(3108) Proposal to conserve Agalinis, nom. cons. against the additional name Esterhazya (Orobanchaceae)

(3108) Agalinis Raf., New Fl. 2: 61. Jul–Dec 1837 [Scrophular. / Orobanch.], nom. cons. Typus: A. palustris Raf., nom. illeg. (Gerardia purpurea L., A. purpurea (L.) Pennell) (typ. cons.). (=) Esterhazya J.C. Mikan, Del. Fl. Faun. Bras.: ad t. [8]. 1822 (mid), nom. rej. prop. Typus: E. splendida J.C. Mikan Agalinis Raf. (New Fl. 2: 61. 1837) currently comprises approximately 65–70 species restricted to the Western Hemisphere, although the exact number of species is unknown due to taxonomic confusion. Across this distribution, North America contains ~40 species of Agalinis, while ~30 species are found in Central and South America. The herbaceous taxa are mostly found in North America, whilst in South America subshrubs and shrubs are predominant, as observed by D'Arcy (in Taxon 28: 419–420. 1979) when synonymizing the Andean Virgularia Ruiz & Pav. Agalinis has long been considered a problematic genus both morphologically and nomenclaturally, which led in the past to its proposed conservation against Virgularia Ruiz & Pav. (Fl. Peruv. Prodr.: 92. 1794), Chytra C.F. Gaertn. (Suppl. Carp.: 184. 1807) and Tomanthera Raf. (l.c. 1837: 65) (Thieret in Taxon 7: 142–143. 1958; D'Arcy, l.c.). The debate surrounding the conservation proposals involving the names Agalinis and Gerardia L. and the typification of the latter caused disagreements between several authors regarding the correct interpretation of the rules of botanical nomenclature. Gerardia was described by Linnaeus (Sp. Pl.: 610. 1753) as including five species: G. flava L., G. glutinosa L., G. pedicularia L., G. purpurea L. and G. tuberosa L. Linnaeus's concept of the genus was rather heterogeneous and Smith (in Rees, Cycl. 16: [155]. 1810) pointed out that G. tuberosa (currently in Acanthaceae) should be the only species of Gerardia, and the other species should receive a new generic denomination and characterization. However, over the years the name Gerardia L. was applied to only a few species in Acanthaceae, while the majority of the species were described in Scrophulariaceae (now in Orobanchaceae). The latter circumscription was adopted in Bentham's revision (in Comp. Bot. Mag. 1: 204–211. 1836), which became known as Gerardia sensu Bentham. The genus was thereafter used in the scrophulariaceous sense in all major taxonomic treatments, but Thieret (in Taxon 5: 58. 1956) successfully proposed the conservation of Stenandrium Nees (in Lindley, Intr. Nat. Syst. Bot., ed. 2: 444. 1836) (Acanthaceae) against Gerardia L., based on the mistaken assumption that Smith (l.c.) inadvertently typified Gerardia L. with G. tuberosa, stating that it "must be the true, though it were the only Gerardia". This was nevertheless the type, having later been effectively designated by Rafinesque (Fl. Tellur. 4: 66. 1838). A subsequent attempt to conserve Gerardia L. sensu Bentham (l.c. 1836) against Gerardia L. (Morton in Taxon 8: 28. 1959) was rejected by the Committee for Spermatophyta (Rickett in Taxon 10: 124. 1961), since by then the use of the name Agalinis had already gained popularity. Esterhazya J.C. Mikan (Del. Fl. Faun. Bras.: ad t. [8]. 1822) was described with a single species, E. splendida, but currently comprises 5 species exclusive to eastern Brazil, with the extremely variable E. splendida extending westwards to Bolivia and Paraguay. Mikan commented on the new genus' affinity with Virgularia (= Agalinis), from which it differed by having filaments villous at the base and apex, along with oval, exserted anthers, woolly-bearded and with a bifid base. Conversely, the filaments of Virgularia were villous only at the base, with included anthers that were sagittate and glabrous. Chamisso & Schlechtendal (in Linnaea 3: 16–18. 1828) included species currently circumscribed in Esterhazya in Gerardia sect. Dargeria, which was characterized by exserted stamens, as opposed to Gerardia sect. Gerardia, whose flowers had included stamens. Martius (Nov. Gen. Sp. Pl. 3: 5. 1829) treated Esterhazya in the synonymy of Virgularia, but the genus was restored by Bentham (in Candolle, Prodr. 10: 514. 1846) and, despite having been considered a synonym of Gerardia by Kuntze (Revis. Gen. Pl. 3(3): 232. 1898), it has since been recognized as a distinct genus. The distinction between Esterhazya and Agalinis has nevertheless been debatable. Bentham (l.c. 1846) differentiated Esterhazya and Gerardia (= Agalinis) on the basis of the position of the stamens (long exserted in Esterhazya vs. included in Agalinis), and on the color of the corolla (red vs. violet to pink). Bentham & Hooker (Gen. Pl. 2: 924. 1876) associated these stamen characteristics with the densely villous anthers of Esterhazya; in addition, these authors pointed out that the leaves of Esterhazya are always entire, whereas Agalinis has entire or incised leaves. In his identification key, Schmidt (in Martius, Fl. Bras. 8(1): 234. 1862) differentiated the two genera based on the position and indument of the stamens, shape of the corolla (tubular-infundibuliform in Esterhazya vs. ventricose-tubular in Gerardia) and base of the anthers (acute vs. short mucronulate or obtuse). Wettstein (in Engler & Prantl, Nat. Pflanzenfam. 4(3b): 90–92. 1891), Barroso (in Rodriguésia 15(27): 53–57. 1952) and Ichaso & Barroso (in Reitz & al., Fl. Ilustr. Catarinense, Escrofular. 1: 74. 1970) differentiated Esterhazya and Gerardia (= Agalinis) mostly based on the position of the stamens. However, the existence of intermediate species presents an obstacle to defining the characteristics used for the separation of these two genera. For example, Agalinis angustifolia (Mart.) D'Arcy has stamens that reach the corolla orifice despite being glabrous, and Esterhazya caesarea (Cham. & Schltdl.) V.C. Souza has a lilac corolla more typical of Agalinis. Likewise, the use of anther base and corolla shape to separate genera as proposed by Schmidt (l.c.) is questionable, as these are quite variable characteristics in Agalinis. The morphological ambiguity between Esterhazya and Agalinis was recently elucidated by phylogenetic analyses (Latvis & al. in Syst. Bot. 49: 455–481. 2024), which revealed Esterhazya nested within an otherwise monophyletic Agalinis within all data partitions. Following nomenclatural priority, new combinations would be required for 40 species of Agalinis widespread throughout the Americas, whilst the conservation of this name against Esterhazya would reduce this number to five. The conservation of Agalinis against Esterhazya thus preserves nomenclatural stability, drastically minimizes the impact on herbarium institutions, and would be the most coherent alternative considering the now well-established history of nomenclatural usage of Agalinis in lieu of Gerardia sensu Benth.