(3004) Proposal to reject the name Oenothera muricata (Onagraceae)

(3004) Oenothera muricata L., Syst. Nat., ed. 12, 2: 263. 15–31 Oct 1767 [Angiosp.: Onagr.], nom. rej. prop.

Lectotypus (vide Fernald in Rhodora 51: 65. 1949): Herb. Linnaeus No. 484.3 (LINN).

Oenothera muricata L. (Syst. Nat., ed. 12: 263. 1767) was described by Carolus Linnaeus with a brief diagnosis indicating its resemblance to O. parviflora L. (Syst. Nat., ed. 10: 998. 1759), but differing by having non-emarginated teeth of the capsule apex and red-coloured papillae covering the stem (“fol. lanceolatis planis, caule purpurascente muricato. Similis parviflorae, sed Fructus os non 8-fidus. Caulis puncta rubra sparsa.”). In addition, the origin of the species was stated to be Canada; however, no specimens or illustrations were cited in the protologue. In 1775, Johan A. Murray (in Novi Comment. Soc. Regiae Sci. Gott. 6: 24–26. 1775) provided a more detailed description and the first illustration of O. muricata. Murray's illustration (l.c.: t. 1; available at https://babel.hathitrust.org/cgi/pt?id=uc1.d0001408673&view=1up&seq=53&seq=53) shows a plant similar to O. parviflora, with numerous side branches within the inflorescence area, divergent sepal tips and a straight tip of the inflorescence axis, and therefore it seems to be consistent with the protologue. Moreover, the illustration was cited in Systema vegetabilium (Syst. Veg., ed. 14: 358. 1784) and in the subsequent edition of Systema naturae by Johann F. Gmelin (Syst. Nat., ed. 13: 607. 1796).

Over the last 200 years, the name Oenothera muricata has been interpreted differently by many taxonomists and geneticists, making its application very confusing. One of the main sources of confusion appeared with the illustration of O. muricata in Flora Danica (Oeder, Fl. Dan. 10(30): t. 1752. 1823, also available at http://www5.kb.dk/images/billed/2010/okt/billeder/object317208/da/). This illustration is significantly different from Murray's illustration and shows a plant with no flowering nor basal side branches, narrow leaves and, what is the most confusing, an explicitly curved top of the inflorescence axis. Such features point to O. oakesiana (A. Gray) J.W. Robbins ex S. Watson & J.M. Coult. (in Gray, Manual, ed. 6: 190. 1890) or O. ammophila Focke (in Abh. Naturwiss. Vereins Bremen 18: 182. 1904) but not to O. parviflora nor any of its closest allies. Following the illustration in Flora Danica, Hugo de Vries (Gruppenw. Artbild.: 37, fig. 11 & t. 7. 1913) interpreted the name O. muricata as referring to those plants characterized by an explicitly curved inflorescence axis. A similar interpretation was presented by Reginald R. Gates (in Rep. (Annual) Missouri Bot. Gard. 1909: 129. 1909). Nevertheless, MacDougal & al. (Mutat., Variat. Relat. Oenotheras: 74. 1907), who cultivated O. muricata from the seeds collected by De Vries in the Netherlands, pointed out some differences between the cultivated plants and O. muricata known from America, as well as noting their significant similarity to the illustration presented in Flora Danica. It is also worth mentioning that the interpretation of the pictures of O. muricata provided by De Vries (l.c., also available at https://www.biodiversitylibrary.org/item/73110#page/49/mode/1up) is very ambiguous. On the one hand, the colour illustration (t. 7) shows a small-flowered plant with a straight inflorescence axis, whereas the specimen visible in the black-and-white picture (fig. 11), demonstrates explicitly a bent upper part of the inflorescence. In both cases sepal tips are not visible enough to conclude on their composition.

To avoid further misapplication of the name Oenothera muricata, Harley Harris Bartlett described a new species, O. syrticola Bartlett (in Cybele Columb. 1: 38. 1914) [currently known as O. oakesiana], which was supposed to replace O. muricata sensu De Vries. Unfortunately, the result turned out to be the opposite of this intention and the name O. muricata started to be assigned to the whole Cernue group, later replaced by O. sect. Parviflorae, consisting of several species, e.g., O. parviflora, O. silesiaca Renner (in Ber. Deutsch. Bot. Ges. 60: 455. 1942), O. syrticola and O. ammophila (Renner, l.c.; Raven in Tutin & al., Fl. Eur. 2: 307. 1968; Cleland, Oenothera Cytogen. Evol.: 309–310. 1972). Moreover, a separate interpretation has been proposed by Soó (in Acta Biol. Acad. Sci. Hung. 3: 226. 1952) and Jávorka & Csapody (Iconogr. Fl. Austro-orient. Eur. Centr.: 25. 1979), who stated that the name O. muricata should be combined with O. salicifolia Desf. ex G. Don (Gen. Hist. 2: 685. 1832; = O. hungarica (Borbás) Borbás in Magyar Bot. Lapok 2: 246. 1903). However, the basis of this statement remains uncertain since it was not supported by original material or any element of the protologue.

In 1949, Merritt Lyndon Fernald (in Rhodora 51: 65. 1949) designated the lectotype of Oenothera muricata from the specimens stored in the Linnean Herbarium. Although the sheet (LINN No. 484.3; image available at https://linnean-online.org/5367/) was annotated by Linnaeus as “muricata”, it shows a specimen that contradicts the protologue and Murray's illustration. The specimen consists only of the upper part of inflorescence with no open flowers, and its appressed sepal tips and straight inflorescence axis make it similar to the common O. biennis L. (Sp. Pl.: 346. 1753). As a consequence of this designation, representatives of the so-called American school of Oenothera taxonomy placed the name O. muricata into the synonymy of O. biennis s.l. (Munz in Rogerson, N. Amer. Fl., ser. 2, 5: 133. 1965; Dietrich & al. in Syst. Bot. Monogr. 50: 293. 1997; Jiarui & al. in Wu & Raven, Fl. China 13: 424. 2007). In contrast, European specialists treated O. muricata as the oldest name referring to O. rubricaulis Kleb. (in Jahrb. Hamburg. Wiss. Anst., Beih. 31(3): 12. 1914 [“1913”]), a microspecies which belongs to the closest allies of O. biennis (Rostański in Acta. Biol. (Katowice) 90: 13. 1975; Rostański & Ellis in Nat. Wales 16: 242. 1979; Jehlík in Slavík, Květena České Republiky 5: 80. 1997), even though O. rubricaulis is known only from Europe and has never been found in Canada. However, later, Krzysztof Rostański (in Watsonia 14: 23. 1982) clearly stated that O. muricata “must be treated as nomen confusum and rejected”.

Based on the analysis presented above, the adoption of Oenothera muricata as the correct name for any of the Oenothera species would be very disruptive to nomenclatural stability. Moreover, the designation of a new type for O. muricata would exacerbate the discrepancy between the narrow and broad species concept of O. biennis. Therefore, to preserve nomenclatural stability, it is here proposed to reject the name O. muricata under Art. 56 of the ICN (Turland & al. in Regnum Veg. 159. 2018).

Acceptance of this proposal would avoid the misapplication of the name Oenothera muricata for the well-established names of Oenothera species. Rejection of this proposal would mean that the name O. muricata would eventually be adopted to replace O. rubricaulis or another well-established name of an Oenothera species.