A new genus, Rockhausenia (Compositae: Senecioneae: Senecioninae)

Werneria Kunth (Compositae: Senecioneae: Senecioninae) was effectively lectotypified by Rockhausen based on Werneria rigida. Two later lectotypifications have apparently been made, firstly (presumably by Cabrera) W. graminifolia and secondly (presumably by Funk) W. nubigena, yet the Index Nominum Genericorum website has no type stated. In removing W. rigida to the then newly recognised Xenophylllum, Funk moved Rockhausen’s type thereby creating a superfluous generic synonym. All species that Funk recognised in Xenophyllum are thus wernerias in the nomenclatural sense. Two species of Xenophyllum described since Funk’s revision are transferred to Werneria: X. funkianum and X. lorochaqui. The remainder of the species of Werneria are here placed in the new genus Rockhausenia (Compositae: Senecioneae: Senecioninae), a generitype is selected, and the corresponding 27 combinations are made at species level; the relevant subgeneric and sectional combinations are also made with two subgenera recognised, one with three sections. The confusion surrounding the generic name Oresigonia is commented on. A nomenclator of all relevant names in Rockhausenia, Werneria and Xenophyllum, is provided.


Introduction
During October and November 2015, the author was fortunate to make a second study visit to the Herbario Nacional de Bolivia (LPB), Universidad Mayor de San Andrés, La Paz, Bolivia, to continue with his work on his Annotated Checklist of the Compositae of Bolivia (Hind 2011). On this occasion, amongst other things, I had the opportunity to go through several more recent collections made in the Parque Nacional Madidi, Prov. Tamayo, Depto. La Paz, Bolivia (material collected as part of a long-term collaboration between LPB and Missouri Botanical Garden -MO). Included within the collections were four of an unusual looking species of what appeared to belong to Werneria Kunth (Compositae: Senecioneae: Senecioninae). Other than one evidently alcohol-treated collection, the other three show distinctive purplish abaxial leaf surfaces, purple, sparsely to moderately, villous pedicels, phyllaries with the apex or apical half coloured purple, and white ray limbs markedly pinkish towards the abaxial surface of the apex. A loan of this material from LPB to K was secured, in order to study the specimens further, and prepare a line drawing of the taxon.
Werneria, in the original sense, can be recognised by the following broad diagnostic description: Plants rhizomatous perennials, or dwarf erect or decumbent subshrubs, rarely of solitary rosettiform plants. Leaves simple, alternate or rosulate, rarely distichous, bases often pseudopetiolate and frequently sheathing, or leaves somewhat fleshy, conspicuously 3-partite and narrowed into a pseudopetiolar base. Inflorescences sessile or scapose, single-headed, usually ebracteate, or bracts foliose to scale-like, glabrous or lanate. Capitula usually heterogamous and conspicuously radiate, or sometimes disciform, or rarely homogamous and discoid; involucres campanulate and ecalyculate; phyllaries uniseriate, linear, bases usually fused, apices acute or obtuse; receptacle flat to convex, glabrous and epaleaceous. Ray florets (when present) few to many, female and fertile, ray limbs often appearing entire or scarcely 3-toothed, white or yellow, often discolorous and reddish, pink, or purplish abaxially. Disc florets few to numerous, hermaphrodite, corollas 5-lobed, usually yellow, sometimes off-white or purplish; style arms truncate with sweeping hairs in a crown. Achenes usually glabrous, sometimes setuliferous, setulae of twin-hairs, these long and villous; carpopodium inconspicuous and annular; pappus setae uniseriate, setae persistent and finely barbellate, usually white, whitish or rarely purplish or pinkish.
A comparison of the material mentioned above with the other species listed and accepted by , Hind (2011), Jørgensen et al. (2005), and Jørgensen (2014), for Bolivia, and those taxa in neighbouring Peru (Beltrán 2017;Beltrán & Leiva 2018), showed no obvious match in the genus Werneria. Analysis of the material in K, and subsequent contemporary studies, have shown this material to be Werneria glandulosa Wedd. (c.f. Calvo et al. 2020a: 33 -35). Rockhausen (1939b: 334) listed the name under 'Species incertae et excludendae. A. Species incertae', with the note that 'Von folgenden Arten hatte ich keine Material und keine Originale vorliegen. Fernerhin was es nicht möglich, dieselben nach der Beschribung sicher einzureihen.' (= 'I had no material and no originals of the following species. Furthermore, it was not possible to classify them with certainty according to the description.'). In adding Werneria glandulosa to the final version of my Annotated Systematic Checklist of Bolivian Compositae (Hind 2011), it became clear that there was a nomenclatural issue with continuing use of the current concepts of Werneria sensu stricto and Xenophyllum V.A.Funk.
Werneria, described in Nova Genera et Species Plantarum (Kunth 1818: 148), contained six species now considered to belong to two genera, Werneria sensu stricto and Xenophyllum (22 spp. -Funk 1997c;Calvo & Moreira-Muñoz 2020). Kunth did not designate a type, nor did he recognise any infrageneric ranks, and the generic description covered the characters of all the species described. The current record for Werneria in Index Nominum Genericorum (Farr & Zijlstra 1996+) indicates that there is no stated or selected type; from my discussion below, this is shown to be incorrect, as suggested by the literature (Rockhausen 1939a;Cabrera 1971;Funk 1997a, Funk 1997c

