These groups are mostly absent from evergreen forest margins and areas of montane grassland that are prone to burning, unless sheltered by rocks in these habitats. Clade II (Plectranthus s.s.) also occurs in naturally fire-free evergreen forest margins or forested gorges in the summer rainfall area of southern Africa and in Madagascar. The clade is largely absent from dry woodland, although c. 10% of species in Africa and Madagascar have been recorded from that habitat, but these are often succulent and lack fire resistant rootstocks. Maurin et al. (2014) suggested a Pleistocene origin, < 5.3 Mya, for the fire-resistant geoxylic life form seen in clade IV and the Coleus clade. This date is consistent with recent species and habitat diversification across habitats in these clades. Ociminae, sister to Plectranthinae, have few evergreen forest margin species and there are about four times as many species of Plectranthinae as Ociminae in the forest margin habitat (Hedge et al., 1998; Suddee et al., 2005; Paton et al., 2009, 2013). More resolved relationships of Ociminae are needed to investigate the pattern of migration and speciation and the relative importance of factors in explaining this difference between sister groups. However, as the relatively few forest species of Ocimineae occur PLECTRANTHUS, COLEUS AND ALLIES CONCLUSIONS The major taxonomic conclusions for this medicinally and horticulturally important group are that Coleus, including Solenostemon with > 270 species, is sister to the remainder of Plectranthinae and merits generic recognition. Plectranthus s.s. (clade II) is a group of only c. 65 species and is sister to a clade comprising the morphologically distinct Tetradenia and Thorncroftia. The species of clade IV need to be moved from Plectranthus into a separate genus, Equilabium. The Madagascan clade III can be recognized as Capitanopsis, with three previously recognized monotypic genera moved into it. Although changes in habitat are significantly fewer than would be expected if there were no phylogenetic pattern, shifts between habitats do occur. The lack of complete species-level phylogenetic trees, even when good taxonomic accounts exist, remains a barrier to understanding the details and frequencies of these changes. Herbaria represent an important resource for such studies and new analytical techniques might provide opportunities for creating such detailed phylogenetic trees from degraded DNA from herbarium specimens (Dodsworth, 2015). However, such studies will still rely on good taxonomic accounts and well-curated herbaria. Information on the medicinal use of plants remains fragmented and difficult to synthesize. Papers that deal with medicinal use tend to be regionally based and rarely present results on the underlying chemistry of the plant, whereas papers focusing on the biochemistry and medicinal use tend to deal with only a few species. Work such as that presented here and by Saslis-Lagoudakis et al. (2011b) provides a framework for further understanding of the relationship between the use of plants and shared biochemical pathways or properties. TAXONOMIC NOVELTIES A conspectus of all species of Coleus, Plectranthus and Equilabium is being prepared separately, including for the first time the formal placing of Anisochilus and Pycnostachys in the synonymy of Coleus. Equilabium is described and the necessary combinations in Capitanopsis are made below. Equilabium Mwanyambo, A.J.Paton & Culham gen. nov. urn:lsid:ipni.org:names:60475121-2 Type species: Equilabium laxiflorum (Benth.) Mwanyambo, A.J.Paton & Culham comb. nov. urn:lsid:ipni.org:names:60475122-2 Basionym: Plectranthus laxiflorus Benth. in E.H.F.Meyer, Comm. Pl. Afr. Austr.: 228 (1838). Lectotype, chosen by Codd (1975): South Africa, © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 et al. (2006) is not significantly different from random, contrasting with previous studies suggesting that medicinal use is not randomly distributed across phylogenetic trees (Douwes et al., 2008; Rønsted et al., 2008; Saslis-Lagoudakis et al., 2011a). However, when all medicinal uses are regarded together as a single character, the pattern of distribution is just significant at the 0.05 level. Several instances of use have been recorded in Tetradenia riparia (Hochst.) Codd. This name was previously used to cover several tropical African species now separated as distinct species following Phillipson & Steyn (2008). It is possible that some of the recorded medicinal use attributed to T. riparia actually applies to T. tanganyikae Phillipson and, if that species is scored as having medicinal use, then the pattern of medicinal use is no longer significantly different from random across the phylogenetic tree. A non-significant pattern of any medicinal use has also been reported in Pterocarpus Jacq. (Fabaceae; Saslis-Lagoudakis