A peer-reviewed open-access journal PhytoKeys 181: 49-64 (2021) & gitesee ate ore $¢PhytoKeys https:/ / Pp hyto keys -pen soft.net Launched to accelerate biodiversity research Clarifying the nomenclatural history of Tovomitopsis, a Brazilian endemic genus of Clusiaceae Lucas C. Marinho!?, Pedro Fiaschi?, André M. Amorim‘, Volker Bittrich® | Universidade Federal do Maranhio, Departamento de Biologia, Séo Luts, MA, Brazil 2. Universidade Es- tadual de Feira de Santana, Programa de Pés-Graduagao em Botinica, Feira de Santana, BA, Brazil 3. Uni- versidade Federal de Santa Catarina, Departamento de Botanica, Florianépolis, SC, Brazil 4 Universidade Estadual de Santa Cruz, Departamento de Ciéncias Bioldgicas, Ilhéus, BA, Brazil § Rua Mario de Nucci, Campinas, SP Brazil Corresponding author: Lucas C. Marinho (Ic.marinho@ufma.br) Academic editor: Manuel Lujan | Received 27 June 2021 | Accepted 10 August 2021 | Published 1 September 2021 Citation: Marinho LC, Fiaschi PR, Amorim AM, Bittrich V (2021) Clarifying the nomenclatural history of Tovomitopsis, a Brazilian endemic genus of Clusiaceae. PhytoKeys 181: 49-64. https://doi.org/10.3897/phytokeys. 18 1.70745 Abstract Tovomitopsis Planch. & Triana is a Brazilian Atlantic Forest endemic genus composed of two species: T. paniculata (Spreng.) Planch. & Triana and T’ saldanhae Engl. An investigation was conducted to clarify the nomenclatural history of Zovomitopsis. We report the results of this investigation, provide an updated description of the genus, and propose lectotypes for 7’ paniculata and its synonyms: Tovomita foliosa C.Presl and Tovomita paniculata Cambess. We also propose lectotypes for 7! saldanhae and for the new synonym Clusia angustifolia Engl. Keywords Atlantic Forest, lectotype, Malpighiales, Neotropics, South America Introduction Tovomitopsis Planch. & Triana is a Brazilian endemic genus currently composed of two species: 77 paniculata (Spreng.) Planch. & Triana and 7’ saldanhae Engl. Both species occur in preserved remnants of Atlantic Forest in southeastern Brazil (Marinho 2021). Tovomitopsis was proposed in 1860 (Planchon and Triana 1860) as a replacement name Copyright Lucas C. Marinho et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 50 Lucas C. Marinho et al. / PhytoKeys 181: 49-64 (2021) for the illegitimate Bertolonia Spreng. (non Bertolonia Raddi 1820) and to accommo- date presumably tetramerous flowered species. Together with Chrysochlamys Poepp. and Tovomita Aubl., these three genera were known by Planchon and Triana (1860: 225) as Les Tovomitées, being differentiated from each other especially by the arrange- ment of sepals on the floral bud: in Chrysochlamys and Tovomitopsis the outer sepals are smaller, exposing the inner sepals in bud, while in 7ovomita the outer sepals are larger, covering the inner sepals and petals. Moreover, Planchon and Triana (1860) indicated that aril anatomy could be useful to differentiate, or at least, better circumscribe Les Tovomitées, but surprisingly this topic has not yet been further investigated. Although some recent studies still indicate floral merosity as relevant to distinguish Tovomitopsis from Chrysochlamys (e.g. Hammel 2010), the latter includes species with four or five petals (e.g. Hammel 1999; Martinez y Pérez et al. 2015), and a clear mor- phological distinction between these two genera is yet missing. Taxonomic treatments and checklists carried out in Mexico (Martinez y Pérez et al. 2015), Central America (e.g. Hammel 1999), northern South America (Kearns 1998), where Chrysochlamys is distributed, and the Brazilian Atlantic Forest (Oliveira-Filho 2006) considered these two genera as congeneric. Molecular phylogenetic evidence shows Tovomitopsis in a politomy with Dysto- vomita and the rest of Clusieae, and thus not very closely related to Chrysochlamys despite their gross morphological similarity (Marinho et al. 2019). Pollen morphology and aril anatomy (Planchon and Triana 1860; Hammel 1999; Stevens 2007; Marinho et al. 2019) have been suggested as promising to distinguish these two genera, but were so far gathered from only a few species of Chrysochlamys. The presence of resin glands in the anther dorsal region of