As part of the formation of the WFO TEN for Ericaceae in 2020, the family backbone was supplied by the WFO (WFO Consortium 2019) as a Darwin Core Archive. This seeded a dataset (Elliott et al. 2020), managed in Padme, a taxonomic database system developed by the Royal Botanic Garden Edinburgh. The family data, provided by WFO, was largely based on “The Plant List” (2013) v1.1. For Erica, updated nomenclatural records were compared to the “International Plant Names Index” (IPNI). Additional names published since 2012 were added to the backbone. Taxonomic placement of species was checked mainly using literature sources for the “Flora of Southern Africa” region (FSA) (Oliver 1984, 1987, 2000; Oliver and Oliver 2002, 2005; Pirie et al. 2017) and online resources, namely Catalogue of the Plants of Madagascar (Rabarimanarivo et al. 2015) and the Botanical Database of Southern Africa (BODATSA) (South African National Biodiversity Institute 2016). Although the South African data was chiefly accessed through the frontend user interface, the yearly checklist data is available in an archived version (South African National Biodiversity Institute 2024).

In 2022, the WFO’s Rhakhis tool (Hyam et al. 2022; Hyam and Elliott 2023) became available, and data in the WFO taxonomic backbone was synced to the Ericaceae classification from Padme. Curation for Ericaceae then transferred from Padme to Rhakhis.

The initial checklist created for the formation of the Global Conservation Consortium for Erica (GCC-Erica), based on the World Checklist of Vascular Plants (WCVP; Govaerts 2022), was compared to the WFO backbone in Rhakhis. Missing names were added to the WFO backbone and conflicts in classification were highlighted and resolved by referring to the literature or the IRHN.

In 1970, The Heather Society undertook the role of International Cultivar Registration Authority (ICRA) for the genera Andromeda L., Bruckenthalia Rchb., Calluna Salisb., Daboecia D.Don, and Erica [these formed a denomination class as defined by the “International code of nomenclature of cultivated plants” 1995 (ICNCP) under the International Commission for the Nomenclature of Cultivated Plants]. The published volumes (Nelson and Small 2000, 2004–2005) were derived from a database that included more details of the names, including bibliographic references to descriptions and illustrations, history, and etymology of the individual scientific and horticultural (cultivar) names. To preserve these data, the entire IRHN database has been archived (see below).

The first volume of the “International Register of Heather Names” (IRHN; Nelson and Small 2000) was published in four parts and contained all cultivar and botanical names that had then been traced within the genera Andromeda, Bruckenthalia, Calluna, Daboecia, and Erica in the broad sense. This list covered species and subspecific taxa of Erica species indigenous in Europe, Asia Minor, the Atlantic islands (Azores, Madeira, and the Canary Islands), and Africa north of the Tropic of Cancer, and their natural and man-made hybrids and their cultivars. These are generally known colloquially as “hardy heaths” or “hardy heathers” because the majority can tolerate the relatively cool climate of the northern hemisphere, north of the Tropic of Cancer.

The second volume (Nelson and Small 2004–2005), also published in four parts, contained names for Erica species and subspecific taxa indigenous to Africa south of the Tropic of Cancer and the islands of the southern Atlantic and Indian Oceans. Many of these are colloquially known (especially in horticulture) as “Cape heaths” although many do not occur within the Cape provinces of South Africa. This volume also included the names of subspecific taxa, natural and artificial hybrids, and cultivars of the Cape heaths. With the re-circumscription of Erica to include previously separate “minor” genera (Oliver 2000), a list of these synonymised genera was included as Annex 1 of Volume 2, and their names, now being synonyms of Erica taxa, were also included in that checklist.

Compilation of the IRHN was a major collaborative effort involving members of The Heather Society (1963–2020), particularly its designated registrars, and sister societies in western Europe and North America. Research and publication of the checklist was funded by The Heather Society with additional financial support from the Stanley Smith Horticultural Trust (UK).

Names data from The Heather Society’s Access 2000® database, used to produce the IRHN (Nelson and Small 2000, 2004–2005), were matched against the WFO backbone. Names data were extracted from this database. To aid in name matching, authorships were modified to reflect standard author abbreviations according to Biodiversity Information Standards (TDWG) (formerly the Taxonomic Databases Working Group), and names without an author string were initially excluded. Natural and artificial hybrid names were added to the WFO when missing as these can have nomenclatural significance. Natural hybrids were placed in the classification where appropriate following the IRHN. Artificial hybrids are unplaced and deprecated, as these are outside the scope of the WFO. The status of “deprecated” was introduced primarily as an internal device in the WFO. It is meant in the modern sense of the word, particularly regarding software: “to withdraw official support for or discourage the use of”. Cultivar names were not processed as they also are beyond the scope of the WFO. This process of bringing in historic names from IRHN raised homonym issues among Latin binomials that were known but not yet resolved. The required replacement scientific names have been dealt with by Nelson et al. (2023). These new names were registered with IPNI (2023) as part of the pre-publication process and subsequently added to the WFO database.