et al., 2011b), although in that study some individual classes of medicinal use were significantly clustered on the phylogenetic tree, unlike in Plectranthinae. The lack of a clear phylogenetic pattern in medicinal use recorded here might be a consequence of relatively sparse and inconsistent depth of sampling across clades and/or the equal scoring of non-homologous medicinally active compounds that may derive from different biosynthetic pathways. The lack of a strong pattern across the whole of Plectranthinae is consistent with the findings of Kelly, Grenyer & Scotland (2014), who suggested that phylogenetic distance is correlated with feature similarity for only short distances along the tree, and those of Rønsted et al. (2012), who suggested that the strength of correlation is dependent on taxonomic scale. Investigations into potential medicinal use could be focused on clades in which interior nodes are optimized as showing this character. Clade B includes the highest number of medicinal species recorded in Lukhoba et al. (2006) Several species in the P. lasianthus–P. lactiflorus clade (clade B, Figs 1, S4) with medicinal use were placed by Codd (1975) in Plectranthus subgenus Calceolanthus Codd. The subgenus was diagnosed by having a calyx with a dense beard of hairs in the calyx throat, although the analysis presented here suggests not all species in this clade share this character, e.g. P. puberulentus K.Morton and P. lanuginosus (Benth.) Agnew. Despite this, species sharing this character are likely to be closely related. Species unstudied for medicinal use and with dense hairs in the calyx throat include P. pentheri (Gürke) van Jaarsv. & T.J.Edwards, P. xylopodus Lukhoba & A.J.Paton, P. ornatus Codd, P. grandicalyx (E.A.Bruce) J.K.Morton and P. otostegioides (Gürke) Ryding (Paton, 2009, 2013). 367 368 A. PATON ET AL. Capitanopsis brevilabra (Hedge) Mwanyambo, A.J.Paton & Culham comb. nov. urn:lsid:ipni.org:names:77165451-1 Basionym: Dauphinea brevilabra Hedge, Notes Roy. Bot. Gard. Edinburgh 41: 119 (1983). Type: Material cultivated in Edinburgh, originally collected in Madagascar, Dist. de Fort Dauphin, Ste-Luce, Hardy & Rauh 2876 (E, holotype). Capitanopsis magentea (Hedge) Mwanyambo, A.J.Paton & Culham comb. nov. urn:lsid:ipni.org:names:77165452-1 Basionym: Madlabium magenteum Hedge, Fl. Madag. 175: 261. 1998. Type: Madagascar, forêt d’Ampandra, 6 km au nord de Vohemar, sur la route vers Ambilobe, Lavranos 28995 (E, holotype P, isotype). Capitanopsis oreophila (Guillaumin) Mwanyambo, A.J.Paton & Culham comb. nov. urn:lsid:ipni.org:names:77165453-1 Basionym: Perrierastrum oreophilum Guillaumin, Bull. Mus. Natl. Hist. Nat., sér. 2, 2: 694. 1930. Type: Madagascar, massif de l’Andringitra, Perrier de la Bâthie 13729 (P, holotype). Synonym: Plectranthus bipinnatus A.J.Paton, Kew Bull. 58: 488. 2003. Type as for Perrierastrum oreophilum ACKNOWLEDGEMENTS Thanks are due to Victor Albert for hosting Montfort Mwanyambo at the University of Oslo and the Norwegian Ministry of Foreign Affairs, SADC Biodiversity Support Programme MWI-04/333 for funding Montfort Mwanyambo during his PhD studies. We thank Bart Wursten (Aeollanthus), Manfred Finckh (Alvesia), Peter Lowry (Capitanopsis) and Neil Crouch (Plectranthus petiolaris) for permission to use photographs in Figure 2. We thank two anonymous reviewers for their constructive comments on an earlier version of the manuscript. REFERENCES Akaike H. 1974. A new look at the statistical model identification. IEEE Transactions on Automatic Control 19: 716–723. Anthoney ST, Ngule CM. 2013. Chemical constituents of infused Plectranthus argentatus leaves. World Journal of Science 1: 151–160. Awas T, Demissew S. 2009. Ethnobotanical study of medicinal plants in Kafficho people, southwestern Ethiopia. In: Ege S, Aspen H, Teferra B, Bekele S, eds. Proceedings of the 16th International Conference of Ethiopian Studies, Vol. 3. Trondheim: NTNU-Trykk Press, 711–726. Bakker FT, Culham A, Marais EM, Gibby M. 2005. Nested radiation in Cape Pelargonium. In: Bakker FT, Chatrou LW, Gravendeel B, Pelser PB, eds. Plant species-level systematics: new perspectives on pattern & process. Ruggel: Gantner Verlag, 75–100. Bascombe K, Gibbons S. 2008. Anti-staphylococcal activity of novel diterpenes isolated from Pycnostachys reticulata. Planta Medica 74: 1047. © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 KwaZulu-Natal, between Umzimkulu and Umkomaas Rivers, Drège 3586 (K!, lectotype). Equilabium is similar to Coleus and Plectranthus in having a strongly zygomorphic corolla and a funnelshaped calyx with a broad upper lip and a lower lip with four lanceolate teeth, and in having stamens contiguous at the point of insertion at the base of the lower corolla lip. It differs from Coleus in having equal corolla lips, rather than the upper lip being