Allionrs Aloe names (Asphodelaceae): nomenclature and typification

The taxa belonging to the genus Aloe published in Synopsis methodica stirpium horii regii taurinensis (Allioni 1760) and in Auctarium ad synopsim meihodicam stirpium horti regii taurinensis (Allioni 1773) were examined. The protologues o f Aloe maculata All. and A. verrucosospinosa All. are analysed and lectotypes designated. The homonymy o f A. succotrina All. w ith A. succotrina Weston (1770) is recognized, and the lectotype o f this last name designated. Epitypes are selected to fix the application of all three names. Short differential diagnoses o f the three species are given and their distribution ranges discussed; distribution maps based on specimens held in the South African National Herbarium (PRE), KwaZulu-Natal Herbarium (NH). Compton Herbarium (NBG) and the South African Museum Collection (SAM) held in NBG are provided.

The taxa belonging to the genus Aloe published in Synopsis methodiea stirpium horli regii taurinensis (Allioni 1760) and in Auetarium ad synopsim methodicam stilpium horti regii taurinensis (Allioni 1773) were examined.The protologucs of Aloe macufata All. and A. verrucosospinosa All. are ana lysed and Icctotypes designated.The homonymy of A. suecotrina All. with A. suceotrina Weston (1770) is recognized, and the lectotype of this last name designated.Epitypes are selected to fix the application of all three names.Short differential diagnoscs of the three species are given and thcir distribution ranges discussed; distribution maps based on specimens held in the South African National Herbarium (PRE), KwaZulu-Natal Herbarium (NH), Compton Herbarium (NBG) and the South African Museum Collection (SAM) held in NBG arc provided.
Synopsis methodica stirpium horti regii taurinensis, published by Carlo Allioni  probably in 1760 (see Stafleu & Cowan 1976), is the first printed catalogue of the plants cultivated in the Botanic Garden of the University of Turin.The names of the species listed in this publication followed the binomial nomenclature system.For 88 of these, Allioni did not find a previously published binomial in the contemporary literature.Instead he gave them a number corresponding to a note.In this note he cited an earlier polynomial, when present, with the corresponding bibliographic citation, or he provided an original description in the absence of citations or as a complement to indications of previous authors.These nomenclatural gaps were later filled by Allioni in Auctarium ad synopsim methodica stirpium horti regii taurinensis (1773, following Stafleu & Cowan 1976).In Auctarium, Allioni provided new binomials for 77 of the previously cited polynomials.Of these, 47 were based on descriptions provided by other authors (Dandy 1970).
L. (Asphodelaceae), are ana lysed and typified.Allioni also described A. succotrina, but this epithet was previously used, for the same taxon, by Weston (1770), thus Allioni's name is a homonym.Weston's name is here typified.
Each species is attributed to its correct taxonomic group and the nomenclature is updated.Diagnostic morphological characters are based on the current taxonomic circumscription of the species and geographical distribution maps are provided for each of the three species.These maps are based on specimens held in the South African National Herbarium (PRE), KwaZulu-Natal Herbarium (NH), Compton Herbarium (NBG) and the South African Museum Collection (SAM) held in NBG.All relevant literature was consulted and Allioni's herbarium examined.In addition, other collections preserved at TO (Bellardi's herbarium and TO-HG) that are taxonomically correlated to the species cited in these two catalogues were checked.
A list of the species referred to the genus Aloe by Allioni is presented in Table 1.It includes the binomials published in Synopsis or in Auctarium, or, when no binomial was used by Allioni, his polynomials as cited in the notes in his books are given; information on the specimens in Allioni's herbarium is also presented.In Synopsis, 13 species are recorded.For four of these (Aloe disticha L., A. spiralis L., A. retusa L., A. variegata L.) Allioni used the Linnean binomial, whereas for the remaining nine he listed in a note the corresponding polynomials assigned by Dillen (1732), J. Commelijn (1697) and C. Commelijn (1701Commelijn ( , 1703)).Of these nine, three were not included in Auctarium: for A. margaritifera L., the Linnean binomial is applied; for A. carinata All. and A. glauca All., the binomials proposed had already been used previously by Miller (1768) (see Chiovenda 1912) and are illegitimate homonyms.A further binomial, A. succotrina All., still currently widely used (see for example Newton 2001a, b), is predated by A. succotrina Weston (1770), and is therefore an illegitimate homonym.Finally, A. maculata All. and A. verucosospinosa [sic] All. are valid binomials.In Allioni's herbarium, specimens of two Aloe species were found which can be assigned respectively to Aloe maculata and to another species, only reported with the polynomial Aloe africana flare rubro ... in Allioni (1760), which could not be positively identified (see Table 1).No other specimens were found in Bellardi's herbarium and in TO-HG.5), that referred to an Aloe described as Aloe Africana cau-lescens foliis spinosis maculis ab utraque parte albicantibus notatis.In Allioni's herbarium there is a specimen labelled Aloe afric.coerulescens [sic} fa!. spin.macu!. ab utraq.parte albicantib.notatis.This is a candidate for a lectotype, however, it is not dated, thus it cannot be ascertained if it can be considered as original material in the sense of the TCBN (McNeill et al. 2006); there is also the difference in the wording, with coerulescens replacing caulescens, but this is probably simply a spelling error.Allioni's reference to the plate of excellent quality in C. Commelijn (1701) makes it preferable to designate this plate as the lectotype of A. maculata.Several authors (Dandy 1970;Webb 1980;Glen & Smith 1995;Newton 2001a, b

Aloe maculata
) considered the use of Allioni's binomial as correct since it has priority over the other binomials.This view was not supported by the proposal made by Gilbert & Demissew (1997)   tural stability, as this binomial is better known than Allioni's.Nevertheless, A. saponaria has not been conserved (McNeill et al. 2006: art. 14).Chiovenda (1912) gave A. obscura Mill.(1768) as the accepted name considering A. maculata to be a synonym.This view has not been followed by any other authors and the correct application of the epithet obscura remains doubtful (Stungo 1996).At present, the binomial A. obscura is considered to be of unresolved application, and it may refer either to a variant of A. maculata (and then becoming the accepted name, taking precedence over A. maculata) or to a hybrid, with A. maculata as one of the parents.However, no plants are known in South Africa that can be matched to this species or to Miller's description, and no specimen, nor natural population, has been found which could allow us to make a comparison (L.Newton pers.comm.).
