The demise of Monechma: new combinations and a new classification in the resurrected genera Meiosperma and Pogonospermum (Acanthaceae)

Recent molecular phylogenetic results have demonstrated that Monechma s.l., a group of plants with ecological importance in the savanna and succulent biomes of sub-Saharan Africa, is polyphyletic with two discrete lineages recognisable. In the present work, we recognise Monechma Groups I and II at the generic rank, which can be distinguished by differences in inflorescence characteristics and seed morphology. The nomenclatural implications of these findings are investigated. The lectotype of Monechma, M. bracteatum Hochst., is a part of a small lineage of plants closely allied to Justicia L. sect. Harnieria (Solms) Benth. for which the earliest valid name is found to be Meiosperma Raf. Hence, Monechma is synonymised within Meiosperma, which comprises six accepted species and two undescribed taxa. The majority of species of former Monechma s.l. are resolved within the second lineage for which the only validly published generic name is Pogonospermum Hochst. This resurrected genus comprises 34 accepted species plus two undescribed taxa. Pogonospermum displays considerable morphological variation and is here subdivided into six sections based primarily on differences in plant habit, inflorescence form, calyx, bract and bracteole venation, and seed indumentum. The new combinations and new sections are validated, and seven accepted species names are lectotypified.


Introduction
With at least 2000 species, the Justicieae (Acanthaceae: Acanthoideae) are the largest tribe of Acanthaceae, and are the source of numerous taxonomic challenges due to complex morphological variation including frequent homoplasy (Daniel et al. 2008;Kiel et al. 2017Kiel et al. , 2018McDade et al. 2018McDade et al. , 2021Darbyshire et al. 2019a;Manzitto-Tripp et al. 2021). Particularly problematic is generic delimitation of Justicia L. and allies in the subtribe Justiciinae (sensu Manzitto- , where molecular phylogenetic data demonstrate that several morphologically discrete genera are nested within Justicia s.l. (Kiel et al. 2017(Kiel et al. , 2018. To maintain a broadly circumscribed Justicia as defined in most modern taxonomic treatments, the entire Justiciinae would have to be treated under Justicia s.l., but this is highly undesirable as it would result in the loss of much valuable taxonomic information. The most favourable alternative option is to disaggregate Justicia and to recognise discrete, wellsupported and well-sampled clades as separate genera, backed by knowledge of taxonomically informative morphological traits (Darbyshire et al. 2019a(Darbyshire et al. , 2020. One such genus nested within the Justiciinae is Monechma Hochst. (or Justicia sect. Monechma (Hochst.) T.Anderson), a striking radiation of African plants. Species of Monechma range throughout sub-Saharan Africa, but are most diverse and abundant in the savannas, deserts, and shrublands of southwestern Africa, especially portions of Namibia and southern Angola (Munday 1983;Darbyshire et al. 2020). In this region, plants of Monechma comprise some of the most dominant elements of landscapes, together with species in three other parallel radiations of Acanthaceae, Petalidium Nees, Barleria L., and Blepharis Juss. (Vollesen 2000;Tripp et al. 2017;Darbyshire et al. 2019bDarbyshire et al. , 2020Darbyshire et al. , 2021. This ecological significance has motivated recent interest in Monechma, where a RADseq phylogenetic analysis which included over 75% of the currently accepted species, resolved relationships amongst this important group (Darbyshire et al. 2020).