shorter than the lower, and in having a symmetrical attachment of the pedicel to the base of the calyx, rather than the pedicel attaching to the calyx asymmetrically, opposite the upper lip of the calyx. It differs from Plectranthus by having a strongly to shallowly sigmoid corolla tube that is parallel-sided at the base (slightly saccate in only one species), as opposed to a straight or downward-curved corolla tube that is usually gibbous or saccate at the base, and in having the lateral lobes of the calyx held midway between the upper lip and lowermost teeth of the lower lip, rather than having the lateral teeth much closer to the lowermost teeth than the upper lip. Perennial or annual, sometimes succulent, subshrubs or herbs or geoxylic herbs, usually aromatic. Leaves simple, sometimes succulent, opposite. Inflorescence thyrsoid, with cymes sessile or pedunculate, bractate, rarely bracteolate, one- to three- (to seven-) flowered; bracts caducous or persistent. Pedicel attaching to calyx symmetrically. Calyx funnel-shaped, straight, twolipped, five-lobed; posterior lobe lanceolate to obovate, sometimes decurrent, usually broader than other lobes; lateral lobes lanceolate or deltoid, held between the posterior and anterior lobes; anterior lobes lanceolate; throat open, glabrous. Corolla strongly two-lipped, fivelobed, white, blue or purple; posterior lip, ascending or erect, four-lobed, median lobes exceeding lateral; anterior lip horizontal, cucullate, enclosing stamens, sometimes frilled at apex, corolla-tube narrow, sigmoid, parallel-sided at base, dilating distally. Stamens four, free at base; anthers ± orbicular. Style apex bifid with subulate lobes. Disc four-lobed with anterior lobe larger. Nutlets, ovoid, glabrous, usually mucilaginous. The new combinations in Capitanopsis are made below: PLECTRANTHUS, COLEUS AND ALLIES Govaerts R, Paton A, Harvey Y, Navarro T, García Peña MR. 2016. World checklist of Lamiaceae. Facilitated by the Royal Botanic Gardens, Kew. Published on the Internet; http://apps.kew.org/wcsp/. Retrieved 15 Jan 2016. Hall T. 2013. BioEdit. Available at: http://www.mbio.ncsu.edu/ BioEdit/bioedit.html. Harley RM, Atkins S, Budantsev A, Cantino PD, Conn B, Grayer RJ, Harley MM, De Kok R, Krestovskaja T, Morales A, Paton AJ, Ryding O, Upson T. 2004. Labiatae. In: Kadereit JW, ed. The families and genera of vascular plants (Lamiales), Vol. 6. Berlin: Springer, 167–275. Hedge IC, Clement RA, Paton AJ. Phillipson PB. 1998. Flore de Madagascar et des Comores, Famille 175: Labiatae. Paris: Muséum National d’Histoire Naturelle. Hoetzel S, Dupont L, Schefuß E, Rommerskirchen F, Wefer G. 2013. The role of fire in Miocene to Pliocene C4 grassland and ecosystem evolution. Nature Geoscience 6: 1027–1030. Holstein N, Renner SS. 2011. A dated phylogeny and collection records reveal repeated biome shifts in the African genus Coccinia (Cucurbitaceae). BMC Evolutionary Biology 11: 28. Juch M, Rüedi P. 1997. Isolation, structure, and biological activities of long‐chain catechols of Plectranthus sylvestris (Labiatae). Helvetica Chimica Acta 80: 436–448. Kelly S, Grenyer R, Scotland RW. 2014. Phylogenetic trees do not reliably predict feature diversity. Diversity and Distributions 20: 600–612. Kaou AM, Mahiou-Leddet V, Hutter S, Aïnouddine S, Hassani S, Yahaya I, Azas N, Ollivier E. 2008. Antimalarial activity of crude extracts from nine African medicinal plants. Journal of Ethnopharmacology 116: 74–83. Lekphrom R, Kanokmedhakul S, Kanokmedhakul K. 2010. Bioactive diterpenes from the aerial parts of Anisochilus harmandii. Planta Medica 76: 726–728. Li H-W, Hedge IC. 1994. Lamiaceae. In: Wu Z-Y, Raven PH. eds. Flora of China, Vol. 17. Beijing: Scientific Press and Missouri Botanical Garden, 50–299. Linder HP. 2014. The evolution of African plant diversity. Frontiers in Ecology and Evolution 2: 38. L u k h o b a C W, S i m m o n d s M S , Pa t o n A J. 2 0 0 6 . Plectranthus: a review of ethnobotanical uses. Journal of Ethnopharmacology 103: 1–24. Maddison WP, Maddison DR. 2015. Mesquite: a modular system for evolutionary analysis. Version 3.05. Available at: http://mesquiteproject.org. Maurin O, Davies TJ, Burrows JE, Daru BH, Yessoufou K, Muasya AM, van der Bank M, Bond WJ. 2014. Savanna fire and the origins of the ‘underground forests’ of Africa. New Phytologist 204: 201–214. Meve U, Liede S. 2002. Floristic exchange between mainland Africa and Madagascar: case studies in Apocynaceae– Asclepiadoideae. Journal of Biogeography 29: 865–873. Moteetee A, Van Wyk BE. 2011. The medical ethnobotany of Lesotho: a review. Bothalia 41: 209–228. Mwanyambo ML. 2008. Phylogeny and biogeography of Plectranthus L’Hérit (Ocimeae: Nepetoideae: Lamiaceae) with emphasis on taxa occurring on the Nyika Plateau Malawi. Unpublished PhD thesis, The University of Reading. Njau EFA, Alcorn JM, Buza