Tovomitopsis could be a synapomorphy of the genus, and the absence of a pistillode in staminate flowers of Chrysochlamys could be also relevant to distinguish these genera (Bittrich and Marinho pers. com.). Tovomitopsis consists of dioecious small trees or shrubs with prop roots and yellow- ish viscous exudate. The opposite leaves are petiolate, entire, chartaceous or coriaceous, with numerous closely arranged veins. The flowers have two pairs of sepals, the outer ones being smaller than the inner ones, and two pairs of whitish petals. Staminate flow- ers have yellow subclavate resiniferous stamens and a pistillode; pistillate flowers have staminodes similar to the stamens and a green-yellowish pistil with expanded stigmas. The fruits are green fleshy capsules that expose seeds with an orange vascularized aril when ripe (Bittrich 2003; Stevens 2007). Although Yovomitopsis includes only two species, the genus has a long taxo- nomic history (see Hammel 1999), with several species floating among the three genera of Les Tovomitées (sensu Planchon and Triana 1860). The type species of the genus, Jovomitopsis paniculata, was described two hundred years ago, but a few nomenclatural issues remain to be addressed. Here, we clarify the nomenclatural history of Zovomitopsis, provide an updated description for the genus, and propose lectotypes for T’ paniculata and its synonyms: Tovomita foliosa C.Presl and Tovomita paniculata Cambess. We also propose lectotypes for 7! saldanhae and its new syno- nym, Clusia angustifolia Engl. Clarifying the nomenclatural history of Tovomitopsis (Clusiaceae) 51 Material and methods This study is based on the analysis of the protologues of 7ovomitopsis names and some of its synonyms, on visits to historical collections in Europe (B, K, M, P, W) and the Americas (A, GH, NY, R, RB; herbaria acronyms according to Thiers 2021), and by analyzing specimens from virtual herbaria. Data on collectors and botanists were ac- cessed in the Taxonomic Literature II website (Stafleu and Cowan 1976-1988). All no- menclatural decisions follow the International Code of Nomenclature for algae, fungi, and plants (Turland et al. 2018). Nomenclature and discussion Tovomitopsis was proposed by Planchon and Triana (1860) as a replacement name for Bertolonia Spreng. [with just one species, B. paniculata, which was initially attributed to Chenopodiaceae (as “Chenopodieae”)], a later homonym of Bertolonia Raddi (Mel- astomataceae). Along with the newly transferred 7! paniculata they described five addi- tional new species, all six of which they felt could be distinguished from Chrysochlamys and Tovomita. Bertolonia paniculata Spreng. was based on a pistillate specimen accord- ing to the illustration provided by the author (see Fig. 1, 1-4b). Sprengel (1821) did not mention either location, number or collector name for this collection. In the “Flora Brasiliae Meridionalis”, edited by Auguste de Saint-Hilaire et al., Jacques Cambessédes (1828) used the same epithet “paniculata” when he published the new species Tovomita paniculata Cambess. ‘This binomial is sometimes mistak- enly interpreted as a new combination for Bertolonia paniculata Spreng. However, Cambessédes is clearly indicated as the author of the Guttiferae monograph at the end of the treatment, and the } sign was used to indicate a new species through- out “Flora Brasiliae Meridionalis”. The complete description of Tovomita paniculata that Cambessédes (1828) provided included stamens, pollen grains and gynoecium, indicating that he studied both staminate and pistillate specimens. The protologue contains the following statement: “in sylvis primaevis prope vicum Aguassu, haud longé ab urbe Rio de Janeiro. Florebat Februario”, and an illustration of a branch with many flowers, a feature found only in staminate specimens, while the illustration details show a pistillate flower and the detail of an ovary, as well as stamens with well- formed anthers (see Fig. 2). Cambessédes (1828) did not indicate a type collection, but one specimen housed at P that was collected by Saint-Hilaire is probably the original material. The specimen P00093861 bears a label indicating the same loca- tion as in the