Phylogenetic results presented in Darbyshire et al. (2020) confirmed earlier results based on evidence from six molecular markers (Kiel et al. 2017) that Monechma s.l. in fact comprises two widely separated lineages, named Monechma Groups I and II. The RADseq study, however, refined delimitation of these two clades from the earlier study by Kiel et al. (2017). Specifically, it revealed that the group of perennial (rarely annual) herbaceous taxa from the savanna biome of tropical Africa, with usually terminal spiciform inflorescences and bracts ± highly modified from the leaves (M. ciliatum (Jacq.) Milne-Redh., M. scabridum S.Moore and allies) is allied to the group of mainly perennial shrublets with ± undifferentiated bracts (M. cleomoides (S.Moore) C.B.Clarke and relatives) primarily from the succulent biome of southern Africa (Darbyshire et al. 2020). This tropical savanna group was not sampled in the earlier study of Kiel et al. (2017) but based on morphological extrapolation, both that study and Darbyshire & Goyder (2019) later surmised that the tropical savanna group was allied to species of the M. debile (Forssk.) Nees complex, which are mostly annual, ruderal species with spiciform inflorescences and highly modified bracts. However, the RADseq analysis demonstrated that M. debile and allies form their own discrete clade, closely allied to Justicia sect. Harnieria (Solms) Benth. Study of a range of morphological traits revealed that the two clades of Monechma under their revised circumscriptions could be separated by differences in inflorescence and seed characters (Darbyshire et al. 2020). It was therefore concluded that these two clades should be elevated to distinct genera, but Darbyshire et al. (2020) refrained from nomenclatural changes.
In the present work, we recognise Monechma Groups I and II at the generic rank as Meiosperma and Pogonospermum respectively, and provide the appropriate new combinations following a nomenclatural review. We also build on the morphological studies of Darbyshire et al. (2020) to propose a formal sectional classification for Monechma Group II (Pogonospermum). This is an early step in the process of redefining the genera within Justiciinae. Moving forward, further detailed studies employing NGS techniques comparable to that presented in Darbyshire et al. (2020), are required across Justiciinae in order to provide a revised classification. To this end, a RADseq analysis of the New World Justiciinae is currently in progress (C. A. Kiel et al. in prep.).

Materials and Methods
The present contribution consists of two parts. The nomenclatural study is based on a review of relevant literature on Monechma and allied genera, including protologues and relevant Flora accounts. The taxonomic study is based on examination of herbarium collections and images of species in situ, in combination with the phylogenetic results of the RADseq dataset of Darbyshire et al. (2020). Most observations were made on herbarium specimens held at COLO, K, RSA and WIND (herbarium abbreviations follow Thiers, continuously updated), with additional observations made via access to digital images of type specimens on JSTOR Global Plants (https:// plants.jstor.org/) and other online repositories of herbarium specimen images.
For species not sampled in Darbyshire et al. (2020), best estimations for taxonomic placement were made based on morphological traits. Type specimens are cited for the accepted names; those seen by one or more of the authors are marked with an "!"; those seen as digital images are marked with an "*".
Species delimitation follows a combination of Hedrén (2006aHedrén ( , 2006b, Munday (1995), Vollesen (2010Vollesen ( , 2015 and Darbyshire & Goyder (2019), with some modifications based on the current study. For each accepted species, the geographic range is derived from herbarium collections, online databases (notably Plants of the World Online, POWO 2020) and relevant taxonomic literature, principally those cited above.
In the section on nomenclatural renovations, a "^" denotes that the species was sampled in the RADseq study (Darbyshire et al. 2020).

Results and Discussion
Applying the correct generic names to Monechma Group I and Monechma Group II Monechma was first described by Hochstetter (1841: 374) who included two species, both based on collections from the expedition to Nubia by Theodor Kotschy: M. bracteatum Hochst. (Kotschy it. nubico 261) and M. hispidum Hochst. (Kotschy it. nubico 239). Later, having seen further cultivated material of Monechma from seeds grown on from Kotschy's collections, Hochstetter (1844: 5) chose to separate Monechma into two genera and described Pogonospermum Hochst. based on his M. hispidum (= P. hispidum (Hochst.) Hochst.) plus a second species, P. ciliare that was based on Justicia ciliaris, which Hochstetter wrongly attributed to Vahl, having overlooked Linnaeus' (1781) earlier application of this name. Hochstetter also erred in using the epithet "ciliare" given that he also included the earlier name J. ciliata Jacq. in synonymy. Hochstetter cited differences in seed characteristics as justifying the recognition of two genera, i.e., Monechma sensu stricto has glabrous seeds, whilst the seeds of Pogonospermum have tufts of trichomes at the base and apex. By splitting Pogonospermum from Monechma, Hochstetter indicated that Monechma was now to be based on M. bracteatum and, indeed, Monechma was subsequently lectotypified with M. bracteatum by Phillips (1951).