J, Chirino-Trejo M, Ndakidemi P. 2014. Antimicrobial activity of Tetradenia © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 Bentham G. 1832. Coleus in Labiatarum genera et species. London: J. Ridgway and Sons, 47–59. Bouckaert R, Heled J, Kühnert D, Vaughan T, Wu C-H, Xie D, Suchard MA, Rambaut A, Drummond AJ. 2014. BEAST 2: a software platform for Bayesian evolutionary analysis. PLoS Computational Biology 10: e1003537. Buerki S, Devey DS, Callmander MW, Phillipson PB, Forest F. 2013. Spatio‐temporal history of the endemic genera of Madagascar. Botanical Journal of the Linnean Society 171: 304–329. Bytebier B, Antonelli A, Bellstedt DU, Linder HP. 2011. Estimating the age of fire in the Cape flora of South Africa from an orchid phylogeny. Proceedings of the Royal Society B: Biological Sciences 278: 188–195. Codd LE. 1975. Plectranthus (Labiatae) and allied genera in southern Africa. Bothalia 11: 371–442. Codd LE. 1985. Flora of southern Africa. 28, 4 Lamiaceae. Pretoria: Botanical Research Institute, Department of Agriculture and Walter Supply, Republic of South Africa. Cook FEM. 1995. Economic botany data collection standard. Kew: Royal Botanic Gardens. Cubey J. 2017. RHS plant finder 2017. London: Royal Horticultural Society. Dodsworth S. 2015. Genome skimming for next-generation biodiversity analysis. Trends in Plant Science 20: 525–527. Douwes E, Crouch NR, Edwards TJ, Mulholland DA. 2008. Regression analyses of southern African ethnomedicinal plants: informing the targeted selection of bioprospecting and pharmacological screening subjects. Journal of Ethnopharmacology 119: 356–364. Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin, Botanical Society of America 19: 11–15. Drew BL, Sytsma KJ. 2012. Phylogenetics, biogeography, and staminal evolution in the tribe Mentheae (Lamiaceae). American Journal of Botany 99: 933–953. Fawole OA, Amoo SO, Ndhlala AR, Light ME, Finnie JF, Van Staden J. 2010. Anti-inflammatory, anticholinesterase, antioxidant and phytochemical properties of medicinal plants used for pain-related ailments in South Africa. Journal of Ethnopharmacology 127: 235–241. Fer I, Tietjen B, Jeltsch F, Trauth MH. 2017. Modelling vegetation change during Late Cenozoic uplift of the East African plateaus. Palaeogeography, Palaeoclimatology, Palaeoecology 467:120–130. Forster PI. 1992. Five new species of Plectranthus L. Hérit (Lamiaceae) from Queensland. Austrobaileya 3: 729–740. Forster PI. 1994. Ten new species of Plectranthus L’Her. (Lamiaceae) from Queensland. Austrobaileya 4: 159–186. Forster PI. 2011. Five new species of Plectranthus L. Hér. (Lamiaceae) from New South Wales and Queensland. Austrobaileya 8: 387–404. Fowler DG. 2006. Traditional fever remedies: a list of Zambian plants. Available at: http://citeseerx.ist.psu.edu/viewdoc/dow nload?doi=10.1.1.527.9450&rep=rep1&type=pdf. Accessed 16 Dec. 2016. Gaudeul M, Rouhan G. 2013. A plea for modern botanical collections to include DNA-friendly material. Trends in Plant Science 18: 184–185. 369 370 A. PATON ET AL. Rønsted N, Symonds MR, Birkholm T, Christensen SB, Meerow AW, Molander M, Mølgaard P, Petersen G, Rasmussen N, Van Staden J, Stafford GI. 2012. Can phylogeny predict chemical diversity and potential medicinal activity of plants? A case study of Amaryllidaceae. BMC Evolutionary Biology 12: 182. Saslis-Lagoudakis CH, Williamson EM, Savolainen V, Hawkins JA. 2011a. Cross-cultural comparison of three medicinal floras and implications for bioprospecting strategies. Journal of Ethnopharmacology 13: 476–487. Saslis-Lagoudakis CH, Klitgaard BB, Forest F, Francis L, Savolainen V, Williamson EM, Hawkins JA. 2011b. The use of phylogeny to interpret cross-cultural patterns in plant use and guide medicinal plant discovery: an example from Pterocarpus (Leguminosae). PLoS One 6: e22275. Shaw J, Lickey EB, Beck JT, Farmer SB, Liu W, Miller J, Siripun KC, Winder CT, Schilling EE, Small RL. 2005. The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. American Journal of Botany 92: 142–166. Shepherd C, Maybry J. 2016. Solenostemon plant named ‘Monkey Island’. U.S. Patent PP27,179. Smitha K, Sunojkumar P. 2015. Notes on the identity and distribution of Plectranthus subincisus (Lamiaceae)—a poorly known species recollected after 150 years in southern India. Phytotaxa: 192: 105–111. Stafford GI, Pedersen ME, van Staden J, Jäger AK. 2008. Review on plants with CNS-effects used in traditional South African medicine against mental diseases. Journal of Ethnopharmacology: 119: 513–537. Suddee S, Paton AJ, Parnell JAN. 2005. A taxonomic revision of tribe Ocimeae Dumort.(Lamiaceae) in continental South East Asia II. Plectranthinae. Kew Bulletin 60: 3–75. Suddee S, Nanthawan Suphuntee N, Sommanussa Saengrit S. 2014. Plectranthus phulangkaensis (Lamiaceae) a new species from Thailand. Thai Forest Bulletin (Bot.) 42: 6–9. Taberlet P, Gielly L, Pautou G, Bouvet J. 