MATERIALS AND METHODS
Data sources: A list of plant names of Thymelaeaceae was compiled based on thorough searches of pertinent botanical literature, herbarium specimen annotations, and online databases of plant nomenclatural data such as Index Nominum Genericorum (ING, 2010), The International Plant Names Index (IPNI, 2010), and Tropicos.org (Tropicos, 2010). Protologues and other relevant publications were gathered and digitized so that the current nomenclatural standing (e.g., validity, legitimacy, priority, typification status) could be determined for each name according to the International Code of Botanical Nomenclature (McNeill et al., 2006). This procedure has uncovered, not unexpectedly, numerous names with some kind of nomenclatural problem, and a large number of those, especially names based on African and Asian types, currently lack effective typifications. In this checklist, names determined to be invalid or illegitimate are preceded by asterisks. A number of these names require formal nomenclatural correction via effective publication following the appropriate guidelines outlined by McNeill et al. (2006).
All data provided in the World Checklist should not be considered definitive and are subject to change as additional information is gathered. Links to digitized original publications for names (i.e., protologues) are often provided within individual name records. Additional literature and supporting references are available for download in pdf format via several online websites such as Botanicus, the Biodiversity Heritage Library, and the Digital Library of the Real Jardín Botánico of Madrid
Limits between Wikstroemia Endl. and Daphne L. are notoriously blurred, but both are recognized as distinct genera according to current conventions. Halda (2001) was the first person to formally treat Wikstroemia as a synonym of Daphne, but this action is totally unjustified without comprehensive analyses of the morphological and molecular data. Two other genera, Eriosolena Blume and Rhamnoneuron Gilg, are closely related to Daphne, but also tentatively retained here as distinct from Daphne and Wikstroemia, respectively. Thecanthes Wikstr. is provisionally maintained as distinct from Pimelea according to the Flora of Australia treatment (Rye, 1990). Gyrinops Gaertn. is retained as a distinct genus pending further molecular study despite some preliminary ITS evidence suggesting a single species of Gyrinops is embedded within a paraphyletic Aquilaria Lam. (Eurlings & Gravendeel, 2005). Given these caveats and exceptions, 50 genera of Thymelaeaceae are recognized in the World Checklist as of January 2010.
Taxonomic evaluation and species diversity: Existing taxonomy is provisionally evaluated when sufficient material for a particular genus or species is available. Assessment proves difficult for rarely collected species and larger genera, especially when types or original material for names have been lost or destroyed or when the taxon belongs to a species complex. In these cases, more information is needed before a decision can be reached. For species that are difficult to circumscribe, previously published judgments of “reputable” authors are adopted.
The taxonomic status of names that cannot be assigned at present and are placed in the “Status unclear” category. Most of these names represent synonyms of existing species, nomina nuda, hybrid taxa, invalid or illegitimate names, or names with untraceable type material. Hybrid names, both natural and horticultural, are provided in the checklist but no serious efforts were made to include a comprehensive list of cultivar names. For further information regarding cultivated species of Daphne see Halda (2001).
Type interpretation and citation: All available sources of original material referenced in the protologue (e.g., earlier publications, illustrations, type specimens, cited exsiccatae) were databased and carefully evaluated for the World Checklist. Publications discussing the typification status of Thymelaeaceae names were scoured for potentially effective typification statements. A thorough search was conducted to identify original material and types at numerous important herbaria (acronyms following Index Herbariorum, IH, 2010), online databases, and libraries. Thymelaeaceae collections in several large herbaria (BM, F, G, K, MO, NY, P, US) were personally inventoried, photographed, scanned, and databased on Tropicos. Additional sources of original material such as type photographs, microfiche collections, and online specimen databases including Aluka (African types), Krypto-S (S), Sonnerat (P), and CJBG (G) were also consulted. At present, Thymelaeaceae collections have been examined from at least 37 different institutions: A, B, BERG, BM, BREM, C, CANB, CAS, CORD, E, F, G, GH, GOET, HAL, HBG, K, L, LE, LINN, M, MEL, MO, NY, P, PI, R, RB, S, SP, SPF, U, UC, US, W, WAG, and Z.
The interpretation of original material and type specimens follows Turland and Jarvis (1997) and McNeill et al. (2006). Information related to collectors, associated institutions, and their herbarium depositories was taken from Taxonomic Literature-2 (TL-2, 2010), the Harvard University Herbaria: Index of Botanists (HUH, 2010), and Index Herbariorum. When the identity of original herbarium material proved dubious, attempts were made to match anonymous inscriptions to the handwriting of the author of the plant name in question by using several publications and repositories with handwriting samples such as Auxilium ad botanicorum graphicem (ABG, 2010) and Handwriting: Linnean herbarium (SMNH, 2010). In some cases personal correspondence of the author was used. Original material, especially that related to older names, often lacks handwritten annotations of the species name by the publishing author(s). The absence of such handwriting does not necessarily preclude the material from consideration as original material because authors did not always personally annotate material used for their descriptions.
Geographic coordinates are assigned, whenever possible, using standard online sources (e.g., Google Earth, The Fuzzy Gazetteer) and regional gazetteers. Within individual name records, specimen-based distribution data and maps can be downloaded or viewed under the collapsible “Specimens” tab of the interface. Specimen records with coordinate data can also be downloaded in the “kml” file format for display in the freely available Google Earth software, but distributional information will sometimes contain collection records with identifications not yet evaluated for the World Checklist. Phenology and elevation data can be viewed graphically by clicking on the “Elevation Chart” and “Phenology Chart” links below the “Specimens” tab, but again displayed information may include a mixture of verified and unverified collections. Verified specimen records bear current determinations by Z. S. Rogers or other family specialists.
Distribution information taken from trusted bibliographic sources is available for most accepted taxa. To view this information, choose “Tropicos” from the “Choose Project” option located in the upper-right hand corner of almost any page on the World Checklist site. Then click on the “Distribution” tab within the Tropicos name record.
Bibliographic citations: Standardized titles and abbreviations for serial and book titles are taken from Botanico-Periodicum-Huntianum (BPH, 2010) and Taxonomic Literature-2, respectively. Author abbreviations for plant names follow Authors of Plant Names (Brummitt & Powell, 1992) and the related author database available on The International Plant Names Index website.
Exsiccatae citations and vouchers: In addition to original material and types, attempts have been made to personally select at least one herbarium voucher for each species accepted in the World Checklist. Vouchers of accepted names, if deposited at MO, have been imaged at high (600dpi) resolution and are displayed within individual name records. Images of types and vouchers from other institutions will be uploaded to the database with institutional permission.
Generic concepts: Most genera are circumscribed and distinguished following the classification of Herber (2003), but there are a few exceptions where different generic limits are adopted based on the author's opinion. For instance, although preliminary molecular evidence suggests that Gnidia is paraphyletic (Van der Bank et al., 2002; Rautenbach & Van der Bank, unpubl. data), a broad delimitation of the genus is employed here that includes Lasiosiphon Fresen. and several other, mostly small, segregate genera, following the most recent African (e.g., Robyns, 1975; Peterson, 1978; Peterson, 2006) and Malagasy taxonomic treatments (Rogers, 2009) of Gnidia. Conversely, the presumably closely related genera, Englerodaphne Gilg (3 African spp.) and Atemnosiphon Leandri (1 sp.), are tentatively maintained as distinct from Gnidia in the World Checklist.