An IRHN archive (Elliott et al. 2023) has been created in Zenodo (the general-purpose open repository developed under the European OpenAIRE program). This contains the complete, unedited IRHN database, the two volumes (eight parts) as published by The Heather Society (in pdf format), and a csv file containing WFO IDs linked to botanical names in the IRHN.

Orthographic variation (particularly in the terminations employed in eponyms and toponyms) has been prevalent in the historic literature for Erica (Nelson and Oliver 2004; Turner 2016). Existing WFO records were corrected in accordance with the “International Code of Nomenclature for algae, fungi, and plants” (ICN, Shenzhen Code; Turland et al. 2018). Some variants that have featured in botanical and horticultural literature and databases (including SANBI’s Red List of South African Plants for threat status; http://redlist.sanbi.org/genus.php?genus=1820) were added and linked to the currently accepted name. We further considered the future consequences of proposals currently under consideration to modify Article 60.8 of the ICN with regard to restricting the correction of names (Proposal 023 by Mabberly 2020; and proposal 024 by Van Rijckevorsel 2020).

In the context of the WFO, unplaced names are those not associated with any taxon as an accepted name or synonym. This could be due a taxonomist having not expressed an opinion on its placement, or the name cannot be resolved to anything in the classification. These records can be found by searching the name parts but are not found by browsing the classification’s hierarchy.

Modern cultivar names as defined under the ICNCP (Brickell et al. 2009), used in the horticultural industry and in gardens lie beyond the scope of the WFO and were therefore omitted from the checklist. Historic Latin form names applied to horticultural selections were removed from the classification and deprecated when there was little doubt that they did not refer to wild species.

Data on cultivar names published before 2001 can be found in the original Access database format and as a csv file in the Zenodo archive (https://zenodo.org/doi/10.5281/zenodo.10255787) (Elliott et al. 2023). The cultivar list will be added to the Catalogue of Life’s Checklist Bank.

For comparison, the number of accepted species, subspecies, varieties, and synonyms were recorded from the “World Checklist of Vascular Plants” (WCVP; v.11.0) (Govaerts 2023), the “Synonymic Checklists of the Vascular Plants of the World” (v.16.4, Sep 2023) (Hassler 2023), and “The Leipzig Catalogue of Vascular Plants” (v.3.01) (Freiberg et al. 2020) using datasets deposited in the Catalogue of Life Checklist Bank (https://www.checklistbank.org/).

Table 2 shows the comparison of numbers of taxa represented across the “World Checklist of Vascular Plants” (WCVP; v.11.0), the “Synonymic Checklists of the Vascular Plants of the World” (v.16.4, Sep 2023), and “The Leipzig Catalogue of Vascular Plants” (v.3.01). There is great variation in the number of names across all taxonomic ranks among these lists, especially in terms of infraspecific taxa and synonym names.

Table 3 lists species and hybrid names in Erica with orthographic variants that have appeared in botanical and horticultural literature and databases, as well as their currently accepted orthography. An indication is also given for names where the acceptance at the Madrid Nomenclature Section of the International Botanical Congress (IBC; July 2024) of current proposals to amend Article 60.8 of the ICN, would result in a reversion to the original spelling as published in the protologue of each name.

Nomenclatural issues have been dealt with by Nelson et al. (2023) as part of the systematics, natural history, and conservation of the Erica (Ericaceae) collection. The following three species are further clarified:

Erica tegetiformis E.G.H.Oliv. in Bothalia 20(1): 46. 1990.

IPNI: urn:lsid:ipni.org:names:941276-1

WFO: wfo-0000673371

Replaced synonym: Erica senilis var. australis Dulfer in Ann. Naturhist. Mus. Wien 66: 32. 1963.

