Solanum villosum Mill.

This common European species was long known as S. luteum (e.g., Hawkes and Edmonds 1972) and this name was in wide use in botanical gardens as the name for this red-fruited taxon. The name S. villosum Lam. was recognised as a synonym for S. luteum. The name S. villosum Mill. was not in use in Europe until resurrected by Edmonds (1979b), but had been used by Stebbins and Paddock (1949) in their treatment of the morelloids of California. Edmonds (1979b), following what is now Article 11.5 of the Code (McNeill et al. 2012), changed the name to S. villosum Mill. citing the explicit synonymisation of S. luteum with S. villosum Mill. by Stebbins and Paddock (1949). Dostál (1949), only a month later than Stebbins and Paddock (1949), synonymised S. villosum with S. luteum, but unfortunately used the name in the Lamarckian sense, although in a letter to Edmonds (quoted in Edmonds 1979b) equated S. villosum Mill. and S. villosum Lam. Many specimens in European herbaria are still identified as S. luteum or as S. alatum (see Knapp et al. 2017 for discussion of S. alatum).

Variation in pubescence and stem characteristics within S. villosum has been recognized at many ranks from subspecies, varieties, and different species (Hawkes and Edmonds 1972; Edmonds 1979b, 1984b; Henderson 1974; Symon 1981; Edmonds and Chweya 1997; Olet 2004; Dehmer and Hammer 2004). The various synonyms and infraspecific ranks reflect the wide morphological variation observed within the species, where indumentum can range from nearly absent (the plants glabrous) to densely pubescent with eglandular and/or eglandular trichomes. The recognition of the glandular variants at infraspecific rank is not supported by AFLP data (Manoko 2007; Olet et al. 2011). In fact, studies have shown that pubescence varies in individuals according to age, where seedlings can exhibit glandular hairs which are lost during growth (Henderson 1974; Seithe 1979; Edmonds 1982). Here we have an inclusive, broad concept of S. villosum that includes this variability but does not recognise it at any rank.

Solanum villosum can be easily recognised from similar species by its orange-red berries (ranging from yellow to bright red) that are slightly elongate and ellipsoid. The only other species present in the Old World with pale yellow fruits is S. palitans, but this taxon differs in being prostrate plant with thinly membraneous 3-lobed leaves and is easy to tell apart. Rest of the morelloids in the Old World have rounded or subglobose black or green berries. Solanum villosum is much more common in southern Europe than is S. nigrum, while S. nigrum becomes more common towards northern and central Europe. These latter two species are difficult to tell apart based on herbarium material alone, especially when mature fruit colour is not available. Calyx lobes of S. villosum are generally longer than those of S. nigrum (relative to width), strongly reflexed at fruit maturity, and the sinus is rounded (as opposed to acute in S. nigrum), and in flower the calyx tube has pale patches that look like small windows or white/translucent areas below the sinus itself. This is particularly easy to see in herbarium specimens. The leaves of S. villosum are similar to those of S. retroflexum in being usually more or less rhomboid with dentate margins, but leaves are more narrowly rhomboid than those of S. retroflexum (i.e., narrower relative to length). Specimens from the southern part of the range of S. villosum and the northern range of S. retroflexum (e.g., Malawi and Mozambique) can be difficult to distinguish in the absence of information on fruit colour (S. retroflexum has matte purple-black berries, also with strongly reflexed calyx lobes in fruit).

Solanum villosum has generally distinctly longer filaments relative to anther length than other species (except S. opacum and S. retroflexum, see descriptions of these species), although large variation can be observed within and between individuals especially in herbarium specimens. We have observed in the field that filaments elongate during flower maturation, where younger flowers in the same inflorescence can be seen with shorter filaments, and older flowers with fully extended and distinctly longer filaments (SK photo 101532). Solanum villosum has a large capitate stigma that does not generally extend much beyond the anthers, in contrast to S. tarderemotum with which it often grows in sympatry in Africa, where the style is long-exserted and often bent and the stigma is smaller.

