Distribution

Endemic to Brazil (Bahia, Minas Gerais, Piauí, Rio de Janeiro). ^{Coutinho & Alves (2020}) also report this species from Pernambuco.

Note

^{Schumann (1886}) cited three syntypes for this name. One (Sellow 1171, B† [= F neg. no. 9563]) is apparently no longer extant. Another is known from a single specimen (Ackermann 451, BR barcode 0000005423972!). The third (Riedel 63) is comparatively well-represented in herbaria and a sheet (LE barcode 00006869!) annotated by the author of the name is designated here as the lectotype. One of us (JGS) was able to dissect a flower on this sheet and can confirm that its floral characters (i.e., unequal pistil and stamen tube lengths, and oblong petals with distinct claws and undulate margins) agree with those given in the original description. The Riedel isolectotype at Kew (K) does not have the collector’s number, but almost certainly is a duplicate of the lectotype or isolectotype now in Saint Petersburg (LE).

A Riedel specimen in Meise (BR barcode 0000005752942!) of Psychotria stachyoides Benth. (Rubiaceae) collected in Campinas, Brazil in 1825 is numbered “331” and the plant also has a ticket attached to it that reads simply “No 63.” This latter might be a field number. Given the different localities and dates (and taxa), we doubt this collection would be confused with the type of Waltheria ackermanniana.

Distribution

Endemic to Brazil (Minas Gerais).

Note

When Saunders (1995b) first proposed this new species in her unpublished thesis, she tentatively designated a collection (Saunders et al. 3167) as the type of this name, but when ^{Coutinho & Colli-Silva (Coutinho et al., 2020}a) validated Waltheria biribiriensis they designated the collection cited above as type.

Distribution

Endemic to Brazil (Alagoas, Bahia, Ceará, Minas Gerais, Pernambuco, Piauí, Rio Grande do Norte). ^{Coutinho et al. (2020}a) also cite a record for Paraíba that we have not confirmed. Collections from Rio de Janeiro are either mislabeled or represent cultivated material. A number of other records in REFLORA Virtual Herbarium (2021) appear to be misidentifications of other species allied to Waltheria ferruginea (see Saunders, 2021).

Note

The locality on the holotype label is Serra do Açuruá, which is east of the Rio São Francisco and south of Xique-Xique, Bahia. None of the other duplicates of Blanchet 2744 have this locality and most have some variation of “Utinga desertum ad fluv. S. Franc.” It is possible that the locality was miscopied on the holotype label, but this cannot be established now.

In the Gray Herbarium (GH) a duplicate of Blanchet 2744 is mounted with a different collection (Gardner 2057) of Waltheria brachypetala from Piauí: the one sheet with two collections has a single barcode hence GH barcode 00056999, p.p.

Distribution

Endemic to Brazil (Bahia, Goiás, Maranhão, Mato Grosso, Minas Gerais, Piauí, Tocantins).

Note

In the protologue of Waltheria bracteosa, ^{St.-Hilaire & Naudin (1842}) wrote “In prov. Goyaz legit Gardner (3607). - Herb. Deless.” and because there is only a single sheet in the Delessert herbarium (G barcode 00358710) we consider it to be the holotype. The date (“1841”) on the label of the holotype appears to be incorrect since Gardner was not in Goyaz in 1841 according to ^{Urban (1906}: 24). Gardner’s numbers are species-distribution, not collection, numbers but the similarities in leaf shape among the duplicates suggest that Gardner 3607 is, in fact, a single gathering and we therefore assume that the duplicates cited above are probably isotypes.

^{Schumann (1886}: 67) provisionally proposed (“ad interim salutata”) the name Waltheria regnellii (see ^{Turland et al. 2018}; Art. 36.1(a)), which was only validly published 22 years later. There are several date discrepancies among duplicates of Regnell III, 276 (the type number of W. regnellii), which lead us to believe that Regnell specimens might also have species-distribution rather than collection numbers. Fryxell (2009) equally perplexed by date discrepancies on Regnell collections came to a similar conclusion and noted that he had been unable to discover a published account of the numbering system employed by Regnell. The isolectotypes listed above all have the same date as the lectotype. The “probable isolectotypes” have dates similar to that of the lectotype or no date at all. One specimen with the type number, however, has a very different date (i.e., UPS-190278! [as “29-XI-1848”]) and we explicitly exclude it from the type of W. regnellii because the discordant date suggests that the material might not be part of a single gathering (see Turland et al., 2018; Art, 8.2). The specimen designated as lectotype (S-R-7493!) agrees best with the protologue, not only with respect to its label data but also leaf blade dimensions that coincide with the maximum measurements given in the original description.

Distribution

Endemic to Brazil (Minas Gerais, Paraná, Rio Grande do Sul, Santa Catarina, São Paulo).

Note

The protologue cites only the locality we record above in quotation marks and does not explicitly give a collector’s name or number. The specimen designated here as lectotype was collected by one of the authors of Waltheria carpinifolia and it agrees with the published description, especially in having leaves slightly scabrous above and gray tomentose below. Furthermore, the lectotype was originally in the “herbarium Richard” (now P), which according to the protologue is one of several collections the authors consulted. No other original material is known to us.

Distribution

Endemic to Brazil (Alagoas, Bahia, Minas Gerais, Sergipe).

Note

The lectotype of Waltheria cinerescens was selected from among three syntypes in Paris (P) collected by St.-Hilaire, each one of which is labeled W. cinerescens and each one of which has the locality data as given in the original description. Two of these sheets include a reference to the St.-Hilaire notebook where the collection was recorded (i.e., “Catal. B1, No 1195”). The specimen (P barcode 02273696 as image!) designated here as lectotype is the more ample of these two and best agrees with the original description by showing a few cordate leaf bases. The specific epithet often is cited incorrectly as “cinerascens.”