In describing Monechma and Pogonospermum, however, Hochstetter had overlooked two earlier generic names of relevance to this group that require consideration here. Firstly, Rafinesque (1838: 64) described Meiosperma based on Dianthera debilis Forssk.
Rafinesque noted that his new genus differed from Dianthera L. (= Justicia in current classifications; with Justicia sect. Dianthera (L.) V.A.W.Graham being exclusively New World as currently circumscribed; Graham 1988) in having single-seeded capsules, which was clearly a mistake as D. debilis is 2-seeded. He went on to note "…thus not even of this family, are the cells monosperm? The cor[olla] is bilabiate but undescribed. Type D. debilis Forsk. Vitm." (p. 64). This suggests that Rafinesque did not see the material in the Forsskal herbarium on which D. debilis was based, rather he was reliant on Vitman's ("Vitm.") description of this species in his Summa Plantarum (Vitman 1789: 4 6 ) , w h o r e c o r d e d t h i s s p e c i e s a s b e i n g monospermous. Vitman's record was, in turn, based upon Forsskål's (1775) description of D. debilis in which he stated in error "capsula bilocularis, monosperma" (p. 9). It is perhaps on the basis of this perpetuated error in recording the number of seeds per capsule that Hochstetter overlooked Meiosperma when describing Monechma. However, an alternative explanation may relate to Rafinesque's work being widely rejected in the biological community at that time (Boewe 2005) and so it may have been dismissed by Hochstetter. Wood et al. (1983) note that there are two specimens in the Forsskal herbarium with the name Dianthera debilis: microfiche 38: II. 3-4 and 38.II.5-6, and that they represent different species. Forsskal 38: II.3-4 is an informative specimen, with an open corolla and mature fruits and seed. This is clearly the specimen from which the name D. debilis has been applied in subsequent works (e.g. Nees 1847; Clarke 1900) and was selected as lectotype by Wood et al. (1983). Forsskal 38.II.5-6, although superficially similar in inflorescence form, is actually referable to Ecbolium violaceum (Vahl) Hillcoat & J.R.I.Wood. Irrespective of these complications, Meiosperma was validly published with D. debilis as the type species. Rafinesque (1838) did not make the new combination for this species in Meiosperma, but this is acceptable under the Botanical Code (Turland et al. 2018).
Secondly, Endlicher (in Endlicher & Fenzl 1839: 81) described Schwabea, with a single species, S. modesta Endl., based on a Kotschy collection from tropical East Africa (probably Sudan) that was grown in cultivation in Charles von Hügel's botanic garden in Hietzing, Vienna. Unfortunately, the Kotschy specimen in question was not numbered, nor was the precise collecting locality cited in the protologue, hence a corresponding herbarium collection is difficult to trace with certainty. Endlicher described this plant as having four didynamous stamens, which would not fit the genera allied to Justicia. However, in de Candolle's Prodromus, Nees (1847) stated that he considered Endlicher to have been mistaken in his recording of the number of stamens and he equated Schwabea with Hochstetter's Pogonospermum, reducing the latter to synonymy. Nees proposed the new combination Schwabea ciliaris (Vahl) Nees, with S. modesta, P. ciliare and P. hispidum all listed in synonymy. He also described a second species, S. spicigera Nees, with two varieties, both based on Heudelot specimens from "Senegambia". In Flora of Tropical Africa, Clarke (1900) disagreed with Nees' interpretation of Schwabea and excluded it from his "Eu-Justicieae" based on it having four stamens. Clarke reverted to Hochstetter's original (1841) circumscription of Monechma, reuniting it with Pogonospermum and with Schwabea sensu Nees (which, confusingly, he used as the name for a proposed section of Monechma). Clarke's view has since prevailed for over a century, except that some authors (e.g. Hedrén 1990Hedrén , 2006b Endlicher's (in Endlicher & Fenzl 1839) Latin description of Schwabea modesta is thorough and, although some details of the plant do seem to align with Monechma ciliatum, there are some anomalies, not least of which is the recording of four didynamous stamens. Whilst some species of Acanthaceae that usually have two fertile stamens can occasionally have aberrant flowers with four stamens perfected (McDade et al. 2021), particularly when in