A set of herbarium and seed bank accessions of S. villosum from Kenya have narrower leaves with entire margins, and larger, branched inflorescences with fruits in which calyx lobes do not recurve as strongly as in more typical populations, but that lack stone cells in the berries. These accessions are somewhat morphologically intermediate between S. villosum and S. tarderemotum, and could present man-made or natural introgression between the two species (see discussion of S. tarderemotum).

The origin of the tetraploid S. villosum has been the focus of previous crossing and cytological studies (e.g., Edmonds 1979a). Molecular (Poczai and Hyvönen 2011), experimental breeding (Tandon and Rao 1966a, 1966b; Venkateswarlu and Rao 1972; Rao 1978), and chromosome binding data (Bhiravamurty and Rethy 1984; Sultana and Alam 2007) suggest that the species is an ancient autopolyploid with regular bivalent formation derived from S. americanum. Edmonds (1979a) suggested that S. villosum is derived from the diploid parents S. americanum and S. sarrachoides, because these two diploids are interfertile. The hypothesis in which a native of the New World, S. sarrachoides, would have given rise to an Old World endemic is not particularly convincing, but current distributions of progenitor taxa do not necessarily have to overlap (e.g., Nicotiana tabacum L., see Chase et al. 2003). The autopolyploid origin of S. villosum, and its contribution to autoalloploid S. nigrum derived from S. villosum and S. americanum, is more likely, considering their sympatry across the entire range of S. villosum. Furthermore, S. sarrachoides is reproductively isolated from all other diploids studied thus far (Edmonds 1977), making it less likely that the species has been involved in allopolyploidy. A possible diploid progenitor could be S. chenopodioides, however, as suggested by Edmonds (1979a), with which S. nigrum and S. villosum share some of their distributions. It is much more likely, however, based on the current molecular, morphological, artificial crossing, and cytological studies that S. nigrum is an autoallopolyploid derived from a genome duplication of the tetraploid S. villosum that comprises genomes of two distinct populations of S. americanum. Morphological characters support this, in that S. americanum and S. villosum both drop their fruits without pedicels.

Edmonds (1979b) superfluously lectotypified S. nigrum var. villosum with a specimen in the Linnean herbarium (LINN 248.19), apparently not realising the name had already been lectotypified completely in accordance with the protologue by Henderson (1974) using the illustration from Hortus Elthamensis (Dillenius 1732; see Jarvis 2007).

Although Miller (1768) used Linnaean names extensively in his Gardener’s dictionary it is unlikely that he had seen Linnaeus’s (1767) 12th edition of Systema naturae where S. rubrum was first published (J.E. Dandy pers. comm. to J.M. Edmonds, quoted in Edmonds 1979b), so S. rubrum Mill. is considered a new name, rather than a citation of S. rubrum L. (nom. utique rej. prop. see Doubtful names in Sarkinen et al. 2018).

Edmonds (2007) typified the names in morelloid solanums coined by J.F. Gmelin (1791) for plants collected by Pehr Forsskål, but for some reason did not designate a lectotype for S. aegyptiacum. We here select the specimen in the Forsskål herbarium (C10013600) that corresponds to Gmelin’s description and is labelled “aegyptiacum a) fructu rubro”. Edmonds (2007) considered “S. aegyptiacum Forssk.” to be not validly published because it lacked a description or diagnosis for the species, although she included one of its varieties in the synonymy of S. memphiticum (as “S. aegypticum b) fructu nigro; foliis integris, villosissimus”). We agree that Forsskål did not validly publish the name S. aegyptiacum, but it was validly published by J.F. Gmelin.

Lamarck (1794) coined his name S. villosum without reference to Miller, and because the more commonly used name for this species was S. luteum until Edmonds (1979b) changed the name (see above), many of the new combinations and infraspecific taxa attributed S. villosum to Lamarck rather than Miller.

