(2778) Proposal to conserve the name <i>Zamites</i> (fossil <i>Cycadophyta</i>: <i>Bennettitales</i>) with a conserved type

(2778) Zamites Brongn., Prodr. Hist. Vég. Foss.: 91, 94. Dec 1828, nom. cons. prop. Typus: Z. gigas (Lindl. & Hutton) Morris in Ann. Mag. Nat. Hist. 7: 116. Apr 1841 (Zamia gigas Lindl. & Hutton, Foss. Fl. Gr. Brit. 3: 45. Jul 1835), typ. cons. prop. Brongniart published the genus Zamites for fossil leaves that he considered as strongly comparable to leaves of the recent genus Zamia. After listing 12 fossil species of Zamia (of which, for one species only, he mentioned a species basionym: in Polypodiolites), he presented four species of Zamites: Z. bechii, Z. bucklandii, Z. lagotis, and Z. hastatus (‘hastata’), all of them with an older basionym in Filicites. The main difference between Zamia and Zamites was the unforked venation in Zamia and the bifurcated venation in Zamites. Originally, species were published in Filicites for fossil leaves that had a fern affinity. Afterwards, it was deduced that several of those species belonged to various other main groups of fossil plants, including, among others, several orders of Gymnosperms. This was, for example, the case with Z. bucklandii, already mentioned by Brongniart (in Orbigny, Dict. Univ. Hist. Nat. 13: 110. 1849) as the type of Otozamites, see below. The generic name Zamites is nowadays regularly used for a considerable number of species. In the past, the first author made searches for typifications of fossil generic names in various important sources such as Miller (N. Amer. Geol. Palaeontol. 1889), Lamotte (in Bull. U.S. Geol. Surv. 924. 1944) and in several books by A.C. Seward. A type designation for Zamites was never met in those searches. Previously, only one source was found for a Zamites type designation, but this typification was not effective. Andrews, in both editions of his Index of generic names of fossil plants (in Bull. U.S. Geol. Surv. 1013: 261. 1955; l.c. 1300: 229. 1970), had a special comment on Zamites: “Owing to innumerable name changes in the cycadophyte leaf genera, it is extremely difficult to cite type species, especially for Zamites. The following is rather arbitrarily suggested: Zamites gigas (Lindley and Hutton) Morris, 1843, p. 24.” Even though Andrews in this record used, very exceptionally*, the term “type”, this choice was not of one of the four species names included by Brongniart in the protologue of Zamites (Art. 10.2 of the ICN; Turland & al. in Regnum Veg. 159. 2018). [*For almost every generic name in his Index, Andrews mentioned a species name as explained in the introduction of his work: “The primary objective has been to cite for each genus a type or representative species that will serve as a much-needed basic reference for paleobotanical taxonomy.”] Only recently, the first author found an earlier and effective type designation. This was in the context of an extended search in Pfeiffer's Nomenclator botanicus of 1871–1875 that started in the summer of 2015 with the generic name Aa Rchb. f. (Xenia Orchid. 1: 18. 1854). The large majority of generic names listed by Pfeiffer concern non-fossil plants. Occasionally a generic name of a fossil plant was encountered, usually without a species name being mentioned, or, if exceptionally so, only in cases in which there was only one original species. The first case of type selection by Pfeiffer for the name of a fossil plant genus was that of Stemmatopteris, seen in May 2019. This was the same month in which we arrived at the conclusion that the best type for Otozamites was O. bucklandii, based on Filicites bucklandii – a choice that had already been made by Brongniart (l.c. 1849), and that we maintained in our proposal to conserve Otozamites against Otopteris (Zijlstra & Van Konijnenburg-van Cittert in Taxon 68: 874–875. 