The issue
Lectotypification of Werneria Kunth There is a categorical statement in Rockhausen's revision of Werneria: 'Gattungstypus ist Werneria rigida.' (Rockhausen 1939a: 249), and later (p. 293) he specified the type of the name W. rigida as a Humboldt & Bonpland collection. His description of the species is not in conflict with Kunth's generic description, and there was no sense of doubt over his statement of the generitype. He placed the species in his 'subgen. Euwerneria (Gay) Rockh. sect. Aciculares Rockh.'; both infrageneric concepts contain the nomenclatural type of the genus (the generitype) and would now be considered as subgen. Werneria and sect. Werneria. This typification appears to have been overlooked, or ignored, by later authors, as many have only cited the second part of Rockhausen's revision, and continue to use a nomenclaturally incorrect concept of Werneria, unaware of the implications (e.g. Linares-Perea et al. 2014;Calvo et al. 2017;Beltrán & Leiva 2018;Calvo & Meneses 2019;Calvo & Moreira-Muñoz 2019;Calvo et al. 2020a, b).
Three further statements of the 'lectotype' of Werneria have been made: 1) Although Cabrera's early revision of the Argentinian species (Cabrera 1948) mentioned nothing about the type, he later (Cabrera 1971: 275) stated that the lectotype was W. graminifolia Kunth. He repeated this statement (Cabrera 1978: 469), whilst still not stating who designated it; it is assumed that it was Cabrera himself. Calvo et al. (2020a: 6) stated that it was Cabrera in his Flora Patagonica account (Cabrera 1971: 275). Werneria graminifolia, an Ecuadorian endemic, occurs in none of Cabrera's treatments of the genus.
2) In recognising Misbrookea V.A. Funk, Funk (1997a: 110, 1997b) stated that the 'type species' was 'Werneria nubigena H. B. K., and [the genus] is referred to hereafter as Werneria s.str.' In both cases she only cited the second part of Rockhausen's paper on Werneria , possibly unaware of the first part where the type citation was made and several new infrageneric taxa described (Rockhausen 1939a). Later, Funk (1997c: 235), in her description and synopsis of Xenophyllum, again mentioned that 'Werneria s.l.' contained at least 'three identifiable groups ... that can be recognized as distinct genera.' One of these 'contains the type species, Werneria nubigena Kunth, and can now be referred to as Werneria s. str. ...'; Xenophyllum dactylophyllum (Sch.Bip.) V.A.Funk was designated as the type of Xenophyllum. In none of the accounts is there any indication who designated W. nubigena as the lectotype of Werneria; it can be assumed to have been selected by Funk, unless evidence is eventually found to the contrary.
3) Freire & Ariza Espinar (2014: 220) stated that the lectotype of Werneria was W. graminifolia, as indicated by Cabrera (1971: 275), but also noted the other citations by Rockhausen (but only citing the second part of Rockhausen's revision, Rockhausen 1939bwhere the type citation does not occur) and Funk (1997b). No additional explanation was provided as to why W. graminifolia was accepted as the lectotype over the selection by Rockhausen. No type was designated or mentioned by Gray (1861), or Blake (1928, when several new species were added in their short synoptic treatments. Tkach et al.'s statement, that Werneria nubigena was the 'type species of Werneria Kunth', with reference to the Index Nominum Genericorum website (Tkach et al. 2016: 101), is also without foundation. The website simply records the type as 'non designatus', and gives no indication of a lectotype (ING, last accessed 10 th October 2020).

The position of Oresigonia
Lessing (1832: 393) appears to have been the first author to mention the generic name Oresigonia, albeit referring to two completely different taxa. He first used it under his treatment of Werneria, alongside Werneria disticha Kunth, in referring to Willdenow's manuscript name, Oresigonia latifolia. It was used lower down on the same page in reference to Culcitium Humb. & Bonpl., albeit as 'Oresigonia Schlechtd. im Mag. naturf. Fr. ' Lindley (1836: 261) also attributed Willdenow's mss. name to Werneria and, immediately after, Schlechtendal's to Culcitium, a position reflected by de Candolle (1838: 323) with the generic name as 'Oresigonia Willd. herb. ex Less.' in the synonymic heading to his treatment of Werneria, and on the following page as 'Oresigonia Schlecht. in Berl. naturf. mag. ex Less.' in the synonymic heading to his treatment of Culcitium. The explanation of this was effectively provided by Blake (1937: 389 -390), as simply an error on Lessing's part, by wrongly referring to Schlechtendal's genus Lasiocephalus (Schlechtendal 1818: 308) (Pelser et al. 2007). Anticona E.Linares, J.Campos & A.Galán (based on Werneria glareophila Cuatrec.) has never been sampled, so its relationship has yet to be elucidated. The generic disposition of the wernerias needs to be reconsidered. Since Rockhausen's lectotypification is valid and, in recognising Xenophyllum, Funk created a superfluous generic synonym for the wernerias sensu stricto. Providing a solution is a purely nomenclatural issue: the transfer of the remaining wernerias to a new genus, Rockhausenia, which is proposed below.