IPNI: urn:lsid:ipni.org:names:77251074-1

WFO: wfo-0000673222

Oliver (1990) raised E. senilis var. australis Dulfer to species level since it is significantly different from E. senilis Klotzsch ex Benth. The epithet ‘australis’ was not available for this taxon at species level because of the earlier name E. australis L. (in Mant. Pl. Altera: 231. 1771) that remains the valid name for one of the European species. Therefore, Oliver (1990) published the new name E. tegetiformis E.G.H.Oliv. for this taxon. We consider this name to be validly published. In other lists it is considered to be not validly published due to the omission of the full reference of the replaced synonym. There is an indirect reference made with the combination “E. senilis Klotzsch ex Benth. var. australis Dulfer: 32 (1963)”. There is only one Dulfer reference in the bibliography of the article by Oliver (1990). While the ICN recommends refraining from this practice (see Rec. 41A.1; Turland et al. 2018), it is permissible to have the full and direct reference separate from the newly published name or combination. The year of the journal volume for the Dulfer reference is cited as 1964 (instead of 1963), but we consider this to be a correctable error under Art. 41.6 (Turland et al. 2018). For these reasons, we treat E. tegetiformis (in Bothalia 20: 46. 1990) as validly published.

Erica heleophila Guthrie & Bolus in Fl. Cap. (Harvey) 4(1.1): 110. 1905.

IPNI: urn:lsid:ipni.org:names:328833-1

WFO: wfo-0000672224

This orthographic issue was dealt with in Nelson and Small (2004–2005) but is revisited here. In the key to species and the protologue (Guthrie and Bolus 1905: 19, 110), the species epithet was published as ‘heliophila’. However, in the Addenda and Corrigenda preceding the Index and the Corrigenda following the Index in the same volume and section of the “Flora Capensis” (Thiselton-Dyer 1909: 1126, 1168), the epithet was amended to ‘heleophila’, changing the meaning of the epithet to “of the marsh” rather than “of the sun”. In the Index (Thiselton-Dyer 1909: 1146), both epithets are listed. It is unclear why this amendment has been largely overlooked. The original publication date of part 1 of volume 4 of “Flora Capensis” was May 1905 (Stafleu and Cowan 1979: 76) and the correction was published in February 1909 (part 6 of the volume), albeit in the same volume and section (volume 4 section 1). The correction of the name E. heleophila is an orthographic (potentially, typographic; Nelson and Small 2004–2005) error permissible under Art. 60.1 (Turland et al. 2018).

The WFO ID of the original orthographic variant can be found in the Table 3.

Erica michellensis Dulfer in Ann. Naturhist. Mus. Wien 67: 85. 1963.

IPNI: urn:lsid:ipni.org:names:329124-1

WFO: wfo-1000055012

Erica saxatilis L.Bolus in Ann. Bolus Herb. 3: 177. 1924. nom illeg. hom. non Erica saxatilis Salisb., Prodr. Stirp. Chap. Allerton: 295. 1796.

IPNI: urn:lsid:ipni.org:names:329562-1

WFO: wfo-0000673189

This orthographic issue was also dealt with by Nelson and Small (2004–2005) but is revisited here. The name E. saxatilis L.Bolus (in Ann. Bolus Herb. 3: 177. 1923) is an illegitimate later homonym of the earlier name E. saxatilis Salisb. (in Prodr. Stirp. Chap. Allerton: 295. 1796) (= E. carnea L.). Dulfer (1963) therefore published a new name for this taxon, namely E. mitchelliensis, with that original spelling. The type collection of E. saxatilis L.Bolus was given as “Cape Province; South-Western Region; Ceres Div., Mitchells Peak, Mitchells Pass, “growing on rocks, rare,” alt. 4500 ft., fl. Dec. 1920, T.P. Stokoe 66” (Bolus 1923: 177). The peak and pass commemorate Charles Cornwallis [baptised Collier] Michell (1793–1851) (Richings 2006), Surveyor-General of the Cape of Good Hope and Superintendent of Works in 1848 when the pass was originally opened (Raper et al. 2014). Dulfer (1963) chose ‘mitchelliensis’ as his epithet, using the “Mitchell” spelling as it was found in the E. saxatilis protologue (Bolus 1923), and derived from the label on Stokoe’s specimens.

Dulfer (1963) also constructed the name with an additional “i” before the -ensis. This has been considered a correctable error in previously published works. The corrected orthography, E. mitchellensis, is used in the South African National Plant Checklist (South African National Biodiversity Institute 2024) and Red List of South African Plants (Turner 2008).

The IRHN (Nelson and Small 2004–2005) further corrected the name to E. michellensis due to the incorrect spellings of “Mitchell’s Peak” and “Mitchell’s Pass”, when they should have been Michell’s Peak and Michell’s Pass as on modern maps of the region. The correction in the IRHN has not been widely adopted or used, but there is nothing in the ICN to suggest that Nelson’s entry in the IRHN is incorrect. We therefore suggest that E. michellensis is the correct orthography to follow for this name. Both orthographic variants are in the WFO as separate entries and synonymised to Erica michellensis.

