Caelospermum versus Coelospermum in Rubiaceae (Gentianales): their etymologies explained

Authors: Brecht Verstraete, Elmar Robbrecht

Abstract

Caelospermum is the original spelling of the generic name, as it appeared in the protologue, but in literature and online databases, the variant spelling Coelospermum is often used. However, the confusion about the spelling is caused by a misinterpretation of the etymology. Here, we demonstrate that the original spelling by Blume should be retained.

Caelospermum versus Coelospermum

Caelospermum Blume is a genus within the Rubiaceae family that is found in Southeast Asia and Oceania and currently consists of 12 species (POWO 2025). The genus is included in the tribe Morindeae, subfamily Rubioideae (Razafimandimbison and Rydin 2024). The circumscription of the genus has been controversial but a broadly delimited Caelospermum can be characterised by the mostly lianescent habit, corolla tubes with narrow longitudinal slits alternating with the lobes, ovaries simple or 2–20(–30)-connate, primarily bilocular with biovular locules, later becoming 4-locular by the development of secondary septa, drupaceous fruits, concavo-convex to planoconvex pyrenes with a narrow slit in the lower part, and flattened seeds with a narrow basal wing (Johansson 1988). Razafimandimbison et al. (2009) confirmed Johansson’s concept of Caelospermum with molecular phylogenetic methods, although enlarging it by treating the Australian genus Pogonolobus F.Muell. as a synonym.

The generic name was originally spelled as Caelospermum in the protologue (Blume 1826: 994; Fig. 1). The genus was mentioned to be closely related to Morinda L. and Gynochtodes Blume and a single species (i.e. the type species), Caelospermumscandens Blume, was described from Nusa Kambangan island in Indonesia. There was no explanation of the etymology of the generic name, but it seems evident that it is derived from Latin “caelum” (= heaven, sky) and “sperma” (= seed). Although Blume mentioned some characters of the pyrenes (Fig. 1), this etymology did not seem to make sense.

Shortly after, the spelling of the generic name was changed by other taxonomists. Both de Candolle (1830: 468) and Richard (1834: 209) corrected the name to Coelospermum, however, without commenting on the reason. Hooker (1873: 119) also used Coelospermum but he mentioned Caelospermum between brackets. Throughout literature, both spelling variants have been used, although it seems that Coelospermum is more prevalent, especially in recent molecular phylogenetic studies (e.g. Razafimandimbison et al. 2008, 2009). The monographer of the genus first used Coelospermum (Johansson 1987). In his revision of the genus, however, he used the original spelling Caelospermum and listed the name Coelospermum in synonymy (Johansson 1988), without explaining why he made this change. Other authors also used both variants in different publications (e.g. Halford and Ford 2004 vs Halford and Ford 2009). Mouly and Fleurot (2021) used the spelling Coelospermum but explicitly mentioned Caelospermum as an orthographical variant. The origin of the change in spelling is found in the interpretation of the etymology. Subsequent authors interpreted the name as derived from Greek “koilos” (= concave, hollow), not Latin “caelum” (= heaven, sky). Since Blume (1826)’s description included the phrase “Pyrenae ... intus concavae ...”, it was interpreted that Blume meant “hollow-seeded” (IPNI 2025).

Besides in literature, the names are also entered in online databases about nomenclature and taxonomy. The International Plant Names Index (IPNI 2025) lists both Caelospermum and Coelospermum but both names have two entries and there is confusion about the correct spelling. In the nomenclatural notes of Caelospermum (https://www.ipni.org/n/34272-1), it says to use Coelospermum. However, the entry Caelospermum (https://www.ipni.org/n/116925-3) remarks that there is an orthographical variant Coelospermum. The entry Coelospermum (https://www.ipni.org/n/327850-2) has a remark in which IPNI chooses to correct Caelospermum to Coelospermum. The entry Coelospermum (https://www.ipni.org/n/85291-3) says the opposite because the remark says that Coelospermum is the orthographical variant. Tropicos (Tropicos.org 2025) is clearer about the Coelospermum (https://www.tropicos.org/name/40037911) is considered to be the orthographical variant of Caelospermum (https://www.tropicos.org/name/40034119). The World Flora Online (WFO 2025) lists Caelospermum (https://www.worldfloraonline.org/taxon/wfo-4000008840) and several subordinate taxa, but all those names are annotated as unchecked names. The status of Coelospermum (https://www.worldfloraonline.org/taxon/wfo-4000005905) and its species are indicated as accepted names. The Plants of the World Online (POWO 2025 only lists Coelospermum (https://powo.science.kew.org/taxon/lsid:ipni.org:names:327850-2).

The interchangeability of cael- (from “caelum”) and coel- (from “koilos”) is also noted for other generic names in angiosperms. Hasskarl (1855: 139) published the name Koilodepas Hassk. (Euphorbiaceae), which is derived from Greek “koilos” (= concave) and “depas” (= beaker) and refers to the cupulate calyx. He later Latinised the name to Coelodepas Hassk. (Hasskarl, 1857). However, Bentham and Hooker (1880: 313) published the name as Caelodepas Benth. & Hook.f., with reference to Coelodepas Hassk. and Koilodepas Hassk. Other examples are Caelebogyne J.Sm. / Coelebogyne J.Sm., Caelestina Cass. / Coelestina Cass., Caelia G.Don. / Coelia Lindl., Caelocline Auct. ex Steud. / Coelocline A.DC., Caeloglossum Steud. / Coeloglossum Lindl., and Caelogyne Wall. ex Steud. / Coelogyne Lindl. (Steudel 1840: 247; IPNI 2025). In almost all cases, the variant coel- is preferred (POWO 2025; WFO 2025). Caelestium Yurtseva & Mavrodiev (Polygonaceae) has no variant because the etymology is explicitly stated in the “… from “caelestis” (= heavenly, divinus)” (Yurtseva and Mavrodiev 2019: 73).

Regarding Caelospermum, the correction of Blume’s spelling of the name is the result of a misinterpretation of the etymology. Caelospermum can indeed be derived from Latin “caelum” but it should be translated as “vault of heaven” (Lewis and Short 1891). In ancient cosmology, the celestial vault is portrayed as a vast solid dome arching above the Earth. The name would then refer to arched pyrenes (Fig. 2), which corresponds well with the characters in the protologue (i.e. “Pyrenae … intus concavae, extus gibbae”) and in Johansson (1988)’s revision (i.e. “endocarp … concavo-convex to planoconvex”). The interpretation that the name would be derived from Greek “koilos” is conceivable but not necessarily correct.

In Rubiaceae, the origins of the generic names by Blume are not always obvious (e.g. Metabolos Blume; Bremekamp 1952: 30) and Blume never explained their etymologies in the protologues, except for eponyms. We will therefore never be able to definitively ascertain what Blume actually intended. Ultimately, both cael- and coel- can refer to a concave feature, and can therefore be used in that sense. However, their etymologies are different, so they should not be considered as orthographical variants. In that case, the original spelling of the name should be retained (ICN Art. 60.1; Turland 2019). In conclusion, the original spelling Caelospermum Blume is correct and should be used.