Novitates neocaledonicae XIV: A third species of Goniothalamus (Blume) Hook. f. & Thomson from New Caledonia and lectotypification of G. obtusatus (Baill.) R. M. K. Saunders Author Munzinger, Jérôme AMAP, Univ. Montpellier, IRD, CIRAD, CNRS, INRAE, F- 34398 Montpellier (France) jerome. munzinger @ ird. fr jerome.munzinger@ird.fr Author Johnson, David M. Department of Biological Sciences, Ohio Wesleyan University, Delaware, Ohio 43015 (United States) Author Saunders, Richard M. K. School of Biological Sciences, The University of Hong Kong, Pokfulam Road, Hong Kong (PRC) text Adansonia 2023 3 2023-08-28 45 20 327 335 journal article 10.5252/adansonia2023v45a20 1639-4798 8334996 Goniothalamus obtusatus (Baill.) R.M.K.Saunders Oxymitra obtusata Baill. , Adansonia; recueil d’observations botaniques 8: 178 ( Baillon 1868 ). — Lectotype : New Caledonia . sin. loc. , 186[?], Vieillard 641 ( P [ P01982592 ], here designated). LECTOTYPIFICATION Baillon (1868: 178) described Oxymitra obtusata Baill. , the basionym of Goniothalamus obtusatus ,based on specimens indicated as being “dans les collections de M.Vieillard”, and following the diagnosis, he stated “ Oritur in Novae-Caledoniae montibus, ubi legit cl. Vieillard ”, which indicates that the type material comprises Vieillard’s collections (plural). Saunders & Wang (2011) interpreted this as Vieillard s.n. and cite the “ holotype ” as being at P.But the Paris herbarium has no fewer than eight herbarium specimens made by Vieillard, all of which thus represent potential syntypes .Moreover, the system used by Vieillard to number his material is well known to be “special and irrational” ( Morat 2010 ) because he assigned a given number to what he thought was a species rather than a single gathering, a situation that was further complicated by the fact that label exchanges are also suspected ( Guillaumin 1942 ). Thus when a species is typified by two or more Vieillard specimens, it is prudent to consider them as probable syntypes or isosyntypes ( Gâteblé et al. 2021 ). Of the eight sheets of Oxymitra obtusata deposited in Paris, three (P01749441, P01982589, P01982591) were previously in the herbarium in Caen (CN transferred to Paris in the 1970s) and were probably not seen by Baillon. Among the five remaining specimens, three have labels with locality information, Balade (P01982590) or Wagap (P01982594, P01982595), which is not indicated in the protologue. The two remaining specimens (P01982592, P01982593) are the only ones with the name Oxymitra obtusata Baill. in Baillon’s hand; P01982592 is in better condition and includes dissected fruits with isolated seeds in a packet, and since it was on the basis of the characteristics of the seed that Baillon used the epithet “obtusata”, this sheet (P01982592) is chosen as the lectotype . Some variations can be observed in Goniothalamus obtusatus (Baill.) R.M.K.Saunders , notably in monocarp shape, with some being globose, wide and with a rounded apex in one, and others narrower with a more acute apex. We were not able to relate this variation to other morphological characters nor to ecological or edaphic differences.