protologue. Presl (1834) proposed Yovomita foliosa C.Presl, as a new species and provided a detailed description and an illustration, but again without indication of the material he used. The author cited only “Habitat in Brasilia ad Rio de Janeiro”. Although he de- scribed the species as a Jovomita, the pair of outer sepals not covering the inner sepals and other floral parts allow us to recognize a Tovomitopsis paniculata specimen in the 52 Lucas C. Marinho et al. / PhytoKeys 181: 49-64 (2021) tre. Cnfd 2.5 Chi el Or de Li ie, Figure |. Lectotype (1—4b) of Bertolonia paniculata Spreng. (= Tovomitopsis paniculata (Spreng.) Planch. & Triana) published by Sprengel (1821) in “Neue Entdeckungen im ganzen Umfang der Pflanzenkunde II”. Clarifying the nomenclatural history of Tovomitopsis (Clusiaceae) Bs) illustration. In 1860, Planchon and Triana described Tovomitopsis and included, in ad- dition to Tovomitopsis paniculata (Spreng.) Planch. & Triana (= Bertolonia paniculata Spreng.), five new species (currently placed in Chrysochlamys), and justified this deci- sion based, in part, on the number of floral parts: while Zovomitopsis was circumscribed to include tetramerous flowers, Chrysochlamys retained the species with pentamerous flowers. In that work, the authors indicated two specimens for 7) paniculata: “Brésil, Rio-de-Janeiro (Aug. de Saint-Hilaire; Sellow)”. In the “Flora brasiliensis”, Engler (1888) provided a description and an illustration of Tovomitopsis paniculata citing eight collections, all of which were collected in Rio de Janeiro state: 1 — Habitat in Brasiliae prov. Rio de Janeiro: Sello in herb. reg. Berol.; 2 and 3 — Glaziou n. 7429, 12466; 4 — in silvis primaevis pr. Aguasta: St-Hilaire; 5 — in silvis umbrosis Capivary: Riedel n. 1318 in herb. Petrop.; 6 — in Serra de Friburgo: Saldanha n. 7044; 7 — in Serra de Tingua: Saldanha n. 5313; 8 — in Serra dos Orgios: Riedel — Flor. Febr. The author also described two new species in the genus: Tovomitopsis spruceana Engl., based on specimens from the Peruvian Amazon, and Tovomitopsis saldanhae Engl., based on specimens from Rio de Janeiro state. The material used in the original description of Tovomitopsis paniculata was not mentioned or indicated by either Planchon and Triana or Engler, but the protologue and subsequent publications provided two important clues to find the type: i) both the description of the pistil and the illustration point to a pistillate specimen; ii) sev- eral points indicate that the type was collected in Rio de Janeiro state. For the locality of his Tovomita paniculata, Cambessédes (1828) mentioned the village of Aguassu, which we believe to be Nova Iguacu (before spelled as /euassi), today a municipality in the metropolitan region of Rio de Janeiro city. Planchon and Triana (1860) cited Rio de Janeiro as the only locality of Zovomitopsis paniculata, and also all specimens cited by Engler (1888) came from the state of Rio de Janeiro. Both Planchon and Triana (1860) and Engler (1888) cited specimens collected by Friedrich Sellow and Auguste de Saint-Hilaire as the basis for their respective descriptions. Sellow col- lected in the state of Rio de Janeiro between 1814 and 1821 (Stafleu and Cowan 1976-1988) and most of his specimens were deposited in B, but there are additional specimens originally from the Miiller-Sprengel herbarium which were purchased by B in 1890 (Stafleu and Cowan 1976-1988). We speculate that Kurt Sprengel, the German botanist who first described Tovomitopsis paniculata (as Bertolonia panicu- lata), likely studied one or more specimens collected by Sellow rather than the ones collected by Saint-Hilaire. During a visit to European herbaria in 2016, a search at B was made for specimens assigned to the names Bertolonia paniculata, Tovomita paniculata, and Tovomitopsis paniculata, but none were found. Likewise, there are no Macbride negatives of such specimens in the Chicago Field Museum. Two relevant Sellow specimens were located in K (K001231050, image seen in Reflora Virtual Her- barium 2021), with the ¢ symbol indicating that it is a staminate specimen, and US (01882513) herbaria, both with floral buds; however, it is uncertain that these 54 