Nomenclature and typification in the endemic genus Codia (Cunoniaceae) from New Caledonia Helen C. Fortune HOPKINS Masons Arms, Hutton Roof, Carnforth, Lancashire, LA6 2PE (United Kingdom) and Royal Botanic Gardens, Kew, Richmond, Surrey, TW9 3AB (United Kingdom) [email protected] ABSTRACT The taxonomic history of Codia is described briefly. All names validly pub- KEY WORDS lished in Codia ,or published under other generic names but referable to Cunoniaceae, Codia , are discussed and lectotypified where necessary. Invalidly published Codia , names are also listed. Codia cinerascens (Pamp.) H.C.Hopkins, comb. et stat. New Caledonia, nomenclature, nov., Codia fusca (Schltr.) H.C.Hopkins, comb. nov., and Codia incrassata typification. Pamp. var. rufinervis (Guillaumin) H.C.Hopkins, comb. nov., are published. RÉSUMÉ L’histoire taxonomique de Codia est récapitulée. Tous les noms valablement publiés sous Codia , ou publiés dans d’autres genres mais devant être rapportés MOTS CLÉS à Codia , sont discutés et au besoin lectotypifiés. Les noms non valablement Cunoniaceae, publiés sont également pris en compte. Codia cinerascens (Pamp.) Codia , H.C.Hopkins, comb. et stat. nov., Codia fusca (Schltr.) H.C.Hopkins, comb. Nouvelle-Calédonie, nomenclature, nov., et Codia incrassata Pamp. var. rufinervis (Guillaumin) H.C.Hopkins, typification. comb. nov., sont ici établis. INTRODUCTION Taxonomic problems in Codia are of two sorts. First are those of species delimitation, which will be Codia , with about 14 species (including some discussed elsewhere. Second are those concerning undescribed), is endemic to the large island of names, authors and types. For instance, some syn- New Caledonia (Grande Terre) and some of the types include material referable to more than one small islands immediately adjacent to it, though taxon (e.g., C . floribunda ). Some varieties, now ele- absent from the Loyalty Islands. The purpose of vated to specific rank, have been recognised previ- this paper is to resolve problems in nomenclature ously under more than one specific epithet (e.g., and typification prior to completion of the “var. cinerascens ”). Some names have been misap- account of the family Cunoniaceae for Flore de la plied in herbaria and/or literature because of lack of Nouvelle-Calédonie. access to types and few published illustrations (e.g., ADANSONIA, sér. 3 • 2005 • 27 (2) : 243-254 © Publications Scientifiques du Muséum national d’Histoire naturelle, Paris. 243 Hopkins H.C.F. C . arborea , C . montana ). Some invalid names have for many years after the original generic descrip- been used sporadically in the literature. Confusion tion, taxonomic works often just repeated the with Pancheria , to which Codia is superficially simi- information given by the F ORSTERS and men- lar as both have flowers in capitula, has lead to a tioned a single species, C . montana . An exception few species being misplaced. to this was L ABILLARDIÈRE (1824) who published Many names in Codia are based on V IEILLARD a detailed description and illustration of C . mont- specimens and the problems associated with these ana (Fig. 1) based on his own material. In the are well known. The numbers on his specimens are second half of the 19th century, B RONGNIART & usually equivalent to a species number rather than G RIS (1862) described four new species referring to a single gathering, and so material ( C . ferruginea , C . floribunda , C . obcordata and under one number may have come from more C . spatulata ), based largely on the collections of than one locality (Wagap, Balard, Gatope, Kanala, V IEILLARD. Some 40 years later, P AMPANINI etc., usually handwritten on the label), or may (1904, 1905) added four more names ( C . albi- have been collected during different periods (either cans , C . albicans var. cinerascens , C . incrassata , 1855-1860 or 1861-1867, usually printed on the C . microcephala ),and S CHLECHTER described label); even with a single locality and date, material C . nitida and Pancheria fusca (actually a Codia ) in may have come from more than one plant and is 1906. The most recent treatment was by occasionally a mixture of more than one species. G UILLAUMIN (1941, 1948), who initially recog- Lectotypification is especially important to fix the nised 10 species, adding C . microphylla in 1943. use of names based on these collections. Both V IEILLARD andB RONGNIART wrote The paper is in several parts. First the taxo- names for proposed new taxa on herbarium spec- nomic history of the genus is described briefly, imens. Some of V IEILLARD’s names were pub- then generic names are discussed, followed by lished by B RONGNIART & G RIS, P AMPANINI and sectional ones. Next all names validly and effec- G UILLAUMIN. Other names have been published tively published in Codia , or considered referable invalidly, especially in works by G UILLAUMIN to Codia , are listed. Information