The collections of Aloe maculata that were available to C. Commelijn are attributed to H.B. Olden land (± 1663-early 1697).A. maculata was cultivated in the Dutch East India Company's Garden at the Cape in 1695, when Olden land was Superintendent and a species with the name Aloe Ajricana caulescens joWs spinosis maculis ab utraque parte albicantibus notatis was listed in his herbarium.Oldenland's name was later used by C. Commelijn in the plate that first depicted this species.It is not known where Oldenland collected the first specimens of A. maculata as the species is widely distributed, from the Cape Peninsula to Swaziland (Reynolds 1950;Van Wyk & Smith 2003), but more than likely it would have been in or near Cape Town since this is a known locality for the species.
The plate that depicts the plant, here designated as lectotype, had already been referred to in the literature as the iconotype of this species (Glen & Hardy 2000).However, the term iconotype has no status in the typification process as fonnalized in the lCBN (McNeill et af. 2006).The plant depicted in the plate lacks an inflorescence and consequently it lacks some of the critical characters needed to easily distinguish it from other spotted-leaved aloes.For this reason, an epitype is designated here.
Geographical distribution: Aloe maculata has one of the widest known distribution ranges of all the aloes.It occurs commonly from the Cape Peninsula along the climatically moderate southern African coastal belt below the Great Escarpment to as far north and northeast as KwaZulu-Natal, Mpumalanga, Limpopo and Swaziland, but apparently does not enter neighbouring Mozambique.It also occurs in the southern African interior in Lesotho and in South Africa's Free State Province, both of which have a harsh, continental climate (Figure 2).(Dandy 1970), but the first valid publication of the name was by Weston (1770).Apart from the short description, Weston gave no other indication (distribution, materials, etc.) for his succotrina, but in the introduction to his publication he referred to Linnaeus (1764).For the description of Aloe succotrina, Weston copied Linnaeus verbatim [the name was not validly published by Linnaeus, who simply copied J. Commelijn's (1697) polynomial, sinking it into the synonymy of A. perfo-Uata].That A. succotrina Weston was based on J. Commelijn's plate was also reported by Brown (1911).We therefore designate the plate published in Horti medici Amstelodamensis (1.Commelijn 1697: fig 48) as the lectotype of A. succotrina Weston.Since Allioni (1773) also referred to the same J.Commelijn figure, his A. succotrina is a later homonym of the name of Weston.
The origin of the name and its misapplication to other, non-South African species, has been discussed in the literature (see Reynolds 1950).Aloe succotrina does not occur on the island of Socotra and it is suggested that the epithet derives from 'succocitrina', referring to the yel-low colour of the sap crystals when pounded (Reynolds 1950), although the purple colour of the dried leaves is one of the distinctive characters of the species (Glen & Hardy 2000;Van Wyk & Smith 2003).
Given that the purple colour of the dry leaves of Aloe succotrina is an important diagnostic character for the species, it could be argued that the illustration (plate) alone is insufficient to allow for the positive identification of the species.It is thought that A. succotrina may have been collected for the first time near Kloof Nek (Reynolds 1950;Smith & Van Wyk 1996).Brown (1911: 142) states that 'Although Aloe succotrina is not mentioned by Thunberg in his Prodromus or Flora, yet in his Travels, English translation, edn 3, vol I: 213, he mentions, under the date of 14th December 1772, when at a farm near Slange River, in the Oudtshoorn Division, that "Here we saw quickset hedges of Aloe succotrina" '.This locality is quite remote from KloofNek, which is in Cape Town.According to Brown (1911), Thunberg probably referred to A. fruticosa, which is now regarded as a synonym of Aloe arborescens Mill.Even though A. succotrina had therefore been known for more than 200 years, this species was only accurately recorded for the first time in the wild in 1906 by Rudolf Marloth.It was then stated to occur in a field of boulders above Newlands, presumably in or near what is today the Kirstenbosch National Botanical Garden, and also near Hout Bay along the coast near Cape Town.A voucher specimen for the occurrence of the species at the fonner locality (Marloth 3967) is therefore here designated as the epitype.
