Fridericia arthrerion (Mart.) L. G. Lohmann

Kaehler, Miriam & Lohmann, Lúcia G., 2022, Taxonomic revisions in Fridericia (Bignonieae, Bignoniaceae) II: the “ Neomacfadya ” clade, Willdenowia 52 (2), pp. 247-271 : 249-252

1. Fridericia arthrerion (Mart.) L. G. Lohmann in Ann. Missouri Bot. Gard. 99: 433. 2014 ≡ Bignonia arthrerion Mart. in Flora 24(2, Beibl.): 47. 1841 ≡ Distictis arthrerion (Mart.) DC., Prodr. 9: 191. 1845 ≡ Arrabidaea arthrerion (Mart.) Baill., Hist. Pl. 10: 28. 1888. – Lectotype (designated as “ holotype ” by Lohmann in Lohmann & Taylor 2014: 433): Brazil, Mato Grosso, Cuiabá, 1833 [s.d.], Herb. Fl. Bras. 515 (= A. L. P. da Silva Manso 18) (M accession code M-86391! [and photo F negative no. 20437!]; isolectotypes: BR barcode BR0000008804723 [image!], G-DC barcode G00014139 [image!], G-DC barcode G00133574 [image!], HAL barcode HAL0116227!, K barcode K000449144!, L barcode L003554 [image!], MO accession code MO-2005583!, NY barcode NY00313130!, P barcode P00468550!, W accession code W-0057658 [image!]). – Fig. 1A–E, 2.

Notes — The protologue of Bignonia arthrerion does not specify who the collector is or the collecting number or the date of collection. It only provides the number 515 from the Herbarium florae brasiliensis . We found several specimens with this herbarium number, including two specimens in BR and M, where most of Martius’s types are deposited. The original set of gatherings used to produce the Herbarium florae brasiliensis ( Martius 1841) has a unique label on its specimens (see Moraes & al. 2014 for a discussion on this subject), which is lacking on most specimens found, except from that in BR. The specimen BR0000008804723 has one label containing the gathering information and number 18, plus a label with the original Martius Herbarium typography. In this label, Martius annotated information about the collector (da Silva Manso), locality of collection and the reference to “Hbr. Fl. Br. n. 515.” When Candolle (1845) published the new combination Distictis arthrerion , he mentioned the specimen deposited in the Herbarium florae brasiliensis citing the same information provided in the label of the specimen at BR. Indeed, Urban (1906: 112) indicated that all gatherings of A. L. P. da Silva Manso were in Martius’s herbarium held in BR, with the duplicates distributed from there. Because there is no indication to the one specimen used by Martius in the protologue, there is a need to select a lectotype. However, Candolle (1845) did not mention the specimens he saw as the type; consequently, he did not designate a type (Art. 7.11, Turland & al. 2018). Besides, Lohmann (in Lohmann & Taylor 2014) used the term “ holotype ” to define the type status of the specimen in M. Following Art. 9.10 ( Turland & al. 2018), we are correcting the use of the term “ holotype ” to lectotype.

= Bignonia hispida DC., Prodr. 9: 152. 1845. – Lectotype (designated here): Brazil, Mato Grosso, Cuiabá, 1832, A. L. P da Silva Manso 71 (G-DC barcode = G00133358 [image!]; isolectotypes: G-DC barcode G00133270 [image!], G-DC barcode G00133371 [image!]).

Notes — In the protologue of Bignonia hispida, Candolle (1845) described the flowers and fruits. There are four exsiccates in G-DC numbered from one to four that may have been used by Candolle to describe this species. In each of the sheets G00133265 (numbered as 2, with flowers) and G00133358 (numbered as 3, with flowers and fruits), there is a label with gathering information, collector and the numbers 70 and 71, respectively. In the specimen G00133271 (numbered as 1, with fruits) there is no label, while the specimen G00133270 (numbered as 4, with flowers) only includes a label with the name B. hispida handwritten (not by Candolle) and no gathering information. Because the specimen G00133358 includes flowers, fruits and the gathering information, we select it here as the lectotype. We do not consider the specimen G00133265 as an isolectotype because the collecting number is different, clearly indicating that it is not a duplicate.