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17:1105–9. Thiers B. 2018. Index herbariorum: a global directory of public herbaria and associated staff. New York Botanical Garden’s Virtual Herbarium. Available at: http://sweetgum.nybg.org/ih/ Vanaja M, Annadurai G. 2013. Coleus aromaticus leaf extract mediated synthesis of silver nanoparticles and its bactericidal activity. Applied Nanoscience 3: 217–223. Waruruai J, Sipana B, Koch M, Barrows LR, Matainaho TK, Rai PP. 2011. An ethnobotanical survey of medicinal plants used in the Siwai and Buin districts of the Autonomous Region of Bougainville. Journal of Ethnopharmacology 138: 564–577. Zhong JS, Li J, Li L, Conran JG, Li HW. 2010. Phylogeny of Isodon (Schrad. ex Benth.) Spach (Lamiaceae) and related genera inferred from nuclear ribosomal ITS, trnLtrnF region, and rps16 intron sequences and morphology. Systematic Botany 35: 207–219. © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 riparia (Hochst.) Lamiaceae, a medicinal plant from Tanzania. European Journal of Medicinal Plants 4: 1462. Nylander JAA. 2004. MrModeltest v2. Program distributed by the author. Uppsala: Evolutionary Biology Centre, Uppsala University. Oliveras I, Malhi Y. 2016. Many shades of green: the dynamic tropical forest–savannah transition zones. Philosophical Transactions of the Royal Society B 371: 20150308. Pagel M, Meade A, Barker D. 2004. Bayesian estimation of ancestral character states on phylogenies. Systematic Biology 53: 673–684. Paton AJ, Springate D, Suddee S, Otieno D, Grayer RJ, Harley MM, Willis F, Simmonds MS, Powell MP, Savolainen V. 2004. Phylogeny and evolution of basils and allies (Ocimeae, Labiatae) based on three plastid DNA regions. Molecular Phylogenetics and Evolution 31: 277–299. Paton AJ, Bramley G, Ryding O, Polhill RM, Harvey YB, Iwarsson M, Willis F, Phillipson PB, Balkwill K, Lukhoba CW, Oteino D, Harley RM. 2009. Lamiaceae (Labiatae). In: Beentje HJ, Ghazanfar SA, Polhill RM, eds. Flora of Tropical East Africa. Lamiaceae (Labiatae): 1–413. Kew: Royal Botanic Gardens. Paton AJ, Bramley G, Ryding O, Polhill RM, Harvey YB, Iwarsson M, Willis F, Phillipson PB, Balkwill K, Oteino D, Harley RM. 2013. Lamiaceae. In: Timberlake J, ed. Flora Zambesiaca, 8,8: 1–331. Kew: Royal Botanic Gardens. Pastore JFB, Harley RM, Forest F, Paton A. van den Berg C. 2011. Phylogeny of the subtribe Hyptidinae (Lamiaceae tribe Ocimeae) as inferred from nuclear and plastid DNA. Taxon 60: 1317–1329. Pokorny L, Riina R, Mairal M, Meseguer AS, Culshaw V, Cendoya J, Serrano M, Carbajal R, Ortiz S, Heuertz M, Sanmartín I. 2015. Living on the edge: timing of Rand flora disjunctions congruent with ongoing aridification in Africa. Frontiers in Genetics 6: 154. Pollard BJ, Paton A. 2009. The African Plectranthus (Lamiaceae) expansion continues. Vale Leocus! Kew Bulletin 64: 259–261. Phillipson PB, Steyn CF. 2008. Tetradenia (Lamiaceae) in Africa: new species and new combinations. Adansonia 30: 177–196. Rambaut A, Suchard MA, Xie D, Drummond AJ. 2014. Tracer v1.6. Available at: http://beast.bio.ed.ac.uk/Tracer Rice LJ, Brits GJ, Potgieter CJ, Van Staden J. 2011. Plectranthus: a plant for the future? South African Journal of Botany 77: 947–959. Ronquist F, Teslenko M, Van Der Mark P, Ayres DL, Darling A, Höhna S, Larget B, Liang L, Suchard MA, Huelsenbeck JP. 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biology 61: 539–542. Rønsted N, Savolainen V, Mølgaard P, Jäger AK. 2008. Phylogenetic selection of Narcissus species for drug discovery. Biochemical Systematics and Ecology 36: 417–422. Appendix 1. Species, voucher specimens and GenBank numbers of materials used in analysis. Sequences marked are being deposited in GenBank Country Collector and Number Herbarium trnL-trnF intron + 3′ exon + spacer rps16 intron trnS-trnG spacer Aeollanthus buchnerianus Briq. Malawi K AJ505434 AJ505327 MH612699 Aeollanthus densiflorus Ryding Kenya K AJ505435 AJ505328 MH612700 Aeollanthus rhemanii Gürke Alvesia clerodendroides (T.C.E.Fr.) Mathew Alvesia rosmarinifolia Welw. Anisochilus harmandii Doan Anisochilus pallidus Benth. Callicarpa americana L. Callicarpa japonica Thunb. Capitanopsis albida (Baker) Hedge Capitanopsis angustifolia (Moldenke) Capuron Capitanopsis cloiselii S.Moore Clinopodium myrianthum (Baker) Ryding Clinopodium vulgare L. subsp. arundanum (Boiss.) Nyman Congea tomentosa Roxb. Dauphinea brevilabra Hedge Malawi Tanzania Brummitt 10401 (cultivated Kew 1970–2734) Mathew 6137 (cultivated Kew 1970–3760) M.L.Mwanyambo et al. 746 Sally Bidgood et al. 4547 MAL K MH612625 MH612626 MH630063 MH630096 MH612701 MH612741 Zambia Thailand Thailand Cultivated Cultivated Madagascar Madagascar Harder et al. 3634 S. Suddee et al. 775 S. Suddee et al. 1080 Cult., Kew 081-84-00507 Cult., Kew 1934–12904 P. Lowry 6255 Clement et al. 2117 K BKF,K,TCD BKF,K,TCD K K P, MO. Silica K K AJ505436 AJ505437 AJ505438 AJ505535 AJ505536 MH612627 AJ505440 AJ505329 AJ505330 AJ505331 AJ505412 AJ505413 MH630093 AJ505333 MH612742 MH612743 MH612744 X X MH612736 MH612737 Madagascar Malawi R.Capuron 20.490-SF M.L.Mwanyambo et al. 771 K MAL MH612628 MH612629 MH630094 MH630062 MH612738 MH612698 Cultivated Cult., Kew 453-79-04649 K AJ505547 AJ505426 X Cultivated Madagascar BHO K AJ505530 AJ505441 AJ505411 AJ505334 X MH612739 Elsholtzia stauntonii Benth. Gmelina philippensis Cham. (G. hystrix Schult. ex Kurz) Hyptis capitata Jacq. Isodon lophanthoides (Buch.