No original material of Astropus tomentosus was located and therefore a neotype is designated here. Astropus Spreng. was described originally as a fourmerous plant, but its author later united the genus with Waltheria as W. astropus and as noted by ^{Garcke (1890}) thereby tacitly admitted his earlier error. Apart from the merosity error, the neotype agrees with the original protologue. Garcke (1890) speculated that Sprengel hesitated to transfer A. tomentosa to Waltheria because of the earlier W. lophanthus G. Forst. (= W. tomentosa), but nothing blocked Sprengel from making a combination and as a consequence W. astropus is an illegitimate superfluous name.

A fragment in the Field Museum (F barcode F0073623F!), which was received from Paris (ex P barcode 06626945 as image! vel P barcode 06626947 as image!), is labeled “Waltheria lantanaefolia St. Hil. & N.” and “Blanchet 1677.” We do not consider this fragment to be type material of W. lantanifolia because the original description of this name does not cite a collector. Likewise, there is a Blanchet sheet in the Delessert herbarium (G-DEL [= F neg. no. 23850]) that has a ticket numbered “1677” and a printed label “Brésil (Bahia), M. Blanchet, 4e envoi, 1834” on which the name “Waltheria lantanifolia” and number “674” are written. This, too, cannot be linked to the original description. The specimen here considered to be the lectotype of W. lantanifolia has on its label the same locality as is given in the original description and the specimen also matches the morphological details presented.

Distribution

Colombia, Venezuela, Brazil (Rio de Janeiro, São Paulo).

Note

^{Schumann (1886}) associated a single collection (Riedel 64) with this name without noting where it was deposited. He did, however, examine the specimen that was at Berlin (B; see F neg. no. 9565), but presumably later destroyed in WWII. The lectotype (LE barcode 0003688!) is the only syntype whose flowers were dissected by one of us (JGS), confirming that the abaxial face of the petals are glabrous along with other details in the original description. The Kew (K) specimen is certainly an isolectotype even though it lacks a collection number: in all likelihood the number was inadvertently omitted when duplicates were distributed by Saint Petersburg (LE).

Distribution

Bolivia, Brazil (Alagoas, Bahia, Distrito Federal, Goiás, Mato Grosso, Mato Grosso do Sul, Minas Gerais, Paraná, Rio Grande do Sul, Santa Catarina, São Paulo), Paraguay, Uruguay, Argentina.

Note

Waltheria communis, W. glabriuscula, W. gracilis, and W. lanata were published simultaneously. ^{Schumann (1886}: 58) was the first to combine them under W. communis, and accordingly this name has priority over the other names when they are considered to be synonyms (^{Turland et al., 2018}; Art. 11.5, Ex. 22). Waltheria douradinha was also published at the same time as W. communis (and the other species enumerated above), but Schumann (1886: 67) recognized it as distinct albeit “incertae sedis.” ^{Cristóbal & Klein (1983}: 30) and Cristóbal (1998: 30) also considered W. douradinha and W. communis to be distinct. Saunders (1998: 280), however, first synonymized W. douradhina under W. communis.

The lectotype (P barcode 02273685!) of Waltheria communis has a label with locality data matching exactly that of the original description and an annotation (“†”) by the author of the name to indicate that he considered the specimen to represent a new species. The specimen also is ample and fertile. A St.-Hilaire specimen from Minas Gerais (SI barcode 000571!), ostensibly type material of W. communis, is excluded as type material because it has a locality (“Deserts de Minas”) different from that given in the original description.

The lectotype (P barcode 02273676 as image!) of Waltheria douradinha has St.-Hilaire’s field ticket and is annotated with the catalogue number assigned by Weddell. The isolectotypes lack field tickets. The label on the lectotype provides a locality that is almost verbatim the locality in the original description and it includes the species name published by St.-Hilaire, all written in his hand. The lectotype also agrees with the original description in having the stems ascending and inflorescence heads slightly pseudo-axillary.

The lectotype (P barcode 02273674 as image!) of Waltheria glabriuscula is the more ample of two syntypes in Paris (P). Its leaves better match the original decription than those of the isolectotype. The lectotype label is annotated “Ne correspond pas au no 673 cat D” even though there is a field ticket on the plant with that number and the letter “d.” In St.-Hilaire’s unpublished notebooks, “673 d” is a species of Gentianaceae (see http://hvsh.cria.org.br/hv?action=byCollect&book=&coll=673;&search=1). The leaves on the lectotype and isolectotype differ one from the other especially in size, but the leaves on both specimens fall within the range of variability seen in this species.

The lectotype (P barcode 02273669! [= F neg. no. 35397]) of Waltheria gracilis has the most ample inflorescence of any of the duplicates of St.-Hilaire 398 located. As indicated in the original description, its inflorescence also has a few axillary flowers in addition to terminal capitulae, a character not apparent in the isolectotypes. Additionally, one of us (JGS) dissected a flower on the lectotype in order to confirm other floral diagnostic characters.

The lecotype (P barcode 02273687!) of Waltheria lanata is an ample specimen, has an original label, and is annotated with the catalogue number assigned by Weddell. It agrees with the protologue, especially in its thick, woody xylopodium (“Radix lignosa, crassa”) with a reddish interior and leaf blades that are plicate with subacute apices and on the distal part of stems villous below and lanate and grayish-glaucescent above.

