686 TAXON 61 (3) • June 2012: 686–687 (2075) Proposal to reject the name Cactus coquimbanus ( Cactaceae ) Urs Eggli1 & Helmut Walter2 1 Sukkulenten-Sammlung Zürich, Mythenquai 88, 8002 Zürich, Switzerland 2 Casilla 175, Buin, Chile Author for correspondence: Urs Eggli, urs.eggli@zuerich.ch (2075) Cactus coquimbanus Molina, Sag. Stor. Nat. Chili: 170, 352. 12–31 Oct 1782 [Dicot.: Cact.], nom. utique rej. prop. Typus: deest. Juan Ignacio Molina (1740–1830) was the first naturalist that applied Linnean nomenclature to Chilean plants (Marticorena, Fl. Chile 1: 15–16. 1995). His list and descriptions of Chilean plants were published in 1782 (Sag. Stor. Nat. Chili), which is 14 years after Molina was forced to leave Chile in 1768 with the expulsion of the jesuits. Molina apparently compiled his plant descriptions largely from memory, which explains the difficulties encountered in the identification of many of his names (Marticorena, l.c.: 16). This also concerns Cactus coquimbanus, which was described from Coquimbo. Molina recognized two species of columnar cacti from Chile, the “ordinary Cactus peruvianus”, and his newly described C. coquimbanus. Since Molina’s C. coquimbanus is the first validly published name for a Chilean columnar cactus, it has priority over any other name, and establishing its correct application is therefore of primary importance. “Cactus peruvianus ” was a name widely applied at the time of Molina and well into the beginning of the 19th century to co- lumnar cacti from South America, but in current classifications and by virtue of lectotypification, Cactus peruvianus L. is a synonym of the Caribbean and Venezuelan Cereus repandus (L.) Mill. (e.g., Anderson, Grosse Kakt.-Lex., 2005 & 2011). Positive identification of both Cactus peruvianus Molina (non L.) and Cactus coquimbanus Molina have been elusive so far. No herbarium material of Molina’s collections has been traced, and no details about his itineraries, etc., have been found in the Torino herbarium, where documents of him are stored (Laura Guglielmone, pers. comm., Oct. 2010). This leaves us with the (i) protologue and the (ii) origin (Coquimbo) as the only sources of information to interpret the name Cactus coquimbanus. To our knowledge four columnar taxa of cacti are native to the region around the city of Coquimbo [Molina with certainty refers to the city of Coquimbo, as the Intendencia Coquimbo was only founded in 1811, the Provincia Coquimbo in 1826]. Using the taxonomy of Anderson (l.c.), these are Echinopsis chiloensis (Colla) H. Friedrich & G.D. Rowley, Echinopsis coquimbana (Molina) H. Friedrich & G.D. Rowley, Eulychnia breviflora Phil. and Eulychnia acida Phil. The minimal description of Cactus coquimbanus (“Cactus erectus, longus, 10-angularis, angulis obtusis, spinis longissimis rectis” / “produce le spine lunghe da otto pollici”) does not allow a positive identification with any one of the four taxa. A comparison of numerous populations of all taxa involved shows considerable overlap in the characters given in the protologue of Cactus coquimbanus (pers. obs.), and it is pos- sible that its description embraces elements of two or more of the taxa present in the region. In the absence of additional data or herbarium material, it is therefore impossible to positively associate Molina’s name either with the genus Echinopsis or with the genus Eulychnia. The correct application of the name Cactus coquimbanus appears to have been problematic at an early time, and already Lemaire (Cact. Gen. Sp. Nov.: 83. 1839) distinguished between “Cactus coquimbanus Molin. Chil.”, treated as a synonym of Cereus eburneus Salm-Dyck [= Stenocereus griseus (Haw.) Buxb. (Anderson, l.c.)], and “Cactus coquimbanus hort. non Molin.”, treated as a synonym of Cereus chi- loensis Colla. Subsequent authors equated the epithet “coquimbanus” (ascribed to Molina, or as “hort. non Molina”) either with a taxon attributable to Trichocereus (a taxonomic synonym of Echinopsis in current classifications) (e.g., Labouret, Monogr. Cact.: 334. 1853, sub Cereus; Schumann, Gesamtbeschr. Kakt., Nachtr.: 23. 1903, sub Cereus; Britton & Rose, Cact. 2: 139. 1920; Berger, Kakteen: 136. 1929; Backeberg & Knuth, Kakt.-ABC: 202. 1936; Borg, Cacti: 138. 1937; Werdermann in Kakt. and. Sukk. 1937: 22. 1937; Backeberg, Cactaceae 2: 1144. 1958; Ritter, Kakt. Südamer. 3: 1111. 1980), to Echinopsis (e.g., Hoffmann, Cact. Fl. Silv. Chile: 86. 1989; Hunt, CITES Cact. Checklist, ed. 2: 51, 187. 1999; Hoffmann & Walter, Cact. Fl. Silv. Chile, ed. 2, 2005; Hunt & al., l.c.), or to Eulychnia (Schumann, Gesamtbeschr. Kakt.: 58–60. 1897, sub Cereus ). The uncertainty in the application of the name Cactus coquim- banus could be resolved by designating a neotype specimen. Such a designation would, however, be based on a completely arbitrary decision. Whatever element would be chosen, some of its features would be in conflict with the protologue and thus the selection could be later superseded under Art. 9.17 of the Vienna Code (McNeill & al. in Regnum Veg. 146. 2006), further adding to the existing confusion. In addition, any neotypification would have another important negative consequence, threatening the stability of a name in current use. If the “minority view” would be supported by such a neotypi- fication, Cactus coquimbanus would become the oldest name for a species of the genus Eulychnia, and would have priority over both Eulychnia breviflora and Eulychnia acida. Neither of these two names has ever been questioned, and replacing either is undesirable and would create additional confusion. If the “majority view” concerning Eggli & Walter • (2075) Reject Cactus coquimbanus two species on the duplicates received in P, identified the one bear- ing trinerved leaves as H. macrophylla and described the other with penninerved leaves as H. wallichiana, making it an earlier name for H. mitrata. Therefore, in order to preserve nomenclatural stability in accordance with ICBN Art. 14.2, conservation of the name H. mitrata against H. wallichiana is here proposed. Acknowledgements The curators of the herbaria mentioned in the text are thanked for providing access to the specimens or for providing loans and digital images. John McNeill is thanked for providing useful comments on a previous version of the manuscript. This work is financed by the National Parks Board, Singapore.