Lucas C. Marinho et al. / PhytoKeys 181: 49-64 (2021) TOVOMITA paniculata. Tuanpin ded £ Figure 2. Illustration of Tovomita paniculata Cambess. published by Cambessédes (1828) in “Flora Brasiliae Meridionalis’. Clarifying the nomenclatural history of Tovomitopsis (Clusiaceae) 55 specimens are duplicates of a presumed type specimen housed in B. Therefore, we selected the illustration provided by Sprengel (1821) as the lectotype of Tovomitop- sis paniculata, as this is the only unambiguous original material known to us. We also choose the illustration provided by Presl (1834) as the lectotype of Tovomita foliosa; and followed Planchon and Triana (1860) and Engler (1888) by consider- ing 7’ paniculata Cambess. as an heterotypic synonym of Tovomitopsis paniculata, rather than a new combination. Tovomitopsis paniculata (Spreng.) Planch. & Triana, Ann. Sci. Nat. Bot., Sér. 4, 14: 262. 1860. = Bertolonia paniculata Spreng., Neue Entdeck. Pflanzenk. II: 110, t. I. 1820 (“1821”). = Tovomita paniculata Cambess., Fl. Bras. Merid. (quarto ed.) 1(8): 315, pl. 64. 1828. Type. lectotype (designated here), [Brazit: Rio de Janeiro] in sylvis primaevis prope vicum Aguassu, haud longé ab urbe Rio de Janeiro. Florebat Februario [1816-1821, A. De Saint-Hilaire s/n] (P! P00093861; isolectotypes: MPU 2-sheets MPU014277, MPU014278). (Fig. 3) = Tovomita foliosa C.Presl, Symb. Bot. (Presl) ii(7). 20. tab. 66. 1834 (1833). Type. lectotype (designated here), illustration in Presl (1834: Tab. 66). (Fig. 4) Type. lectotype (designated here), illustration in Sprengel (1821: Tab. I). (Fig. 1, 1-4b) Notes about Tovomitopsis saldanhae Tovomitopsis saldanhae was described by Engler (1888) based on specimens from the Serra dos Orgaos (Rio de Janeiro state) as “Habitat in Brasiliae provincia Rio de Janeiro, in Serra dos Orgdos ad Theresopolim: J. de Saldanha n. 6777, 6780, 6781, Glaziou n. 13576 in herb. Eichler’. His description and the illustration point to the use of both staminate and pistillate specimens. In the same contribution, Engler also described Clusia angustifolia Engl. based on Saldanha 7335, which was collected in the same locality of 7. saldanhae. The specimen (pistillate) clearly matches with specimens of 7. saldanhae, especially the oblanceolate leaves with dark dots below, four petals and two series of resiniferous staminodes with subapical antherodes. As the specimen of Clusia angustifolia deposited in B was destroyed during World War II in 1943, we select a duplicate at R as the lectotype. Tovomitopsis saldanhae was later transferred to Chrysochlamys as C. saldanhae (Engl.) Oliveira-Filho, but Oliveira-Filho (2006) did not mention or indicate a type. Chrysochlamys saldanhae is now a synonym of T! saldanhae (Marinho 2021). Among the 7’ saldanhae syntypes, we chose A. Glaziou 13576 (P01901232, Fig. 5), which is the best preserved specimen housed at P, as the lectotype. Lucas C. Marinho et al. / PhytoKeys 181: 49-64 (2021) penlasai WBiuAdoo sas THE NEW ton i ‘A ) = “a TANICAL, GARDEN pa fh No, oe a > = OS “Possible TYPE ioeertolevicn Panicula tea Seren eue Eutdeck. Phlauzent, 2: no)920 Conf Bent sh Aer Cuguey | SiS eee tS | iy NO) “aie ae ae | | Be Je Pa -'S —— a AN Gooey | Be cael | ese area 4grE, Ec SCPOPD | oe tte bal } spn eA tog $ = a ee / : ie i in f+ et Ce ee a ae ara ; BRESIL. — Province de Rio DE JANEIRO. | IF, Bo & & Jo Bee Voyage d’Auguste de Saint-Hirare, de 1816 4 1824. eo | Lee catal. D7 | Figure 3. Lectotype of Tovomita paniculata Cambess. (P00093861) housed at P. Clarifying the nomenclatural history of Tovomitopsis (Clusiaceae) 57 Geb OF ast | asaie Say WH LEN g Li ferro Wil Jos. Shatit se. Figure 4. Lectotype of Zovomita foliosa C.Presl (= Tovomitopsis paniculata (Spreng.) Planch. & Triana) published by Presl (1834) in "Symbolae botanicae, sive, Descriptiones et icones plantarum novarum aut minus cognitarum ". 58 Lucas C. Marinho et al. / PhytoKeys 181: 49-64 (2021) uséum i P01901232 . pepiesal }uBuAdOS a = THE NEW YORK BOTANICAL GARDEN > Negative Noe S/OG Specimen Noe Gla foun SZFOL 0123 x Zs ; 7 ¢ =—_ pee | ay No et | ISOTYPE HERB. MUS. PARIS. SYN TYPE 2 de “Teveni lopsis sin | dake rise Ena. LEEPD ; a [S/ +NG jhe bts. (25 : £51. “ 3 6 [ef ,