The situation described by Saunders (1995b: 466; see also her annotation on KW barcode 001001726!) where the type of Waltheria boliviensis was in a bundle of specimens that was dropped during the “Great War” and its label was replaced inadvertently with that from another specimen necessitates the designation of a lectotype (^{Turland et al., 2018}; Art. 9.3). While Saunders may have been correct in stating that the specimen in KW (KW barcode 001001726!), which is now labeled Jurgensen 121, is by virtue of its morphology the specimen from Bolivia (not Mexico) that Turczaninow used to describe W. boliviensis, it is impossible to confirm this and for all intents and purposes the specimen examined by Turczaninow was lost when the bundle was dropped. A duplicate in Geneva (G barcode 00358717!) of D’Orbigny 920, the type collection, is designated here as lectotype. Its label data agree with the protologue and the specimen agrees with the original description especially in having subcordate leaves and angular-striate peduncles below the terminal capitulae.

We assume that Schumann examined Pohl 1325, the type of Waltheria communis var. dietrichii, in Berlin (B) and that the speciman is no longer extant. The lectotype (K barcode 0000380984!) for this varietal name agrees more precisely with the original description than do the isolectotypes, notably in having the leaf blades 7 × 2 cm and inflorescence capitulae long-pedunculate. The isolectotypes in Vienna (W 0071716! and W 0071717!) each have two collector numbers [“Pohl 2169 (Diar. no. 1325)”] and are the source of the locality data that were unknown (“loco non indicato”) to Schumann. Pohl’s name is not recorded on the Kew (K) sheets, only his collection number.

The lectotype (K barcode 000380978!) of Waltheria communis var. henningsii, which was examined by Schumann, was received at Kew (K) on exchange from Berlin (B). Several field tickets link it to the original description. The specimen was selected from among several syntypes because it best illustrates the original description and is a more complete specimen with ample fertile material.

^{Schumann (1886}) cited two syntypes when he described Waltheria communis var. platyphylla: Mosén 417 and “Regnell n. III. 275 ex parte.” A collection of the former annotated by Schumann as “Waltheria communis St. Hil. γ platyphylla Schum.” is selected as the lectotype (S14-46290!). It agrees with the original description, especially in that the plant is suberect, tomentose or tomentellous, and the leaf blades are 6-7 cm long, almost as wide, and with a coarsely serrate margin. The Regnell collection, presumably another part of it (“ex parte”), was also cited by Schumann (1886) as material examined for W. communis var. henningsii.

The lectotype (F barcode F0073617F!) of Waltheria communis var. tomentella was received by the Field Museum (F) on exchange from Berlin (B). Similarly, the possible isolectotype specimen in Paris (P) also was received from Berlin (B), but for this latter specimen there are not enough data to establish unequivocally that it is a duplicate of the lectotype. One of the syntype collections (D’Orbigny 920) of this varietal name is also the collection number designated here as the lectotype of W. boliviensis. The lectotype of W. communis var. tomentella agrees with the original description in having ascending stems branched from the base, tomentellous ovate leaves with stellate hairs when mature, and 5-6 mm long sparingly ciliate stipules.

The varietal name Waltheria communis var. vulgaris is based on numerous syntypes. The lectotype (P barcode 02273681 as image!), a sheet of Blanchet 3664, is the only specimen that strictly agrees with the statement in the original description that the stems of this variety do not exceed 10 cm in length. The type locality is stated to be “in provincia Bahia.” The labels on two isolectotypes (P barcode 02273683! and W 0070967!), however, have a somewhat more precise locality (viz., “Jacobina”). In the protologue of this varietal name, ^{Schumann (1886}: 59) erred in citing the non-existent “Waltheria vulgaris St-Hil.” as a synonym.

The lectotype (F barcode F0073629F!) of Waltheria prostrata has mature cymes that clearly show the unusual four-sided inflorescences described in the protologue. The protologue also states the type locality is “Habitat in provinciis meridionalibus Brasiliae.” Curiously, the lectotype and an isolectotype in Paris (P barcode 02273678!), both received on exchange from Berlin (B), are labeled “Brasilia australis.” The material in Berlin (B; see F neg. no. 9573), which presumably was destroyed in WWII, was labeled simply “Brasilia.” Barring additional information explaining this discrepancy, we assume the type locality is as given in the protologue.

The lectotype (G barcode 00381018!) of Waltheria communis var. hirta was annotated by Schumann, one of the coauthors of the name, as “Waltheria communis St. Hil. forma [sic] nova.” A separate annotation by Hassler, the other coauthor, corrects this to “var. hirta K. Sch. et Hassler.” The lectotype, but not the isolectotype, has handwritten locality and descriptive (plant height and flower color) data that match what is in the protologue.

Only one sheet (G barcode 00381062!) of the type collection (Hassler 6747b) of Waltheria communis var. velutina was located. Schumann annotated it as “forma ad hoc incognita.” Hassler, however, wrote the published varietal name on the specimen’s label. We cannot establish that this collection consists of a single sheet, hence our lectotype designation.

The holotype (RBR barcode 00037633 as image!) of Waltheria communis var. paraibana matches exactly the unnumbered figure, a photograph, published in the protologue. The orientation of the left hand branchlet, however, was changed after publication but this appears to be nothing more than the random placement of the branchlet by a plant mounter.

Distribution

Endemic to Brazil (Bahia, Goiás, Maranhão, Tocantins).

Distribution

Endemic to Brazil (Minas Gerais). Reports of this species from Bahia probably represent misidentifications of Waltheria ackermanniana.

Note

Only one specimen labeled Waltheria erioclada was found in the de Candolle herbarium (G-DC) and we assume it is the only material de Candolle examined, hence the holotype. It agrees with the original description, especially in its oval leaves. The name is often placed in synonymy under W. indica (see e.g., ^{Coutinho, Colli-Silva & Pirani, 2020}), but this infrequently collected species shows morphological affinities to W. ackermanniana and W. glazioviana K. Schum., not to W. indica or allied species.