-Ham. ex D.Don) Kudô Lavandula maroccana Murb. Lavandula minutolii Bolle Melissa officinalis L. Mentha suaveolens Ledeb. Nepeta racemosa Lam. Ocimum filamentosum Forssk. Cultivated Cultivated Wagstaff, s.n. Hardy & Rauh 2876 (cultivated Kew 1998–2417) Wagstaff 356 Cult., Kew 381-74-02999 BHO K AJ505526 AJ505527 AJ505406 AJ505407 X X Wongprasert et al. s.n. H.J.M. Bowen 3993 Upson s.n. Upson s.n. Wagstaff 88-09 Cult., Kew 1970–3169 Z.Jamzad s.n. Brummitt 18993 BKF,K,TCD RNG AJ505449 MH612630 AJ505337 MH630131 MH612795 MH612796 RNG RNG BHO K TARI K AJ505461 AJ505462 AJ505529 AJ505541 AJ505432 AJ505466 AJ505347 AJ505348 AJ505410 AJ505418 AJ505325 AJ505352 X X X X X MH612793 Thailand Malaysia Cultivated Cultivated Cultivated Cultivated Iran Kenya 371 Taxon PLECTRANTHUS, COLEUS AND ALLIES Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 APPENDICES 372 Appendix 1. Continued Country Collector and Number Herbarium trnL-trnF intron + 3′ exon + spacer rps16 intron trnS-trnG spacer Ocimum tenuiflorum L. Origanum vulgare L. Thailand Cultivated K K AJ505473 AJ505543 AJ505358 AJ505422 X X Orthosiphon rubicundus (D.Don) Benth. Orthosiphon schimperi Benth. Plectranthus acaulis Brummitt & Seyani Plectranthus adenophorus Gürke Plectranthus africanus (Scott-Elliot) A.J.Paton Thailand Suddee 893 Chase 13334 (Cult., Kew 000-69-19317) Suddee 809 K AJ505477 AJ505360 X Malawi Zambia M.L.Mwanyambo et al. 769 M.L.Mwanyambo et al. 742 MAL MAL MH612631 MH612633 MH630130 MH630064 MH612794 MH612702 Tanzania Democratic Rebublic Congo Kenya K.&T.Pocs 87063 Masens da Musa Y. 632 K K MH612634 MH612635 MH630105 MH630103 MH612763 MH612758 Mrs. S.F. Polhill 327 K MH612636 MH630067 MH612705 Thailand Malawi Kenya Suddee et al. 868 M.L.Mwanyambo et al. 762 Lukhoba et al. 501 BKF, K, TCD MAL K AJ505498 MH612637 MH612638 AJ505376 MH630088 MH630115 MH612746 MH612727 MH612773 Thailand Suddee et al. 869 BKF AJ505499 AJ505377 X Tanzania Cultivated Cultivated Tanzania Madagascar Cultivated K RNG K K K K MH612639 MH612640 MH612641 MH612642 MH612643 AJ505501 MH630068 MH630117 MH630122 MH630104 MH630095 AJ505379 MH612706 MH612775 MH612782 MH612759 MH612740 MH612780 S. Africa Tanzania S.Bidgood et al. 1918 T.T.Aye s.n. Cult., Kew 1999-14 Sally Bidgood et al. 3413 Cult., Kew 1988–3186 Chase 8514 (Cultivated K-1970–3559) Balkwill et al. 10880 Sally Bidgood et al. 3335 J, K K AJ505502 MH612644 AJ505380 MH630070 MH612760 MH612708 Zimbabwe T.Müller 3592 K MH612645 MH630087 MH612726 Cultivated Chase 13336 (Cultivated K-1991–6) Cult., Kew 1970–3233 P.I.Forster PIF15180 Brummitt 9700 (Cult., 1991–6) K AJ505532 AJ505409 MH612730 K K K MH612646 MH612647 AJ505504 MH630098 MH630120 AJ505382 MH612748 MH612778 MH612761 Plectranthus agnewii Lukhoba & A.J.Paton Plectranthus albicalyx S.Suddee Plectranthus alboviolaceus Gürke Plectranthus alpinus (Vatke) O.Ryding Plectranthus amboinicus (Lour.) Spreng. Plectranthus annuus A.J.Paton Plectranthus argentatus Blake Plectranthus barbatus Andr. Plectranthus betonicifolius Baker Plectranthus bipinnatus A.J.Paton Plectranthus buchananii Baker Plectranthus calycinus Benth. Plectranthus candelabriformis Launert Plectranthus chimanimaniensis S.Moore Plectranthus ciliatus E.Mey Plectranthus coeruleus (Gürke) Agnew Malawi Plectranthus congestus R.Br. Australia Plectranthus crassus N.E.Br. Malawi A. PATON ET AL. Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Taxon Country Collector and Number Herbarium trnL-trnF intron + 3′ exon + spacer rps16 intron trnS-trnG spacer Plectranthus daviesii (E.A.Bruce) Mathew Plectranthus dissectus Brenan Plectranthus diversus S.T.Blake Plectranthus djalonensis (A.Chev) A.J.Paton Plectranthus ecklonii Benth. Plectranthus elegans Britten Plectranthus emirnensis (Baker) Hedge Plectranthus esculentus N.E.Br. Malawi M.L.Mwanyambo et al. 792 MAL MH612648 MH630099 MH612749 Malawi Australia Zambia J.D.