cultivation, this is rare in taxa that lack staminodes, as is the case in the Justicioid Lineage. Further, Endlicher recorded the seeds of S. modesta as having hygroscopic trichomes, which would not fit with M. ciliatum, unless he misinterpreted the bristly trichomes of that species. Endlicher (in Endlicher & Fenzl 1839) suggested a possible affinity with his own genus Russeggera Endl. (= Lepidagathis Willd.) and this could potentially fit with the description, although no known species of Lepidagathis from East Africa fit with the description of S. modesta. Meisner (1840) subsequently reaffirmed this close relationship by placing Russeggera and Schwabea in a separate tribe of Acanthaceae, Russeggereae. The available evidence therefore suggests that Clarke (1900) was correct in his assertion that Nees (1847) erroneously equated Schwabea with Pogonospermum, and we consider Schwabea to be an unplaced name Indeed, it was treated as such in the recent reclassification of Acanthaceae (Manzitto- ).
Monechma ciliatum is somewhat unusual in its seed indumentum in comparison to other known species of Monechma Group II, but it is nevertheless confirmed as a member of that clade in the study of Darbyshire et al. (2020), which rejected the exclusion of M. ciliatum from Monechma Group II (P < 0.001; see Table 2 of Darbyshire et al. 2020). In following Clarke (1900) and rejecting Schwabea s.s., Pogonospermum is therefore the earliest name for Monechma Group II sensu Darbyshire et al. (2020). The word root of the name Pogonospermum ("bearded seeds") is not ideal in view of the fact that, although several species in this clade have pubescent seeds, only those of M. ciliatum could be described as "bearded". We are nevertheless obliged to adopt this generic name for this clade if it is to be recognised as a discrete genus, as proposed below.
Should the generic name Monechma be conserved? An alternative to resurrecting both Meiosperma and Pogonospermum for Monechma Groups I and II would be to propose to conserve Monechma over one or other of these names. The obvious choice here would be to propose to conserve Monechma over Meiosperma given that the types of these two names are congeneric. As Meiosperma has not been used since Rafinesque's (1838) original publication of the name, it is plausible that a proposal to conserve Monechma over Meiosperma might be successful. However, there are problems with this choice. Firstly, as the large majority of diversity within Monechma s.l. falls within Group II (Pogonospermum), this would not result in much greater nomenclatural stability, as the names for only three currently accepted species (Monechma bracteatum, M. debile and M. monechmoides (S.Moore) Hutch., representing only half of the currently accepted species in Group I and only 7.5% of the species diversity across Monechma s.l.) would be retained. Secondly, the name Monechma debile is problematic as the circumscription of this species has varied widely in the past and, in its broadest sense, has been applied to most members of Monechma Group I, including to the more widespread and abundant taxon and type of Monechma, M. bracteatum (e.g. see Munday 1995;Ensermu 2006). Given this confusion, it is not greatly desirable to maintain the name Monechma debile.
The second alternative would be to propose to conserve Monechma over Pogonospermum. This would have the advantage of maximising nomenclatural stability given that 26 of the 34 currently accepted species in Group II have a name available in Monechma. However, this would require a proposal to change the lectotype of Monechma, from M. bracteatum to M. hispidum. The resultant significant change in circumscription of Monechmaincluding the preclusion of the most widespread, common and familiar taxa, i.e., M. debile and its allieswould likely create significant confusion for non-specialists over application of the name Monechma. Given that many species in Group II are highly range-restricted and littleknown species and only rarely cited in the literature (e.g., a search in May 2021 for Monechma in Google Scholar returns only 1,160 results), it would seem inappropriate in this case to propose the rather drastic change in the lectotype, given the controversy over such a proposal. Furthermore, such a change would run contrary to the valid decision made by Hochstetter (1844) in terms of the segregation of Pogonospermum from his Monechma as discussed above.