Distribution

Endemic to Brazil (Bahia, Goiás, Minas Gerais, Tocantins).

Note

Apparently two different gatherings were combined to form the collection distributed as Blanchet 2685: material from Serra da Jacobina (the type locality) and from Barra. The holotype (KW 001000119!) has the locality as given in the original description. More importantly, Turczaninow annotated this specimen, which was in his personal herbarium, with his new species name and the specimen agrees with the original description. The locality on the label of Blanchet 2685 in the de Candolle herbarium (G-DC [G barcode 00358718]!) is “Villa di [sic] Barra.” Barra, a locality on the west bank of the Rio São Francisco, is ca. 400 km distant from Serra da Jacobina. The de Candolle sheet therefore is not considered to be type material. The type status of a sheet in Vienna (W 0071719!) is equivocal because it does not include locality data.

Distribution

Endemic to Brazil (Minas Gerais).

Note

The lectotype (P barcode 02273702!) and the isolectotype in Montpellier (MPU barcode 016436!) of the name Waltheria ferruginea have field tickets with the number “116” (or “911”?) tied by string to the specimens. The significance of this is unknown because we have not located the relevant entry in St.-Hilaire’s notebooks (see ^{Dwyer, 1955}; ^{Pignal et al., 2013}). In any case, when ^{Coutinho et al. (2020}a) designated a lectotype they either overlooked or suppressed this detail.

^{Schumann (1886}: 55-56) considered Waltheria brachypetala to be synonymous with W. ferruginea and consequently the latter name often has been misapplied to the former and other related species. Waltheria ferruginea now is considered to be narrowly endemic and limited to the vicinity of Grão Mogol and Cristalia, Minas Gerais (see Saunders, 1995b, 2021; ^{Amorim et al., 2009}; ^{Coutinho et al., 2020}a).

Distribution

Endemic to Brazil (Bahia, Goiás, Minas Gerais, Tocantins).

Distribution

Bolivia, Brazil (Mato Grosso).

Distribution

Endemic to Brazil (Rio de Janeiro).

Note

“Glaziou n. 8723” is the only collection cited in the protologue of Waltheria glazioviana. Schumann examined material in Berlin (B) that subsequently was destroyed in WWII (see F neg. no. 9567). Nonetheless, a number of duplicates exist and the one in St. Petersburg (LE!) is designated here as the lectotype. Flowers on this specimen were dissected by one of us (JGS) and the dissection confirms that they have the thrum morphology described in the protologue.

Distribution

Mexico, El Salvador, Honduras, Nicaragua, Costa Rica, Panama, Colombia, Venezuela, Brazil (Mato Grosso). This species was relatively recently introduced into Brazil (1997 or before) where it is known from a single collection made in disturbed vegetation (^{Coutinho et al., 2019}).

Note

The lectotype (PR 3273!) of Waltheria glomerata was selected from two specimens collected by Haenke, both of which are original material and now deposited in Prague (PR). The lectotype agrees with the original description especially in having bracts equal in length to the calyx, the lateral ones lanceolate and acute, and the middle one oblong and 2-3-lobed, and most importantly the petals equal in length to the calyx.

The lectotype (US barcode 00479032!) of Waltheria rhombifolia was selected from four syntypes cited in the protologue that were stated to be in “herb. nat. C.R.” No type material was located in the Museo Nacional de Costa Rica (CR) and as a consequence the lectotype is selected from material in the Donnell Smith, or author’s, herbarium. The lectotype has two labels: one with mostly handwritten data and the number “4038” and the other with a printed distribution label that has both a distribution number and the collector’s number (i.e., “8091. Waltheria brevipes, Turcz. ... Leg. Tonduz, (n. 4038 herb. nat. Cost.)”). The specimen clearly shows the obovate -or oval- rhombic leaf blades that are noted in the original description.

The name Waltheria brevipes Turcz. has been widely misapplied to W. glomerata, but W. brevipes is considered here to be a synonym of W. preslii Walp., a species considered to be endemic to Mexico. In W. preslii the petals are twice the length of the calyx, which is distinct from the relative lengths of the petals and calyx in W. glomerata.

Distribution

Endemic to Brazil (Minas Gerais).

Distribution

Endemic to Brazil (Mato Grosso).

Distribution

A cosmopolitan weedy species found in all Brazilian states.

Note

In the present conspectus, we have adopted a narrow concept of Waltheria indica and chosen to list only the New World synonyms, which are the ones of most concern to anyone interested in the Brazilian flora. Several names that have been synonymized under W. indica in Brazil, however, are excluded in our interpretation of the species. Waltheria erioclada considered to be a synonym of W. indica by ^{Coutinho, Colli-Silva & Pirani (2020}), is recognized here as distinct (see above). Waltheria laevis D. Wolf ex Schrank and Riedlea berteroana Balb. ex DC. (as “Riedleia berteriana”), both cited by ^{Schumann (1886}: 64) as synonyms of W. americana (= W. indica), are synonyms of W. glabra Poir., a Caribbean species not known to occur in Brazil (^{Dorr, 2012}) and therefore not treated here.

Waltheria indica and W. americana were published simultaneously. ^{Brown (1818}: App. 5: 478, 484) apparently was the first to unequivocally combine the two names, and he adopted W. indica for the combined species. When the two names are considered synonyms, Brown’s choice must be adopted (see ^{Turland et al., 2018}; Art. 11.5, Ex. 22).