Diagnostic characters: shrubby aloe with tall, erect, unbranched inflorescences and greyish green leaves that, once dry, are persistent and purple, and with white, deltoid teeth on the margins.
Geographical distribution: the species is restricted to the southwestern Cape of South Africa.It typically occurs in fynbos vegetation on and against sandstone rocks and cliffs from the Cape Peninsula eastwards to Hermanus (Figure 4).Taxonomic and nomenclatural notes: the original spelling was Aloe verucosospinosa.Following article 60.1 of the lCBN (McNeill et al. 2006) the orthographically correct name is A. verrucosospinosa.No original material of this species was found in Allioni's herbarium.Wjinands (1983: 124) used plate 46 in C. Commelijn (1706) as the lectotype of Aloe humilis and A. verrucosospinosa.However, Allioni referred to page 77 in C. Commelijn (1703) to which plate 26 is attached; although plate 46 (C.Commelijn 1706) apparently refers to the same plant, it is not simply a reproduction, since it differs in several details and includes an inflorescence which is missing from plate 26.Thus, plate 46 in C. Commelijn (1706) cannot be considered as original material of A. verrucosospinosa in the sense of Art.9.2 of the ICBN and consequently it cannot be used as its lectotype.We designate here, as lectotype of Aloe verrucosospinosa All., plate 26 in Praeludia botanica (c.Commelijn 1703).

Aloe verrucosospinosa
Although the vegetative characters shown in the plate 26 are consistent with the concept of A. humilis, and the same specimen was apparently later depicted with an inflorescence, it may be confused with other species which occur in, or approximate to, its distribution area, especially since two look-alike miniatures (A.aristata Haw. and A. brevifolia Mill.) also have leaf tubercles and marginal teeth.For this reason, an epitype is designated here.It is possible that the first collections of Aloe humilis were made near Oudtshoorn, the westernmost locality of the species on the route taken by Oldenland (Reynolds 1950).A collection made by Pole Evans in 1915 under the number 194 (PRE), at Mossel Bay, a locality ± 80 km from Oudtshoorn which is the westernmost occurrence of the species along the coast, is here designated as epitype.It was cited by Reynolds (1950) under A. humilis.
As was the case with Aloe maculata, the material of 'Aloe Afric.humilis spin.et verrucis obsita' that was available to Commelijn, and was later used for the descriptions of A. humilis and A. verrucosospinosa, is attributable to Oldenland.Oldenland accompanied the Schrijver expedition towards the eastern interior of the Eastern Cape in 1689.The expedition reached the vicinities of Oudtshoorn and Uniondale, travelling into the arid interior as far east as 30 km northwest of Aberdeen.Aloe humilis was one of the species that was planted in the Dutch East India Company's Garden at the Cape, and was growing there in 1695 when Oldenland was Superintendent.The name 'Aloe Africana humilis spinis inermibus et verrucis obsita' was already given by Oldenland and later used by Commelijn (Reynolds 1950).Allioni's binomial is not cited by Newton (200 Ia,b) nor is it listed in TROPICOS 2008, the IPNI 2008 or World Checklist of Selected Plant Families (WCSP) 2008.However, Dandy (1970), Wijnands (1983: 124), APCD 2008 and Plants of southern Africa (PO SA) 2008 cited A. verucosospinosa [sic] as a later synonym of A. humilis (L.) Mill., an interpretation with which we agree.
Diagnostic characters: miniature aloe with leaves carrying soft, harmless tubercles and teeth, and short, stout inflorescences with large, loosely packed, pencil-shaped flowers.
Geographical distribution: the species has a wide distribution range in the southern and eastern Cape of South Africa, where it occurs in the arid interior in sparse shrublands in the Little and Great Karoo, as well as on coastal hills in renosterveld or in dry, subtropical thicket, from Mossel Bay in the west to Grahamstown in the east (Figure 4).
We are very grateful to Prof. Len Newton (Nairobi) for valuable nomenclatural and taxonomic advice.Comments by Dr Urs Eggli (ZUrich) and an anonymous referee greatly improved the manuscript.Ms Hester Steyn of the South African National Biodiversity Institute, Pretoria, is thanked for compiling the distribution maps.The curators of the National Herbarium (PRE), Pretoria, the Compton Herbarium (NBG, Cape Town), which also
FIGURE 2.-Geographical distribution of Aloe maculata All. based on specimens held in the National Herbarium (PRE), KwaZulu-Natal Herbarium (NH), Compton Herbarium (NBG) and the South African Museum Collection (SAM) held in NBG.