In conclusion, it is our opinion that neither option for conserving Monechma is desirable and thus we propose to accept Meiosperma and Pogonospermum, and adopt the necessary nomenclatural changes.

Generic delimitation and infrageneric classification in Meiosperma and Pogonospermum
A summary phylogeny placing the newly resurrected genera Meiosperma (Monechma Group I sensu Darbyshire et al. 2020) and Pogonospermum (Monechma Group II sensu Darbyshire et al. 2020) in the context of related clades in Justiciinae is presented in Fig. 1, based on the results of the RADseq phylogeny in Darbyshire et al. (2020). In the Nomenclatural Renovations section below, we provide the new combinations in Meiosperma and Pogonospermum for all currently recognised members of Monechma or Justicia section Monechma. Table 1 provides a summary of these two genera and how they relate to the nomenclature in Darbyshire et al. (2020). The two genera can be separated by application of the key below.

Key to distinguish the reinstated genera Meiosperma and Pogonospermum
Inflorescences of axillary or both axillary and terminal spikes; bracts elliptic, ovate or obovate, often imbricate, inflorescence units at each axil often > 1-flowered; seeds 2 -3 mm diam., lenticular with a sharp rim, ± symmetrical in cross section and lacking a prominent ridge on one side, glabrous*.  Fig. 8 in Darbyshire et al. (2020).

Meiosperma
Meiosperma, as currently defined, comprises six accepted species and two undescribed taxa (Fig. 2, Table 1). These species are closely allied morphologically (although M. tetrasperma is notable for having 4-seeded capsules, the other species having only 2-seeded capsules), and no infrageneric taxa are proposed. Species delimitation in this group is challenging, with significant infraspecific variation noted in several of the widespread species as currently delimited. A full revision of Meiosperma is desirable, ideally supported by an expanded phylogenetic study. Both Kiel et al. (2017) and Darbyshire et al. (2020) noted a close relationship between Meiosperma (Monechma Group I) and a paraphyletic Justicia sect. Harnieria (Fig. 1). However, Darbyshire et al. (2020) noted that monophyly of sect. Harnieria cannot be rejected by the results from the RADseq analysis. Meiosperma and species in sect. Harnieria are easily separated by differences in seed sculpturing: seeds of the former are smooth whilst those of the latter are variously tuberculate (see Fig. 14 Kiel et al. 2017). It is also worth noting that some species of sect. Harnieria, such as J. diclipteroides Lindau and J. heterocarpa T.Anderson, are heterocarpic, and have winged indehiscent 1seeded fruits in addition to the typical explosively dehiscent 4-seeded capsules (Hedrén 1989). In such cases the seeds in the indehiscent fruits are smoother than in the dehiscent fruits but still have some tubercles (see Hedrén 1989, Fig . 14D). This phenomenon of heterocarpy is not observed in Meiosperma.
In addition to the difference in seed sculpturing, the axillary inflorescences of Meiosperma are often more developed, forming spikes with broad imbricate bracts whilst those of sect. Harnieria are usually fasciculate. However, there is some overlap in inflorescence form, and Hedrén (1990) noted similarities in the structure of the cymose units within the inflorescences of members of these two groups. Darbyshire et al. (2020) provide further discussion on morphological similarities and differences between the two. Additional RADseq sampling in sect. Harnieria would elucidate the relationship between these taxa. Depending on the outcome of further studies, either sect. Harnieria will need to be elevated to generic status (Harnieria Solms) or Meiosperma will need to be expanded to include sect. Harnieria and so include taxa with both smooth and sculptured seeds. This would increase the size of Meiosperma significantly, as sect. Harnieria has c. 50 recognised species in Africa alone (Hedrén 1989, and with subsequent additions and modifications in e.g Vollesen 2010Vollesen , 2015Champluvier 2013).