The name S. alatum Moench has been commonly used for densely pubescent eglandular populations of S. villosum (Edmonds 1979b). There was considerable confusion over the identity of the type of S. alatum; Edmonds (1979b) typified it with an element she in the same publication excluded from her concept of S. alatum. The name has been proposed for conservation with a conserved type (Knapp et al. 2017) to overcome this disruptive typification; the conserved type proposed by Knapp et al. (2017) is cited here. 

Solanum suffruticosum apparently was a “herbarium” name of Peter Schousboe, a Danish botanist who worked in Tangiers in the early part of the 19th century. The epithet was used and attributed to Schousboe by Willdenow (1809) who referred to plants grown in Berlin that originated from northern Africa (seeds sent by Schousboe?); we have selected the sheet in the Willdenow herbarium with annotations of the name “Solanum suffruticosum” and “Hort. bot. Berol. W.” as the lectotype for this name. Almost 50 years later, Johan Lange, working in Copenhagen, also validated S. suffruticosum, but with a “?” as to the name (“d) Solanum suffruticosum Schousb.?”, and citing a different collection from Spain (C10019275), that we here designate as the lectotype. Lange’s name is a later homonym and illegitimate.

Dunal (1813) cited a single element for his S. incertum, an illustration in Rheede von Draakestein’s Hortus Malabaricus (1690) with the common name “Nelen-Tsundja”; we here designate this plate as the lectotype for S. incertum. The accompanying description states the berries are orange.

Solanum kitaibelii and S. flavum are names coined by Schultes (1814) based on specimens of Pál Kitaibel; S. kitaibelii as a replacement for Kitaibel’s “Solanum incanum” that Schultes realised was pre-occupied (“Hr. Prof. Kitaibel nannte diese Art S. incanum; da aber Flora peruviana schon ein S. incanum hat, so nannte ich es nach meinem hochverdienten Freunde, der es zuerts in Ungern fand” – Professor Kitaibel named this species S. incanum, but there is already one in the Flora Peruviana, so I name it after my esteemed friend, who found it first in Hungary). Edmonds (1979b) cited specimens as holotypes for these names, effectively lectotypifiying them. She cited a specimen in the Willdenow herbarium (B-W04421010) as the “holotype” of S. flavum, it is more likely however, that Schultes used specimens now held in M (Schultes’s herbarium at Landshuth, where he worked at the time of description of these two taxa, is now part of M). The lectotype for S. kitaibelii has the name in Schultes’s hand (M-0166013), but the isolectotype for S. flavum does not (M-0164200).

We have designated a specimen with the exact locality as the lectotype of S. ochroleucum; the G-DC duplicate is better preserved (G00144474).
Döll’s various infraspecific taxa described in his Rheinische Flora (1843) as “Formen” (here taken as the rank of forma) are mostly new combinations of previously published epithets, but his forma angulosum is based on several cited localities in Germany. We have not been able to locate specimens for these collections, and defer typification until more in-depth searches are done. It may be that these names are based on field observations and not herbarium material.

Solanum sinaicum was described by Boissier (1849) using a collection of his own from Arabia; he did not cite a herbarium, so we are lectotypifying this name with a specimen in G (G00343377). Although the protologue mentions “baccis nigricantibus” the plants are unmistakably S. villosum; we suspect the fruit colour was cited from the dry specimen.

The spelling of Richard’s (1850) name S. plebeium has sometimes been corrected to “plebejum” (e.g., Edmonds 2012). This is not necessary and incorrect, “plebeium is the nominative neuter singular of “plebeius”, meaning “commoner” or “of the common people”.