2019). In February 2020, the first author had almost reached the end of searching through Pfeiffer and came to Zamites. For this name, Pfeiffer listed as type Z. bucklandii – and, on checking back, for Otozamites had “= Zamites”! It immediately became clear that Pfeiffer's type designation of Zamites presented a problem for Otozamites, making it a later synonym of Zamites (by lectotypification). It is, however, also a problem for Zamites because this name is still in regular use in a sense different from Otozamites, and not including Z. bucklandii. Very shortly after this discovery, the Utrecht University buildings were closed because of the coronavirus pandemic, and it was only possible in June 2020 to enter the building briefly and access the literature enabling this proposal to be prepared. Looking back at all four original species of Zamites, it became clear that Z. bechii and Z. lagotis also belong to Otozamites; Z. hastatus probably does not, but the position of that species is unclear as it has not been recorded since Brongniart (Prodr. Hist. Vég. Foss.: 95. 1828). Thus, it is necessary to select as type a species name not included in the protologue, and we decided to propose Z. gigas as conserved type. Andrews (l.c. 1955, 1970) had already attempted to designate this name as type, reflecting his knowledge of the way in which Zamites has long been used. In his Zamites entry, Andrews (l.c. 1955, 1970) also noted: “See discussion in Seward, 1917, p. 529–532.” Seward (Fossil Pl. 3: 529–532. 1917) presented a thorough overview on how the species of Zamites had been treated in the past by many important palaeobotanists, and ended with a detailed generic description. In the paragraph dealing with the genus Williamsonia (l.c. 1917: 423), he mentioned his own 1900 description and discussion of Zamia gigas (Cat. Mesoz. Pl., Jurass. Fl. 1: 177–189. 1900), in which, following Carruthers (in Trans. Linn. Soc. London 26: 693. 1870), he included this species in Williamsonia Carruth. (l.c.: 691), based on leaves of Zamia gigas and flowers, now known to be female, associated with them in fossil collections in Yorkshire (England) by Williamson (in Trans. Linn. Soc. London 26: 663–674. 1870). In 1917, Seward, although typifying Williamsonia by W. gigas (≡ Zamia gigas) (l.c. 1917: 421), took the view (l.c. 1917: 423) that Zamites should only be used when solely the leaves are known. The fact that W. gigas is homotypic with Zamites gigas requires a solution through another proposal. Some important species of Zamites are: Z. bohemicus Velen. (in Abh. Königl. Böhm. Ges. Wiss., ser. 7, 3: 6. 1889), Z. brevipennis (Oishi) H. Takim. & al. (in Paleontol. Res. 12: 135. 2008), Z. buchianus (Ettingsh.) Seward (Cat. Mesoz. Pl., Weald. Fl. 2: 79. 1895), Z. choshiensis T. Kimura & Ohana (in Proc. Japan Acad., B 61: 352. 1985), Z. decurrens C.A. Menéndez (in Bull. Brit. Mus. (Nat. Hist.) Geol. 12: 18. 1966), Z. feneonis (Miq.) Ettingsh. (in Abh. K. K. Geol. Reichsanst. Wien 1(3.3): 9. 1852), Z. occidentalis Newb. (in Macomb, Rep. Explor. Exped. Santa Fé, New Mexico 1859: 142. 1876), Z. persica Boureau (in Mém. Mus. Natl. Hist. Nat., ser. 2, 30: 226. 1950), Z. quiniae T.M. Harris (Yorkshire Jurassic Fl. 3: 3. 1969), Z. vachrameevii Doludenko (in Trudy Geol. Inst. Akad. Nauk S.S.S.R., ser. 2, 178: 67. 1969), and Z. weberi Seward (in Trudy Geol. Komiteta, ser. 2, 38: 11. 1907). Many dozens of species of Zamites are recognized, all of them from Mesozoic layers, commonly Jurassic, from many parts of the world: Europe, Asia, North and South America. If Z. bucklandii should remain the type of Zamites, all those leaf species (including the above-mentioned) that are now in Zamites would have to be placed under a new generic name, since we do not know an existing name that is available and suitable for these species. Therefore, we propose to retain the genus Zamites in its currently accepted sense with Z. gigas as the conserved type. GZ, https://orcid.org/0000-0001-9894-9967 JHAvKvC, https://orcid.org/0000-0001-5833-3439 We thank John McNeill (Royal Botanic Garden, Edinburgh, Scotland & Royal Ontario Museum, Toronto, Canada) for his valuable comments to improve this proposal.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.