Pogonospermum
Pogonospermum is a larger and morphologically more diverse genus than Meiosperma (as currently delimited see above), with 34 species currently recognised and with at least two undescribed taxa. Darbyshire et al. (2020) discussed the infrageneric variation in this group and noted two major clades, the first (the "ciliatum/scabridum" clade) with pubescent seeds and the second with glabrous seeds. This second clade was further subdivided into four clades that can be separated based on variation in inflorescence form, venation of the calyx lobes and bracteoles, and seed shape. In the Taxonomic Renovations section, we formally recognise these clades as sections of Pogonospermum (Table 1), which can be separated using the key below. The "ciliatum/scabridum" clade of Darbyshire et al. (2020) is subdivided here, with Pogonospermum ciliatum (the type species of the genus) treated in its own section, separate from sect. Savannicola which contains the other species with pubescent seeds; these two sections are easily separated by differences in plant life form and seed indumentum. Therefore, six sections are recognised in Pogonospermum (Figs 1 & 3 The largest section in Pogonospermum is sect. Aridicola that comprises 17 species, many of which are confined to the deserts and xeric bushlands of the succulent biome in southwest Africa (Munday 1983;Darbyshire et al. 2020). In Darbyshire et al. (2020), P. incanum (Nees) I. Darbysh. & Kiel was resolved as  Munday (1995). These species all share rather fleshy and narrow, obovate to linear-oblanceolate or linear-terete leaves, and an indumentum comprising complex trichomes that are either biramous, as in P. incanum and sparsely so in P. saxatile, or are anvil-shaped, i.e. hairs with a short stalk terminated by a single arm, as in M. crassiusculum and P. robustum (Munday 1995). It would be desirable to include these species in a future expanded RADseq dataset to confirm or refute their alliance to P. incanum. The other species of sect. Aridicola lack these complex trichomes on the leaves, although P. calcaratum (Schinz) I. Darbysh. & Kiel has branched trichomes on the stems (Munday 1995).
Pogonospermum incanum and its putative allies share a striking resemblance to the southern African Justicia cuneata Vahl (see Immelman 1995). Justicia cuneata comprises three subspecies, two of which have glabrous leaves, but the third, subsp. hoerleiniana (P.G.Mey.) Immelmann from southwest Namibia, has dense swollen anvil-shaped trichomes throughout its vegetative portions. Justicia cuneata has rarely been collected in fruit and we have not seen mature capsules or seeds of this species; Immelmann (1995) describes the capsules as 1seeded, clavate and hard in texturean unusual arrangement for Justicia and indeed for members of Justiciinae in general although, as noted above, indehiscent single-seeded fruits are known in some heterocarpic species of Justicia sect. Harnieria (Hedrén 1989). Justicia cuneata has not, to our knowledge, been included in any molecular phylogenetic studies to date and it should be considered a priority for future studies to determine whether it is allied to the P. incanum group or if (as is quite possible) this represents a case of convergent evolution.

Nomenclatural Renovations in Meiosperma and
Pogonospermum Below we provide a species inventory for the two resurrected genera Meiosperma and Pogonospermum, formalise the necessary new combinations and select lectotypes where required and appropriate. We also formally describe the newly recognised sections within Pogonospermum.