Bignonia pearcei Rusby in Mem. Torrey Bot. Club 6: 100. 1896 ≡ Arrabidaea pearcei (Rusby) K. Schum. ex Urb. in Repert. Spec. Nov. Regni Veg. 14: 302. 1914 ≡ Fridericia pearcei (Rusby) L. G. Lohmann in Ann. Missouri Bot. Gard. 99: 442. 2014, syn. nov. – Lectotype, first step (designated as “ holotype ” by Gentry 1977b: 40): NY; second-step (designated here): Bolivia, La Paz, Guanai-Tipuani, Apr–Jun 1892, A. M. Bang 1391 (NY barcode NY000313146; isolectotypes: A barcode A00091987 [image!], E barcode E00259235 [image!], F accession code F-163739 [image!], G barcode G00009125 [image!], G barcode G00009126 [image!], K barcode K000403832!, M – barcode M0086363!, MICH barcode MICH1115826 [image!], MIN barcode MIN190896 [image!], MO accession code MO-2005581!, NY barcode NY01404746!, US accession code US-47713 [image!], US accession code US-47714 [image!], US accession code US-1322423 [image!]).

Notes — In the protologue of Bignonia pearcei, Rusby (1896) listed two gatherings. Bang 1391 is the first one listed, followed by a statement “Also collected by Pearce at Palta” without mentioning the collecting number or the herbarium where this specimen was deposited. We considered the differentiation of these two sentences as an indication of the type. Indeed, Gentry (1977b) indicated the gathering Bang 1391 in NY as holotype; however, he did not specify which of the two sheets deposited in NY he was referring to. NY00313146 has a stamp from the Columbia College Herbarium, where Rusby worked and where the first and better-quality set of Bang’s gatherings was housed ( Rusby 1893: 2). This specimen also has a printed label indicating “ Bignonia pearcei sp. n. ” and Gentry’s handwritten annotation indicating it as holotype. We select this specimen here as the second-step lectotype. “ Fridericia arthrerion” (Cham.) L. G. Lohmann in Forzza & al., Cat. Pl. Fung. Brasil 1: 764. 2010. Designation not validly published (Art. 41.1).

Morphological description — Lianas (sometimes scandent shrubs), high climbing. Branches terete to subquadrangular, striate and lenticellate, grey (young branches black) with ferruginous trichomes evident when dry, pubescent to villous, with simple trichomes; without interpetiolar ridge, with interpetiolar glandular fields; prophylls of axillary buds triangular, not apiculate, 1.1–2.3 mm long. Leaves 2(or 3)-foliolate; petiole 1.2–5.2 cm long, puberulous to villous, with simple trichomes; petiolules in two lengths, central 1–5.6 cm, lateral 0.5–2 cm, villous, with simple trichomes; blade chartaceous, smooth, margin revolute, central leaflet ovate to elliptic, 11–18.5 × 2.6–4 cm, base rounded, apex long apiculate, lateral leaflets ovate, 9–25 × 4.6–7.5 cm, base rounded, subcordate to truncate, apex acuminate, adaxially not vernicose, puberulous to velutinous or villous (sometimes hispid), with simple trichomes throughout, abaxially pubescent to villous, with simple and glandular peltate trichomes throughout, venation palmate actinodromous basal, secondary veins raised, tertiary veins raised, without pocket and trichome tuft domatia. Inflorescences terminal and axillary, thyrse with bud abortion, with 3–5 orders, first-order peduncles 1–4.5 cm long, second-order 1.5–2.1 cm, third-order 0.3–1.1 cm, fourth-order 0.1–0.4 cm, fifth-order c. 0.2 cm, pubescent to villous, with simple and glandular capitate trichomes; bracts linear, c. 4 mm long, caducous; bracteoles linear, c. 2 mm long, caducous; pedicels 0.3–0.5 cm long. Calyx tubular to campanulate, not costate, truncate, 0.4–0.7 × 0.3–0.4 cm, glabrescent to velutinous, with simple and glandular capitate trichomes shortly stalked, chartaceous, pink to dark red, without glands. Corolla infundibular, slightly curved, furrowed, 2.3–3.5 cm long, 0.6–1.2 cm wide at tube mouth, externally villous, without glands, lobes 0.5–0.8 × 0.6–0.7 cm, margin rounded and undulate, lilac to dark red. Androecium with all stamens included; longer filaments 1.6–1.7 cm long, shorter filaments 1.1–1.4 cm; stamin- ode 4–5 mm long; anthers 2–3 mm long, connectives not protruding (sometimes protruding c. 0.1 mm). Gynoecium with ovary cylindric, 2–2.5 × 0.7–1 mm, angled, lepidote, style 1.4–1.7 cm long; stigma lanceolate; nectar disk annular and smooth under ovary, c. 2 × 2.4 mm. Fruit linear, flat, margins slightly raised, central ridge slightly raised, valves coriaceous, rough, 24–36 × 1–1.4 cm, villous becoming glabrescent, with simple and glandular capitate long-stalked trichomes; septum coriaceous. Seeds oblong, body semicircular (irregular), 0.8–0.9 × 1.5–2.6 cm, wings hyaline, 0.8–1 cm wide, margins crisped.