^{Gillis (1974}: 100-101) was the first to designate a lectotype for Waltheria indica. ^{Verdcourt (1995}: 418) independently designated the same Hermann specimen as the lectotype of this name, but his designation was superfluous. ^{St. John (1976}) and ^{Verdoorn (1981}) considered material in the Linnean herbarium (LINN 852.3 and LINN 852.2, respectively) to be the type of W. indica, but neither of the specimens they cited is original material and in both cases their designations were published subsequent to that of Gillis.

^{Gillis (1974}: 101) was the first to designate a specimen in the Linnean herbarium (LINN 852.1) as the lectotype for Waltheria americana. Subsequently, ^{St. John (1976}) discussed this specimen and a second one (LINN 852.2) in connection with the type of W. americana, but he failed to distinguish between them and therefore did not effectively designate a lectotype for W. americana, which in any case would have been a superfluous designation.

Waltheria arborescens is a superfluous name for W. indica. In the protologue ^{Cavanilles (1788}) cites both W. indica and W. americana in synonymy and he provides indirect references to the original publication of both names by citing their treatment in the Systema Plantarum (^{Reichard, 1780}), which links back to the original protologues of both Linnean names.

Waltheria fruticosa Røttb. is a nomen nudum. The author compared his plant to W. indica L. (as “Americana L.?”), but he did not provide a description or diagnosis and his text simply discussed the plant’s medicinal properties (see ^{Turland et al., 2018}; Art. 38.1(a), Art. 38.3), which is insufficient for valid publication.

The single specimen cited in the protologue of Waltheria elliptica has not been located. The collection was made in the East Indies and sent to Lamarck by Sonnerat (““Habitat in India Orientali. V.S. communicatam à D. de Lamarck, ex collectis a D. Sonnerat””). Curiously, the protologue of W. microphylla Cav. (not treated here) uses identical language to describe its type locality and the sole specimen associated with that name. This suggests there was a lapsus by Cavanilles that dates back to when both names were published simultaneously. A search of the microfiche of the Lamarck herbarium (P-LA) failed to yield material suitable for a lectotype. Similarly, ^{Garilleti (1993}: 148), in a survey of Cavanilles-related material in the Linnean (LINN) and Madrid (MA) herbaria, did not find original material of either name that can be unequivocally associated with either Sonnerat or Lamarck. Therefore, the illustration (t. 171, fig. 2) published with the original description of W. elliptica is the only original material known and designated here as the lectotype.

The protologue of Waltheria corchorifolia states “Hab. in Rio Janeiro in Brasilia. (Herb. Juss.).” The only specimen in the Herbier d’Antoine Laurent de Jussieu (P-JUSS) that agrees with these data is one collected by Dombey, which is selected here as the lectotype (P barcode 00672023 as image!).

The holotype (KW barcode 001000115!) of Waltheria surinamensis is a specimen examined and annotated by the author of the name. ^{Turczaninow (1863}) compared this new species to W. paniculata Benth. However, further study might show it to be more closely related to W. martii Colla and therefore not part of W. indica s.str.

Distribution

Colombia, Venezuela, Guyana, Brazil (Acre, Amazonas, Rondônia, Roraima).

Note

^{Bailey (1940}) designated a lectotype for Waltheria involucrata when he wrote “Material of my 1247 was compared for me in 1930 by the late N. E. Brown with the Bentham type at Kew ....” Bailey (1940) had cited the type collection (“Schomburgk no. 772”) in a previous paragraph. Inasmuch as Schomburgk 772 was the only collection associated with the name in Bentham’s protologue and because there are two specimens at Kew, one each in the herbaria assembled by Bentham and Hooker, Bailey’s mention of the “Bentham type at Kew” narrowed his designation of lectotype to a single sheet.

The text associated with “Waltheria macropoda Klotzsch ex Ri. Schomb.” states simply that it is a “perennial herb,” which does not satisfy the requirements of the ICN (^{Turland et al., 2018}; Art. 38.1) for a description or diagnosis. The plant was collected by Richard Schomburgk “Auf den Ufern des Zuruma’ (i.e., “On the banks of the Zuruma [= Suruma]”) in Guyana. Although we have not seen any material, ^{Schumann (1886}: 52) cites a Richard Schomburgk collection of W. involucrata with this same locality (i.e., “ad Suruma: Rich. Schomburgk n. 753”).

Distribution

Endemic to Brazil (Bahia, Goiás, Minas Gerais, Pernambuco, Piauí, Sergipe).

Notes

Two syntypes, Blanchet 2579 and Gardner s.n., are cited in the protologue. The former is designated here as the lectotype of Waltheria macropoda. The sheet (KW barcode 001000116!) containing the lectotype, however, includes both syntypes: two elements (“A” and “C”) correspond with Blanchet 2579 and one (“B”) with Gardner s.n. The Blanchet material agrees with the protologue: 3-4 inflorescences (capitulae) in the upper axils of young leaves, densely congested 3-bracteolate flowers, and petals twice the calyx in length. An isolectotype at Kew (K barcode 000380967!) is attributed to Moricand, not Blanchet, but the locality, collection number, and specimen agree with the lectotype. Since Moricand had material of the Blanchet collection in his herbarium, which was given to Geneva (G) in 1908, we assume the person who sent the specimen to Kew as exchange recorded Moricand’s name and not that of the collector.

In her unpublished thesis, Saunders (1995b) incorrectly considered Waltheria macropoda Klotzsch ex Ri. Schomb. (= W. involucrata) to be a valid name and treated W. macropoda Turcz. as a later homonym. She consequently proposed a replacement name derived from an annotation (viz., “Waltheria Martiana Schum.”) that Schumann attached to a specimen of Blanchet 2579 (F barcode F0073624F!). Schumann subsequently reannotated this specimen as W. macropoda Turcz. and recognized Turczaninow’s species in the Flora Brasiliensis (Schumann, 1886: 66-67) while suppressing his unpublished herbarium name. Before Saunders realized that the Schomburgk name was a nomen nudum, she annotated a number of specimens of W. macropoda Turcz. as “W. martiana.”