&E.G.Chapman 7195 J.R.Clarkson&V.J.Neldner 10114 G.Pope, A-R Smith & D.Goyder 2120 T.T.Aye s.n. J.D.Chapman 6065 R.A.Clement et al. 2057 K K K MH612650 MH612651 MH612652 MH630078 MH630118 MH630107 MH612716 MH612776 MH612765 RNG K K X MH612653 MH612654 MH918091 MH630089 MH630090 X MH612728 MH612729 MAL MH612655 MH630109 MH612767 Plectranthus glabratus (Benth.) Alston Plectranthus glandulosus Hook.f. Plectranthus goetzii Gürke Plectranthus gracilis Suesseng. Plectranthus gracillimus (T.C.E.Fr.) Hutch. & Dandy Plectranthus guerkei Briq. Plectranthus hadiensis (Forssk.) Sprenger Plectranthus hockii De Wild. Plectranthus ignotus A.J.Paton Plectranthus lactiflorus (Vatke) Agnew Plectranthus lanuginosus (Benth.) Agnew Plectranthus lasianthus (Gürke) Vollesen Plectranthus laxiflorus Benth. Plectranthus leptophyllus (Baker) A.J.Paton Plectranthus longipes Baker Plectranthus masukensis Baker Plectranthus melleri Baker Plectranthus modestus Baker India M.L.Mwanyambo et al. 707 J.Klackenberg & R.Lundin 176 K MH612657 MH630097 MH612747 Nigeria Malawi Malawi Tanzania J.D.Chapman 4015 M.L.Mwanyambo et al. 765 E.Phillips 3636 S.Bidgood & K.Vollesen 3229 K MAL K K MH612658 MH612659 MH612660 MH612661 MH630084 MH630076 MH630075 MH630108 MH612722 MH612714 MH612713 MH612766 Tanzania Ethiopia L.Festo 723 M.G.Gilbert et al. 9265 K K MH612662 MH612663 MH630101 MH630121 MH612752 MH612779 Tanzania Tanzania Ethiopia Goyder et al 3904 S.Bidgood et al. 2544 I.Friis et al. 8874 K K K AJ505443 MH612664 MH612665 AJ505335 MH630110 X MH612753 MH612768 X Tanzania J.M.Grimshaw 9341 K MH612666 MH630124 MH612785 Botswana P.A.Smith 2090 K MH612667 MH630123 MH612784 Ethiopia Tanzania M.G.Gilbert et al. 9245 S.Bidgood et al. 4207 K K MH612668 MH612669 MH630083 MH630102 MH612721 MH612754 Tanzania Tanzania Cameroon Zambia O.Hedberg 6825 S.Bidgood et al. 792 R.Letouzey 14998 M.L.Mwanyambo et al. 752 K K K MAL MH612670 MH612671 X MH612672 MH630077 MH630071 MH630113 MH630106 MH612715 MH612709 MH612771 MH612764 Cultivated Malawi Madagascar Malawi 373 Taxon PLECTRANTHUS, COLEUS AND ALLIES Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Appendix 1. Continued 374 Appendix 1. Continued Country Collector and Number Herbarium trnL-trnF intron + 3′ exon + spacer rps16 intron trnS-trnG spacer Plectranthus mollis (Aiton) Spreng. Plectranthus montanus Benth. Plectranthus oertendahlii T.C.E.Fr. Plectranthus parishii Prain Plectranthus parviflorus Willd. Plectranthus parvus Oliv. India Cultivated Cultivated Thailand Cultivated Kenya K K K BKF, K, TCD RNG K MH612673 AJ505538 AJ505534 AJ505511 MH612674 MH612675 MH630072 AJ505383 MH630091 AJ505390 MH630119 MH630081 MH612710 MH612781 MH612731 MH612745 MH612777 MH612719 Plectranthus pauciflorus Baker Plectranthus petiolaris E.Mey. ex Benth. Plectranthus pinetorum A.J.Paton Plectranthus puberulentus K.Morton Plectranthus pubescens Baker Plectranthus punctatus (L.f.) L’Hér. subsp. edulis (Vatke) A.J.Paton Plectranthus rungwensis A.J.Paton Plectranthus sallyae A.J.Paton Plectranthus sanguineus Britten Plectranthus schizophyllus Baker Plectranthus scutellaroides (L.) R.Br. Plectranthus shirensis (Gürke) A.J. Paton Plectranthus sigmoideus A.J.Paton Plectranthus stenophyllus Baker Plectranthus stenosiphon Baker Plectranthus stolzii Gilli Plectranthus sylvestris Gürke Plectranthus tenuicaulis (Hook.f.) J.K.Morton Plectranthus termiticola A.J.Paton Tanzania S. Africa J.Klackenberg & R.Lundin 260 Chase 8518 (Cult., 1996-1453) Chase 3380 (Cult., 1969–5789) Suddee 1144 T.T.Aye s.n. M.G.Gilbert & Mesfin Tadessa 6713 Mrs. M. Richards 22953 Univ. of Natal K-1996–2729 K K MH918093 AJ505512 MH630085 AJ505391 MH612723 MH612725 Malawi Kenya Malawi Kenya D.J.Goyder & A.J.Paton 3660 Mathew 6830 (Cult., 1970–3784) M.L.Mwanyambo et al. 778 Lukhoba et al. 505 K K voucher, MAL K MH918094 AJ505507 MH612676 MH612677 MH630080 AJ505386 MH630074 MH630116 MH612718 MH612783 MH612712 MH612774 Tanzania Tanzania Malawi Malawi Trinidad Tanzania L.B.Mwasumbi 16222 S.Bidgood et al. 2661 Brummitt s.n. (Cult., 1970–2072) M.L.Mwanyambo et al. 796 Barnard et al. 193 S.Bidgood et al. 4725 K K K MAL RNG K MH612678 MH612679 AJ505513 MH612680 MH612681 MH612682 MH630086 MH630114 AJ505392 MH630125 MH630127 MH630128 MH612724 MH612772 MH612762 MH612786 MH612788 MH612789 Malawi Tanzania Malawi Tanzania Malawi Angola E.G.Chapman 264 S.Bidgood et al. 1951 R.K.Brummitt et al. 16050 M.Richards 9880 R.K.Brummitt 12332 C.Henriques 323 BM K MAL K K K MH612683 MH612632 MH612684 MH612685 MH612687 MH612688 MH630126 MH630066 MH630079 MH630082 MH630112 MH630129 MH612787 MH612704 MH612717 MH612720 MH612770 MH612790 Zimbambwe Heany Teachers Training College H.69 Chase 13332 (Cult., 563-87-04012) J.A.Mlangwa 631 R.I.Ludanga 1301 K MH612689 MH630073 MH612711 K AJ505533 AJ505405 MH612750 K K MH612690 MH612691 MH630100 MH630069 MH612751 MH612707 Plectranthus thyrsoideus (Baker) B.Matthew Plectranthus triangularis A.J.Paton Plectranthus vesicularis A.J.Paton Tanzania Tanzania A. PATON ET AL. Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Taxon Taxon Country Collector and Number Herbarium trnL-trnF intron + 3′ exon + spacer rps16 intron trnS-trnG spacer Plectranthus viphyensis Brummitt & Seyani subsp. zebrarum (Brummitt & Seyani) A.J.Paton Plectranthus welwitschii (Briq.) Codd Plectranthus xerophilus Codd Plectranthus zombensis Baker Prostanthera petrophila B.J.Conn Prosthanthera nivea Benth. Pycnostachys reticulata (E.Mey.) Benth. Pycnostachys umbrosa Perkins Pycnostachys urticifolia Hook. Rosmarinus officinalis L. Salvia nilotica Juss. ex Jacq Tectona grandis L.f. Tetradenia fruticosa Benth. Tetradenia nervosa Codd Tetradenia tanganyikae Phillipson Malawi M.L.Mwanyambo & E.S.Kathumba 797 MAL MH612692 MH630065 MH612703 Cultivated S.Africa Malawi Cultivated Cultivated S. Africa Cult., 1999-15; Cult., 1989-1322; Hardy 3966 J.L. Balaka & K. Kaunda 378 Chase 6980 Chase 6975 Cult., Kew 1999–2425 K K K K K AJ505505 AJ505515 MH612693 AJ505525 AJ505524 AJ505516 AJ505384 AJ505394 MH630111 AJ505404 AJ505403 AJ505395 MH612791 MH612792 MH612769 X X MH612755 Kenya S. Africa Cultivated Malawi Cultivated Madagascar Madagascar Malawi K K K MAL BHO K K MAL AJ505517 AJ505518 AJ505546 MH612695 AJ505528 AJ505519 AJ505520 MH612696 AJ505396 AJ505397 AJ505425 MH918092 AJ505408 AJ505398 AJ505399 MH630092 MH612757 MH612756 X MH612697 X MH612732 MH612733 MH612734 S. Africa Cultivated Mathew 6067 (Cult., 1970–3755) Cult., Kew 1999–2426 Cult., Kew 1973 14217 M.L.Mwanyambo et al. 728 Waimea 73P172 Hardy 2910A (Cult., 1989-1324) Hardy 2910B (Cult., 1993–3116) M.L.Mwanyambo & E.S. Kathumba s.n. L.McDade LM1281 Chase 13331 (Cult., 1975-1177) J K AJ505521 AJ505544 AJ505401 AJ505423 MH612735 X Cultivated Chase 8757 (TCMK 15) K AJ505539 AJ505416 X Thorncroftia longifolia N.E.Br. Thymus serpyllum L. var. citriodorum (Pers.) Becker Vitex trifolia L. PLECTRANTHUS, COLEUS AND ALLIES 375 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Appendix 1. Continued 376 A. PATON ET AL. Appendix 2. Medicinal use of sampled species not recorded in previous review of the medicinal uses of Plectranthus (Lukhoba et al., 2006) Use Category (Cook, 2005) Reference Aeollanthus buchnerianus Aeollanthus densiflorus Anisochilus harmandii infection/fever skin infection/fever Moteetee & Van Wyk (2011). Awas & Demissew (2009). Lekphrom, Kanokmedhakul, Kanokmedhakul (2010). Plectranthus argentatus cure for colds in children eye and skin diseases tonic – antimalarial and antimycobacterial properties stomach pain, inflammation Anthoney & Ngule (2013) Plectranthus scutellarioides Plectranthus scutellarioides headache, coughs malaria, diarrhoea Plectranthus shirensis Plectranthus sylvestris Pycnostachys reticulata Pycnostachys urticifolia Pycnostachys urticifolia Tetradenia riparia traditional fever remedy anti-inflamatory pain-related ailments mental diseases antibacterial antimicrobial gastro-urinary, infection fever, skin respiratory, pain infection/fever, gastro-urinary, infection/fever infection/fever pain other mental infection/fever infection/fever Waruruai et al. (2011) Kaou et al. (2008) Fowler (2006) Juch & Rüedi (1997). Fawole et al. (2010) Stafford et al. (2008) Bascombe & Gibbons (2008) Njau et al. (2014) SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher's web-site: Table S1. Amplification of plastid DNA markers: primers used and thermocycling conditions. Figure S1. Dated analyses produced by BEAST. Bars indicate 95% highest posterior density interval (HPD). Numbers give mean ages for crown nodes of named clades. P – Plectranthus clade comprising Alvesia, Aeollanthus and Plectranthus clades I–IV and C – Coleus clade comprising Coleus clades A and B. Figure S2. Consensus tree with geographical distribution optimized: sub-Saharan Africa (blue); Madagascar (green); Asia (yellow) and Australia (black). Grey indicates proportion of trees in which optimization is unresolved. Red indicates proportion in which node is absent. P – Plectranthus clade comprising Alvesia, Aeollanthus and Plectranthus clades I–IV and C – Coleus clade comprising Coleus clades A and B. Figure S3. Consensus tree with habitat optimized: dry woodland (green); evergreen forest margins (blue); montane grassland (black); seasonally flooded grassland (white). Grey indicates proportion of trees in which optimization is unresolved. Red indicates proportion in which node is absent. P – Plectranthus clade comprising Alvesia, Aeollanthus and Plectranthus clades I–IV and C – Coleus clade comprising Coleus clades A and B. Figure S4. Consensus tree with medicinal use optimized: medicinal use recorded (black); medicinal use not recorded (white). Grey indicates proportion of trees in which optimization is unresolved. Red indicates proportion in which node is absent. P – Plectranthus clade comprising Alvesia, Aeollanthus and Plectranthus clades I–IV and C – Coleus clade comprising Coleus clades A and B. © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society, 2018, 188, 355–376 Downloaded from https://academic.oup.com/botlinnean/article/188/4/355/5139384 by guest on 16 December 2021 Species