Phenology — Produces flowers from December to May. Fruits were collected in May, August, September and November.

Distribution and habitat — Fridericia arthrerion is distributed through W Amazonia. This species occupies dry and wet Amazonian forests and ecotone areas between the Cerrado and Chaco. It is distributed through Bolivia (El Beni, Cochabamba, La Paz, Pando and Santa Cruz), Brazil (Acre, Mato Grosso and Pará), Colombia (La Guajira), Ecuador (Napo, Pastaza and Sucumbios) and Peru (Amazonas, Cuzco, JunÍn, Loreto, Madre de Dios, Pasco, Puno, San MartÍn and Ucayali).

Conservation status — Fridericia arthrerion is categorized as Least Concern (LC) based on its Extent of Occurrence (1,987,141 km 2) and Area of Occupancy (92,500 km 2).

Remarks — Fridericia arthrerion can be easily recognized by its tubular and short calyx, unique features among the species of the “ Neomacfadya ” clade, as well as the pubescent to villous branches and villous fruits that become glabrescent. Vegetative material is often confused with F. nigrescens (Sandwith) L. G. Lohmann due to the puberulous to velutinous or villous leaflets, usually becoming black when dry. However, F. nigrescens has branches and leaflets covered by dendritic trichomes, a synapomorphy of the “Sampaiella” clade, within which it is placed ( Kaehler & al 2019). In contrast, F. arthrerion lacks dendritic trichomes and shows exclusively simple trichomes instead, like the remaining species of the “ Neomacfadya ” clade. Fridericia nigrescens further differs from F. arthrerion by the long petioles (4–11 cm vs 1.2–5.2 cm in F. arthrerion ), bullate leaflets (vs smooth in F. arthrerion ) and strongly asymmetric base of the lateral leaflets (vs rounded, subcordate to truncate in F. arthrerion ). Fridericia arthrerion can also be confused with F. egensis (Bureau & K. Schum.) L. G. Lohmann ; see discussion provided under F. egensis .

Gentry & Grose (2007) indicated that Arrabidaea arthrerion (≡ Fridericia arthrerion ) may occur in Suriname. This observation was based on Bureau & Schumann (1896: 51), who mentioned a specimen in BR that was wrongly attributed to Martius and was likely collected by Wullschlaegel. Neither Bureau & Schumann (1896) nor Gentry & Grose (2007) were able to find this specimen, nor were we able to locate it. This record seems to be misidentified considering the occidental Amazonian distribution pattern of F. arthrerion .

In the phylogeny of Fridericia ( Kaehler & al. 2019), two gatherings of F. arthrerion were retrieved in different clades. The first was placed within the “ Neomacfadya ” clade (Lohmann 521, as F. egensis ), while the second was placed within the “ Fridericia s.str. ” clade (Kileen 4015). After a thorough study of Lohmann 521, we noted that this gathering has coriaceous and narrow fruit (c. 1.4 cm wide) as found in other F. arthrerion specimens, whereas F. egensis has woody and wider fruits (2.3–2.7 cm wide), indicating that Lohmann 521 actually corresponds to F. arthrerion . While Kileen 4015 was correctly identified, we suspect that there might be an error or contamination with the sequence included in the molecular phylogenetic study by Kaehler & al. (2019), given that the morphology does not match its phylogenetic placement. The phylogenetic placement of this material should be verified in future studies.

Here we synonymize Fridericia pearcei under F. arthrerion due to their shared morphologies, as noted by Gentry (1977b: 40) and Sandwith (in sched.). Fridericia pearcei was a widely used name for the plants with long, multicellular trichomes. However, the specimens from Ecuador lack these trichomes, as noted by Gentry (1977b). On the other hand, F. arthrerion was the name used for specimens with black leaves and ferruginous indumentum when dry, features that combined were found in a few specimens. After careful comparison of the types, and a thorough examination of around 100 specimens, we realized that this species includes a broad morphological variation of indumentum and colour of dried leaves, with a continuum between the extremes. The indumentum is ferruginous but varies from puberulous to completely villous leaves, calyces and fruits, whereas the dried leaves vary from green or brownish green to black. However, the shape of these structures is constant. Therefore, we follow Gentry’s and Sandwith’s suggestions and consider Bignonia pearcei as a synonym of F. arthrerion (Mart.) L. G. Lohmann.