Waltheria macropoda is sometimes confused with W. bracteosa and W. operculata Rose.

Distribution

Endemic to Brazil (Espírito Santo, Rio de Janeiro).

Note

Three syntypes of Waltheria maritima are in Paris (P). All three were annotated by St.-Hilaire who appended a mark (“†”) following the name on each label to indicate that he considered the material to represent a new species. All three also have field tags or labels (viz., “288 S”). The lectotype (P barcode 02273710 as image!) is the specimen with the exact wording of the locality in the original description: the isolectotype labels have slight variations in wording. Additionally, the lectotype has leaves that match exactly the dimensions cited in the protologue. A fragment of Waltheria maritima in the Field Museum (F barcode F0073625F!) lacks collecting data, but presumably it was removed from the lectotype or one of the isolectotypes in Paris (P) and not Montpellier (MPU).

The text of a descriptive holographic label in Colla’s hand on the specimen designated here as lectotype of Visenia ferruginea (Fig. 1) substantially agrees with the original description published by Colla (1833). This, as noted by Fryxell (1976) in a similar context, underscores the identification of the specimen as the type of the name in question. In this case, the label text states that the specimen differs from V. scabra Colla, not that it is that species, and similar language is in the protologue. ^{Moraes et al. (2013}: 33), however, selected this specimen (Fig. 1) as the lectotype of V. scabra even though it is in serious conflict with the latter protologue and is here superseded (^{Turland et al., 2018}; Art. 9.19(c)). Visenia scabra is described as having leaves that are petiolate and round with crenate margins. The specimen designated here as lectotype of V. ferruginea, in addition to being linked to the protologue via its holographic label, has ovate rather than round leaves that are unevenly coarsely toothed, not crenate. “Melochia 1” is a tentative identification on Colla’s label.

Fig. 1 Lectotype of Visenia ferruginea Colla (= Waltheria maritima A. St.-Hil.) in the Colla herbarium (TO). The text of the holographic label in the lower-right corner substantially agrees with that of the protologue. The label on the left (“1. Melochia ...”) added by Martius provides information on the type locality, Mucuri, and suggests that the specimen was collected by Wied. Goldberg’s annotation appears to have created confusion concerning the specimen’s correct identity. The holographic label states how the specimen differs from V. scabra Colla, not that it is that species. (Reproduced with permission). Color version at http://www.ojs.darwin.edu.ar/index.php/darwiniana/article/view/953/1241 

^{Goldberg (1967}), who revised Melochia, examined this specimen and identified it as Waltheria. Unfortunately, he also wrote “Type of Visenia scabra?” on his annotation slip, which may be the source of the subsequent confusion with V. scabra (q.v.).

The lectotype (P barcode 02273712!) of Waltheria aspera is one of the three syntypes cited by ^{Schumann (1886}) who probably examined all three in Berlin (B). A possible duplicate of one syntype (“Princ. Nuwied n. 455”) was located by ^{Moraes et al. (2013}: 32) in Turin (TO), but they indicated that the specimen lacks a number. All material of another syntype (Sellow 1127) evidently was destroyed in WWII (see F neg. no. 9564). Therefore, the syntype (Glaziou 6097) found in Paris (P) is the best choice for lectotype, especially as it has the collector and number given in the protologue. The specimen also was annotated by Schumann who provided locality data not given in the protologue. The name Waltheria aspera has been widely misapplied to specimens of W. cinerescens. Moraes et al. (2013) considered W. aspera to be a synonym of W. scabra (Colla) P.L.R. Moraes & Guglianone, a combination founded on what we argue is, in fact, a species of Passifloraceae (see “Excluded and unresolved names” below).

Distribution

Mexico, Colombia, Venezuela, Guyana, Bolivia, Brazil (Alagoas, Amazonas, Bahia, Ceará, Espírito Santo, Goiás, Maranhão, Mato Grosso, Mato Grosso do Sul, Minas Gerais, Pará, Paraíba, Pernambuco, Piauí, Sergipe), Paraguay, Argentina.

Note

The protologue of Waltheria martii states simply “Communicata a cl: Martio sine nomine specifico ex Brasilia.” ^{Moraes et al. (2013}: 32), however, noted that many of the specimens in Colla’s herbarium ostensibly collected by Martius were, in fact, collected by Wied and they selected one of them as the lectotype. Moraes et al. (2013: 32) considered W. martii to be a synonym of W. indica, but they relied on an overly broad concept of the latter species. Pin flowers of W. martii have a long exserted elongate-plumose stigma whereas those of W. indica have a penicillate stigma equal to or slightly longer than the stamens.

In the protologue of Waltheria albicans, ^{Turczaninow (1858}) clearly erred when he wrote “Blanchet 2791” instead of “2691.” The former collection is the type of Eugenia flavescens var. parvifolia O. Berg (Myrtaceae). The latter collection is the type of W. albicans and the lectotype (KW barcode 001000121!), designated here, is a sheet annotated by the author of the name. The locality data visible on the specimen are not complete, which suggest Turczaninow had more than one sheet available to him when he prepared the original description. In any case, the lectotype itself agrees with the original description. There are conflicting dates on several of the duplicates of Blanchet 2691 and not all of them have dates. The conflicts, however, are probably perceived rather than real. Some dates clearly refer to the date material was received by a herbarium and not when it was collected by Blanchet.