It is acknowledged that species delimitation requires further research in some groups of both Meiosperma and Pogonospermum. Furthermore, some species currently placed within Justicia may ultimately   Clarke (1900: 220). Monechma scabrinerve C.B. Clarke (1900: 215 Vollesen (2010Vollesen ( , 2015 for a list of examples.   NOTE. Hedrén (2006b)  This species has been treated within Monechma debile by some authors, for example Munday (1995). , but the five-lobed calyx (fide Clarke 1900) does not fit with that species, and examination of the type specimen at BM reveals that the seeds of this species match those of Meiosperma. It may be a depauperate form of M. monechmoides but this requires further investigation. The isotype at LISU represents slightly better material than that at BM. NOTE. Pfeiffer (1874) lectotypified Pogonospermum with P. ciliare by reference to "Justicia ciliaris Vahl". Although Pfeiffer wrongly attributed that name to Vahl rather than to Linnaeus (1781), it seems highly likely that this error was based on reference to Hochstetter (1844) who also attributed J. ciliaris to V a h ls e e d i s c u s s i o n a b o v e . P f e i f f e r ' s lectotypification is therefore acceptable.
Pogonospermum sect. Pogonospermum RECOGNITION. Erect annual herbs. Inflorescences of short or more elongate terminal spikes with bracts markedly to only slightly differentiated from the leaves; bracts, bracteoles and calyx lobes all conspicuously pale-ciliate. Calyx lobes with only the midvein prominent, margins pale-hyaline. Seeds oblate in face view, one side flattened, the other side convex and with a central ridge, with tufts of bristly multicellular trichomes at apex and on one side of hilum at base, the two tufts with trichomes orientated in opposite directions. Fig. 3A. http://www.ipni.org/urn:lsid:ipni.org:names:77254447-1 RECOGNITION. Perennial herbs with a ± woody base and rootstock. Inflorescences of terminal spikes, sometimes also with additional spikes in the distalmost leaf axils; bracts markedly differentiated from the leaves, linear, lanceolate, to narrowly elliptic, oblong or oblanceolate; bracts and bracteoles, single-veined or sometimes with the two lateral veins also prominent, margins sometimes markedly paler than rest of surface. Calyx lobes with only the midvein prominent, sometimes with two parallel lateral veins also more faintly visible, margin often pale-hyaline. Seeds ± oblate in face view, one side flattened, the other side convex and with a central ridge, at first white sericeous-puberulous, these trichomes sometimes sparse or absent on mature seeds. Fig. 3B RECOGNITION. Perennial herbs with a ± woody base and rootstock. Inflorescences of terminal spikes, sometimes also with additional spikes in the distalmost leaf axils; bracts markedly differentiated from the leaves, linear or lanceolate to narrowly oblanceolate. Calyx lobes (and usually also bracts and bracteoles) with three veins prominent and often also differently coloured to the paler intercostal surface. Seeds ± globose in face view, only slightly flattened, glabrous. Fig. 3D RECOGNITION. Shrublets, perennial herbs or more rarely annual herbs, often with intricate branching, mature branches sometimes woody and gnarled, foliage often dense. Inflorescences of axillary single-flowered cymes, bracts undifferentiated from the leaves or only slightly differentiated, flowers sometimes together forming a poorly defined, leafy terminal spike. Calyx lobes with only midvein prominent or with no veins prominent, margins sometimes hyaline. Seeds rounded or ovate in face view, one side flat or concave to slightly convex, the other side more markedly convex and usually with a central ridge, sometimes mottled or intricately patterned, glabrous.  NOTE. The multiple synonyms reflect the considerable variation within this widespread species, but it was found to be monophyletic in the RADseq study of Darbyshire et al. (2020).
Monechma molle E. Mey. ex C.B.Clarke (1901: 69). Justicia dregei J.C. Manning & Goldblatt (2014: 46). DISTRIBUTION. Africa: Namibia, South Africa. NOTE. The two type specimens at K are mounted on the same sheet; K000378762 derives from Hooker's herbarium whilst K000378763 is from Bentham's herbarium. As Nees states that he saw the collection in Hooker's herbarium, the former is here selected as lectotype. Additional collecting locality information, believed to have been added at a later date by Charles Baron Clarke, is noted on both the Kew specimens: "between Holgat River and the Orange River.
[alt.] 1000 -1500 ft. Little Namaqualand". Nees also saw a duplicate in "h. Drèg[e]" that is here assumed to have been in Berlin and was likely to have been destroyed during World War II.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/ licenses/by/4.0/.