The WFO IDs for the orthographic variants can be found in the Table 3.

Erica triceps Link in Enum. Hort. Berol. Alt. 1: 371. 1821.

IPNI: urn:lsid:ipni.org:names:329771-1

WFO: wfo-0000673442

Erica adunca Benth. in Prodr. 7: 618. 1839.

IPNI: urn:lsid:ipni.org:names:328152-1

WFO: wfo-0000671312

Although previously treated as separate species (e.g., in Schumann et al. 1992: 193 and 195), Oliver (in Oliver 2012 and in Oliver and Forshaw 2012) treated these two names as synonymous. The name E. triceps is the older of the two names and thus has priority (Art. 11.1; Turland et al. 2018). However, a note in the ID aid from Oliver (Oliver and Forshaw 2012) stated that any type in Berlin was destroyed during the Second World War and the description in the protologue (Link 1821: 371) is insufficient to definitely associate it morphologically with E. adunca or any other known species. No type was explicitly designated in the protologue, but original material derived from the plant cultivated at the Berlin Botanical Garden in 1808 can be assumed to have been in Herb. B. We have not traced other original material. Dulfer (1964: 116 no. 370) noted that the habitat of the species was the Cape of Good Hope (“Hab. in promont. bon. sp.”), repeating information published by Link (1821: 371 no. 3731: “Hab. in Pr. b. sp.”) in the protologue. However, Dulfer (1964) did not directly cite any extant specimen as a type.

Nomenclatural resolution depends on the relative use of E. triceps and E. adunca for the species as currently circumscribed. Use of the older name, E. triceps, and its application is uncertain due to the lack of original material and the ambiguity of the description in the protologue (Link 1821: 371). Should E. triceps and E. adunca be regarded as synonymous, unequivocal use of the younger name, E. adunca, would require formal rejection of the older E. triceps under Art. 56 (Turland et al. 2018). Such action at this stage would be premature as further investigation might reveal information that can clarify the application of E. triceps. We believe the best course of action is to regard E. triceps as an insufficiently known name that cannot be applied to any extant taxon with certainty, and we do not treat it as a synonym of E. adunca. We apply only the name E. adunca to the South African taxon that has previously been treated as either E. adunca or E. triceps, the latter probably a misapplication.

Immediate work is required to reduce the number of unplaced names, by placing names that can be traced to wild plants in the classification and deprecating those of horticultural origin.

As of June 2024, many names have a taxonomic reference, i.e. the citation from where the taxonomic concept or circumscription is derived (see Berendsohn 1995) but more references are needed. These references are currently omitted because the relevant publications lack a doi or stable URL to link with, which is a requirement for references in Rhakhis. Use of a taxonomic concept reference is implemented throughout the Ericaceae TEN and follows the best practice adopted by the Caryophyllales TEN (Fassou et al. 2022; Korotkova et al. 2021).

The Erica checklist will be continually edited, when appropriate, via the WFO Rhakhis tool to contribute to the Ericaceae TEN, the wider WFO project and support various activities of the GCC-Erica. The six-monthly releases, apart from providing achievable deadlines for incremental improvements, also allow for a stable citable taxonomy that can be referenced and compared across time through the WFO Plant List API.

The Erica classification was extracted from the December 2023 WFO Plant List release to synchronise the classification in the Erica identification aid (Oliver et al. 2024). Synchronisation to future WFO Plant List releases will continue.

In working through the developing WFO pipeline with Catalogue of Life (CoL), the Erica checklist will be incorporated into the annual CoL Checklist and from there can be utilised by the Global Biodiversity Information Facility (GBIF).

The authors have declared that no competing interests exist.

No ethical statement was reported.

No funding was reported.

ACE: Data Curation, Investigation, Writing - Original draft, Writing - Review and Editing. SPB: Data Curation, Writing - Review and Editing. RRK: Data Curation, Writing - Review and Editing. ECN: Data Curation, Investigation, Writing - Review and Editing. MDP: Data Curation, Investigation, Writing - Review and Editing.

Alan C. Elliott https://orcid.org/0000-0002-8677-5688

Stoffel P. Bester https://orcid.org/0000-0001-6961-5052

Ronell R. Klopper https://orcid.org/0000-0002-0948-5038

E. Charles Nelson https://orcid.org/0000-0002-3909-4388

Michael D. Pirie https://orcid.org/0000-0003-0403-4470

All of the data that support the findings of this study are available in the main text.