Waltheria martii and W. albicans appear to be conspecific. In addition to being similar in gross morphology, the two species share a long exserted elongate-plumose stigma in pin flowers. Waltheria martii is the earliest validly published name for this species and thus has priority over W. albicans.

Distribution

Endemic to Brazil (Mato Grosso, Mato Grosso do Sul).

Note

This species is often confused with Waltheria polyantha K. Schum.

Distribution

Mexico, Honduras, Colombia, Venezuela, Bolivia, Brazil (Alagoas, Bahía, Ceará, Mato Grosso do Sul, Minas Gerais, Paraíba, Pernambuco, Piauí, Rio de Janeiro, Rio Grande do Norte, Tocantins), Paraguay. ^{Coutinho, Colli-Silva & Pirani (2020}) also cite records from the Brazilian states of Maranhão, Mato Grosso, and Sergipe, but we have not confirmed these reports.

Note

^{Schumann (1886}) confused this species with Waltheria macropoda Turcz.

Distribution

Endemic to Brazil (Rio de Janeiro).

Note

Two syntypes, Glaziou 3874 and Pohl 1354, are cited in the protologue of Waltheria petiolata. The former collection is designated here as lectotype. Two sheets of this collection were found and there is evidence that a third had been in Berlin (B, see F neg. no. 9570) but was destroyed in WWII. The lectotype (C barcode 10019408!) is the only sheet among the extant original material that was annotated by Schumann with the name W. petiolata.

Uncertainties in identifying duplicates of Pohl 1354 made this syntype less suitable for selection as the lectotype of Waltheria petiolata. ^{Schumann (1886}: 61) cited this syntype as “Pohl dupl. n. 1354.” The significance of “dupl.” is unclear, although it is occasionally applied to other Pohl collections in Schumann’s treatment of Waltheria. Kew (K) has two sheets of Waltheria petiolata received from Vienna (W) both of which have this number but no collector identified on either sheet (K barcode 000380974!, K barcode 000380975!). What appears to correspond to these specimens in Vienna, however, is a specimen collected by Schüch (W 0070935!) and not Pohl.

Distribution

Endemic to Brazil (Minas Gerais).

Note

Two syntypes are cited in the protologue: “Sello n. 1938, 1437” and “Martius.” It is not clear why the Sellow collection has two numbers, but both numbers are on the lectotype (F barcode F0073627F!) label and penciled on a determination slip on the no longer extant Berlin specimen (B; see F neg. no. 9572) seen by Schumann. The Berlin specimen lacks a label per se or at least one cannot be seen on the photograph. Inasmuch as the Sellow collection in the Field Museum (F) is the only original material located, it is designated here as the lectotype.

According to ^{Urban (1906}: 106), Sellow visited the type locality, Serra da Itambé, only once on 9-X-1818.

Distribution

U.S.A., Mexico, Guatemala, El Salvador, Honduras, Brazil (Alagoas, Bahia, Ceará, Minas Gerais, Paraíba, Pernambuco [including Fernando de Noronha], Piauí, Rio Grande do Norte, Rio de Janeiro, Sergipe).

Note

Waltheria rotundifolia is based on material that was cultivated in the botanical garden in Munich, of which Schrank, the author of the name, was the first director (^{Stafleu & Cowan, 1985}). According to the protologue, the seed was from Brazil (“Brasilia, unde D. de Martius semina misit. Colitur in Caldario”). The lectotype (M barcode 0209849 as image!) is the only original material found in Munich (M) that is explicitly labeled as being cultivated from seed from Brazil. This specimen also agrees with the original description.

The name Waltheria pohliana is based on two syntypes: “Regnell n. III. 274” and “Pohl dupl. n. 1337.” The specimen (W 0071731!) designated here as lectotype is one of the duplicates of the latter collection. Morphologically it matches well the material Schumann had of this collection in Berlin (B, see F neg. no. 9571), which had been acquired from Vienna (“Hrb. Musei Palat. Vindob.”) with greatly simplified locality data (“Brasilia”). The Berlin duplicate evidently was destroyed in WWII.

The lectotype of W. pohliana is labeled “2935 (d[iar]. no. 1337)” but all of the duplicates were distributed by Vienna (W) as “Pohl 1337,” which is the only Pohl number cited by ^{Schumann (1886}: 62). The probable isolectotypes are not numbered and one of them has no label data at all.

Distribution

Endemic to Brazil (Bahia).

Distribution

Endemic to Brazil (Bahia).

Note

^{Schumann (1886}: 54) cited a single collection (“Sello 188”) from “inter Victoria et Bahia.” He presumably examined it in Berlin (B), but the specimen did not survive WWII. No duplicates have been found despite an extensive search of herbaria. The only other original material is the illustration, which therefore is designated here as lectotype.

Distribution

Endemic to Brazil (Minas Gerais).

Note

When Saunders (1995b) first recognized this species as new in her unpublished thesis, she tentatively designated a collection (Arbo et al. 5086) as the type of her proposed new name, but when ^{Coutinho & Colli-Silva (in Coutinho et al., 2020}a) validated Waltheria terminans they designated the collection cited above as type.

Distribution

Endemic to Brazil (Bahia, Goiás, Mato Grosso, Tocantins).

Note

There are two specimens of Waltheria vernonioides in Stockholm (S) marked “holotype” and “isotype,” respectively. The ICN (^{Turland et al. 2018}; Art. 7.10), however, requires that a type designation be effectively published and the mere annotation of a herbarium sheet does not meet this requirement. We therefore designate the specimen (S-R-7494!) labeled “holotype” as the lectotype because its printed label includes the collection number as well as the habitat data that are given in the protologue. The possible isolectotype (S13-3153!) has two labels: a printed label with 3-V-1903 as the date (entered in ink by hand) and a penciled label with 27-IV-1903 as the date. The reason for this date discrepancy is unknown. Furthermore, neither of these two labels includes the habitat data (“In dumetis subhumidis solo argillaceo”) provided in the protologue.