Synoptic treatment of Rockhausenia
I have refrained from providing a key to species as there is a perfectly adequate key to all species in Calvo et al. (2020a), and a country-specific key is available for Peru by Beltrán (2016) and Argentina (Freire & Ariza Espinar 2014). Adequate keys to taxa originally considered to be in Xenophyllum were provided by Calvo & Moreira-Muñoz (2019) expanded from Funk (1997c).
In the following synoptic treatment of Rockhausenia, under each accepted name synonyms are provided in chronological order. For each name references are provided, and a type citation given for all basionyms. The type citation is a classical one, and is simply a transcription from the basionym protologue. The location of known types is stated (acronyms following Thiers [continuously updated]) and, when available, barcode numbers of type material given. In instances where institutional barcodes unnecessarily start with six or more zeros the barcode number is concatenated to the remaining digitsthe suggestion that institutions could ever have that number of specimens when fully digitised is, frankly, ludicrous. Herbarium or manuscript names are also provided where they are relevant to the history of taxa concerned and have appeared in the literature. A full nomenclator of names is also provided.
RECOGNITION. Species of Rockhausenia are mostly low, perennial mat-or cushion-forming perennials, very rarely aquatics or marginals, although sometimes also found as isolated rosettiform plants/individuals. Species usually have entire (rarely pinnatifid/lobed) concolorous leaves, and characteristically have solitary, sessile or pedicellate, capitula, an involucre of basally fused uniseriate phyllaries, and discolorous ray floret limbs that are usually white (rarely yellow) adaxially and pinkish or purplish abaxially; three species (of the 27 in the genus) possess discoid capitula. In contrast, Werneria spp. usually form dense or loose hummocks or mat, with the rhizomes and upper stems densely leafy for some length, with the exposed leaves green and those within the tussock or mats often brown or blackish (or sometimes whitish). The capitula in Werneria are heterogamous and radiate (very rarely are the rays greatly reduced or absent and the heads effectively disciform), ray limbs are usually white or very rarely violetpurple or yellow. Misbrookea, a rhizomatous rosettiform monotypic genus from Bolivia and Peru, is easily recognised by the long (3 -5 mm) whitish strigose hairs found on both leaf surfaces and on the involucres (vs glabrous leaves and involucres or villous scape and involucres in few Rockhausenia spp.), and style arm apices with long multicellular hairs. DISTRIBUTION. The genus Rockhausenia is a New World endemic, essentially restricted to South America, currently with 27 species of the high Andes (of Argentina, Bolivia, Chile, Colombia, Ecuador, Peru and Venezuela), with the exception of the most widespread species, Rockhausenia nubigena (Kunth) D.J.N.Hind, which is found from southern Mexico and Guatemala south to southern Bolivia.

New combinations in Rockhausenia
For convenience, the taxa appear in the order treated by Rockhausen (1939a, b) together with several additional species added where appropriate. Chilena (Gay) 4: 216 (1849). Note: The symbol used by Rémy is a typographical section symbol | §| Elsewhere Rémy used the abbreviation 'Secc.', or even in expanso, 'Seccion', with the || symbol as a clear lower rankin Cardamine for example (1: 109), or Nasturtium (1: 117)where it was certainly used as a section! However, the word 'Seccion' has been used as another rank, below tribe, and above genera. Werneria Kunth subgen. Anactis (J.Rémy) Rockh. (Rockhausen 1939a: 265 de Candolle 1911de Candolle , 1914Cogniaux et al. 1910;Lingelsheim et al. 1909aLingelsheim et al. , 1909bPax 1908 DISTRIBUTION. Bolivia (La Paz), Peru (Cusco, Puno). NOTE. As this species is still relatively poorly illustrated, bar the colour images provided by Alfredo Fuentes (LPB) to Calvo et al. (2020a), which he also provided to me after my earlier study visit in 2015, I have provided a black and white line illustration to allow some of the critical details to be seen, and comparison made with similar speciessee Fig. 1.

Declarations
Conflict of interest. The author declares no conflict of interest.
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/.