Dunal (1852) described three varieties of S. miniatum; var. villosissimum and var. stenopetalum based on European specimens and var. subserratum based on a specimen from “Arabia”. Edmonds (1979b) discussed the identities of these varieties, but did not typify them. Dunal cited the same collection Castagne s.n. from “Montaud prope Salon” for both var. stenopetalum and var. villosissimum; we have lectotypifed var. villosissimum with the other specimen cited that does not overlap with that used for var. stenopetalum (G00144520). Material in Dunal’s herbarium at MPU (MPU091757) labelled as “Arabie” cited by Edmonds (1979b) is labelled as var. villosissimum and numbered “No. 26” and is undoubtedly original material of var. subserratum. This sheet is here designated the lectotype for this name, despite it having an annotation label with the incorrect infraspecific name. In describing S. minutiflorum he (Dunal 1852) cited two collections “Pannonia circa Bannat – (Terenbinki l.c.)” and a specimen from Richard. We have chosen the latter as the lectotype because it has two duplicates; neither Edmonds (1979b)  nor we were able to locate the other syntype. 

The names coined by the Hungarian botanist Pál Kitaibel were published post-humously by Kanitz (1863) and are largely illegitimate (some were used in earlier publications by Schultes, see above). Solanum flavum and S. canescens were published as alternative names “Solanum flavum vel canescens” and therefore have the same type; we have selected the sheet in the Kitaibel herbarium at BP with the locality on the label matching that in the protologue (“E Cottu Thurócz. v. Borsodiensi”, BP [Herb. Kit. fasc. IX: 105]).

In his Manual flora of Madeira (Lowe 1872) coined several names we here consider synonyms of S. villosum, both at the specific (S. patens) and infraspecific level (S. villosum var. velutinum, S. villosum var. laevigatum). Edmonds (1979b: 215) lectotypified var. velutinum citing the type as the “same as the holotype of Solanum villosum Lam.” Lamarck (1794) was cited along with many other works, including Linnaeus’s (1753) S. nigrum var. villosum, so arguably var. velutinum is an illegitimate superfluous name because S. nigrum var. villosum was cited in synonymy and the epithet “villosum” should have been taken up. We here re-lectotypify var. velutinum with a specimen cited by Lowe and represented in BM (BM000056199). Edmonds (1979b)  lectotypified S. patens with “Lowe 547 BM!”, but we exclude all but the sheet labelled by J.M. Edmonds as “lectotype” (BM000083209) as type material; other specimens at BM labelled “Lowe 547” have different collecting dates and localities.

Edmonds (2012) designated wild-collected material as the lectotype of S. hildebrandtii stating that in Bitter’s re-description of this species in Solana africana (1913a: 467-468) he had designated a holotype. However, Braun and Bouché clearly stated that the material they used was from cultivation, and Bitter (1913a) did not re-typify S. hildebrandtii with wild-collected material, he merely cited it as something he had seen. This typification is corrected here to neotype from lectotype.

Two un-numbered collections were cited in the protologue of S. nigrum var. aurantium (Maire 1928); Maire s.n. (coll. 4 Apr 1922) and Maire s.n. (coll. 26 Mar 1923). We have selected the duplicate in Morocco (RAB-44657) of the former as the lectotype because it has numerous duplicates. We have only been able to find a single duplicate (MPU001957) of the collection made in March 1923.

Blom (1935) cited two collections (H. Christoffersson s.n., coll. 1930; C. Blom s.n., coll. 1935) and several herbaria in his description of S. villosum var. miniatibaccatum; we have selected the GB duplicate of his un-numbered collection from 1935 as the lectotype.

The names coined by Pojarkova (1955b) and here regarded as synonyms of S. villosum were first published in Russian only (Pojarkova 1955a), but later in the same year validated with Latin descriptions and type specimen designations.
Solanum purpureilineatum (Sabnis and Bhatt 1972) was described using two specimens with two different collecting numbers from a single gathering. Rather than lectotypifiying this name we are treating this a a two sheet holotype (see McNeill et al. 2012: 14-15, Art. 8.3 Ex. 4).