Distribution

Mexico, Colombia, Venezuela, Guyana, Bolivia, Brazil (Acre, Alagoas, Amapá, Amazonas, Bahia, Ceará, Distrito Federal, Espírito Santo, Goiás, Maranhão, Mato Grosso, Mato Grosso do Sul, Minas Gerais, Pará, Paraíba, Pernambuco, Rio Grande do Norte, Rôndonia, Roraima, São Paulo, Sergipe, Tocantins), Paraguay.

Note

The lectotype (P barcode 0019001179!) of Waltheria viscosissima has a field ticket that ties the sheet to the relevant entry in St.-Hilaire’s notebooks and the locality written on its label agrees with the information given in the protologue. The sheet, amply fertile, also matches the range of leaf sizes provided in the original description. Other duplicates of St.-Hilaire 1452 have less documentation or one or more leaves larger than described. Although one of the two Paris (P) and the sole Montpellier (MPU) isolectotypes of W. viscosissima lack the collector’s number, they have the same label data as the lectotype: the Montpellier specimen even has the same handwriting. The fragment at the Field Museum (F) was removed from a St.-Hilaire specimen in Paris (P), but there is no other information now associated with this fragment.

The lectotype of Waltheria hirsuta (PR 3271A!) is the more complete of two Haenke sheets that are original material and now in Prague (PR). It agrees with the protologue except the petals are undoubtedly yellow, not purple (“purpurea”). Almost certainly Presl who never saw the plants in the field, mistook the dried petal color for the color of living flowers.

Note

A careful comparison of the published protologue of Visenia scabra with the descriptive holographic label on the specimen in the Colla herbarium (TO) designated here as lectotype shows that their texts are virtually identical. Thus, this specimen, which also is labeled “3. Melochia ...,” is the only material suitable for a lectotype of V. scabra. ^{Goldberg (1967}: 354), who revised Melochia, stated that the description of V. scabra was “inadequate for a determination” despite having annotated this specimen in 1960 as Turneraceae (= Passifloraceae) and “Type of Visenia ferruginea?” ^{Moraes et al. (2013}) designated the specimen that we consider to be the type of V. ferruginea (see “Note” above sub Waltheria maritima) to be the lectotype of V. scabra. They seem to have relied on Goldberg’s annotations on these specimens rather than original material to come to their conclusions. Goldberg annotated the specimen labeled “1. Melochia” as “Type of Visenia scabra?” and the one labeled “3. Melochia” as “Type of Visenia ferruginea?” In any case, the choice of lectotype for V. scabra made by Moraes et al. (2013) must be superseded because it is in serious conflict with the protologue (^{Turland et al., 2018}; Art. 9.19(c)).

Note

The lectotype, also labeled “4. Melochia,” agrees with Malvastrum americanum based on its stem vestiture, leaf shape and venation, and floral epicalyx. ^{Moraes et al. (2013}) cite a second Wied collection (Brazil, Bahia, Rio de Bellome [= Rio Jequitinhonha], 1816, Prinz zu Wied s.n., BR barcode 0000013306403 as image!), which also is M. americanum.

Previously, ^{Goldberg (1967}: 354) excluded Visenia spicata from Melochia and stated simply that it “belongs to the Malvaceae.” ^{Moraes et al. (2013}: 32) were also unsuccessful in resolving the identity of this name.

Note

This name previously was considered to be a synonym of Waltheria indica L. (see e.g., ^{Jarvis, 2007}: 927). However, the only original material (Jarvis, 2007: 927) is the tip of an inflorescence of Melochia spicata and it is designated here as the lectotype.

Note

An original pen and ink drawing for this species is preserved in the Biblioteca Nacional do Brasil, Rio de Janeiro (cat. no.: mss1198656_008; see: http://objdigital.bn.br/acervo_digital/div_manuscritos/mss1198656/mss1198656_008.jpg).

The description is inadequate to identify this species. Moreover, while the description mentions sessile flowers, the original drawing (and published plate) shows pedicillate flowers.

Note

An original pen and ink drawing for this species is preserved in the Biblioteca Nacional do Brasil, Rio de Janeiro (cat. no.: mss1198656_010; see: http://objdigital.bn.br/acervo_digital/div_manuscritos/mss1198656/mss1198656_010.jpg).

The description is inadequate to identify this species and describes a plant with solitary flowers. Solitary flowers do not occur in Waltheria. They are usually in pairs within cymose clusters.

Note

An original pen and ink drawing for this species is preserved in the Biblioteca Nacional do Brasil, Rio de Janeiro (cat. no.: mss1198656_009; see: http://objdigital.bn.br/acervo_digital/div_manuscritos/mss1198656/mss1198656_009.jpg).

The description is inadequate to identify this species because it details a plant that is pentagynous or with five styles. Waltheria species have a solitary style.

Note

The published Vellozo plates were prepared and issued posthumously and there is no evidence that Vellozo ever saw them. Original pen and ink drawings that serve as the basis for these published plates, however, are archived in the Biblioteca Nacional do Brasil, Rio de Janeiro and are thought to be original material.

^{Urban (1883}) considered Waltheria terminalis to be a synonym of Turnera ulmifolia var. cuneiformis (Juss. ex Poir.) Urb. While ^{Arbo (2005}) did not mention the Vellozo species, she did accept Urban’s variety but at species rank.