from the proto- (1911a, b). logue about the type(s) is given in quotation It is remarkable how many of the species recog- marks and a lectotype designated where neces- nised now are based on the early collections of sary. Digital images of types are being collected V IEILLARD. Other significant collectors from the and will be available on CD from the author and same period include B ALANSA, D EPLANCHE and on the web in due course. Invalid names are in a P ANCHER. Among the relatively recent collec- separate list. Finally an index to all names indi- tions, those by G UILLAUMIN, B AUMANN and cates their status and synonymy. Note that litera- H ÜRLIMANN in 1951 are sometimes poorly ture references only discussed immediately after pressed and some are sterile, but they are impor- their citation accompanying a Latin name are not tant because they frequently represent several repeated in the references at the end. individuals in the same population or are from localities close to one another, thus giving a pic- ture of local variation. The largest number of col- TAXONOMIC HISTORY lections is by M ACK EE, who collected more than 160 numbers of Codia . The genus was discovered and described by Johann Reinhold F ORSTER and his son Johann Georg Adam F ORSTER, who visited New GENERIC NAMES Caledonia in 1774 as naturalists on Captain JamesC OOK’s second voyage of discovery in the CODIA J.R.Forst. & G.Forst. Pacific. C OOK was the first European to “dis- Char. Gen. Pl.: 59, tab. 30 (1775). cover” New Caledonia. Pfeifferago Kuntze, Revis. Gen. Pl. 1: 227 (1891). Because the genus is endemic to New Caledonia, few botanists have worked on it and T YPE . — Codia montana J.R.Forst. & G.Forst. 244 ADANSONIA, sér. 3• 2005 • 27 (2) Codia (Cunoniaceae) in New Caledonia F IG. 1. — Codia montana J.R.Forst. & G.Forst. Reproduced from L ABILLARDIÈRE (1824). ADANSONIA, sér. 3 • 2005 • 27 (2) 245 Hopkins H.C.F. The first edition of the F ORSTERS’ book, with Although B RONGNIART & G RIS (1862) divided only two copies known, was published on 29 No- Codia into two sections containing species with vember 1775 (F ARR et al . 1986; see http:// petals (sect. Eucodia = sect. Codia ) and those ravenel.si.edu/botany/ing/). Numerous copies without (sect. Codiopsis ), the possession of petals exist of the two subsequent editions from 1776 appears to be variable in several species and so (S TAFLEU & C OWAN 1976), and this date has this character cannot be used reliably to distin- sometimes been given as that for publication of guish two groups. the name Codia . K UNTZE (1891) published Pfeifferago as a nomenclatural synonym to replace Codia , consid- VALIDLY PUBLISHED NAMES ering the latter to be a variant of the same word as R UMPHIUS’ name Codiaeum but referring to a 1. Codia albicans Vieill. ex Pamp. different plant. Both names are derived from the Greek word “kodeia” or “kodia”, meaning a head, Ann. Bot. Rome 2: 104 (1905). — Type (as given or Codiaeum may perhaps be derived from in protologue): “Ad montes prope Wagap [Vieillard, Herb. de la Nouvelle Calédonie, 1861-1867, n. 582] “kodiho”, the vernacular name for Codiaeum (B-B, C-I, D C)”. — Lectotype (here designated): variegatum in Ternate in the Moluccas in collibus circa Wagap, 1861-1867, Vieillard 582 (Q UATTROCCHI 2000). However, the name Codia (G!; iso-, FI [photo]!, G! [date “1868”, see below], is legitimate as R UMPHIUS’ name was published K! [2 sheets], P! [4 sheets]; probable iso-, BM!, prior to 1753 and thus cannot be considered for NY [image]! barcode no. NY00356073, P!; other sheets exist at L and MEL [ fide H OOGLAND, card priority, and Codia is not now thought to be suf- index]). ficiently similar to Codiaeum to be problematic. The name Pfeifferago commemorated the P AMPANINI indicated that two of the type German botanist Louis Karl GeorgP FEIFFER specimens were in the Barbey-Boissier herbarium (1805-1877) (Q UATTROCCHI 2000). (B-B) and the de Candolle herbarium (D C) at G. However, since these collections do not refer to Flora Orientalis nor to either DE C ANDOLLE’s SECTIONAL NAMES Prodromus nor the Monographiae Phaneroga- marum , they are now in the general herbarium at 1. Sect. Codia G (see http://www.cjb.unige.ch/). The sheet cited as C-I refers to the herbarium Centrale Italiano at Sect. Eucodia Brongn. & Gris, Bull. Soc. bot. FI which is also now incorporated into their gen- France 9: 76 (1862), nom. inval. (Greuter et al . 2000: eral herbarium (see http://sciweb.nybg.org/ Arts 21.3 and 32.1(b)). — “Flores petalis praediti” ( C . montana , C . floribunda , C . ferruginea ). science2/ IndexHerbariorum.asp). Types of other species of Codia described by P AMPANINI are Guillaumin (1941) – C . montana , C . albifrons , similarly in the general herbaria at G or FI. C . ferruginea (note this last lacks petals); Guillaumin The handwritten date “1868” on the sheet at (1948) – as listed in 1941.