Our mission is to eliminate the need for repeated reference to primary sources for basic bibliographic information about plant names.
About the index
The International Plant Names Index (IPNI) is a nomenclatural index of names of vascular plants published under the Code of Nomenclature for algae, fungi and plants (ICN).
The records in IPNI originally came from three sources in 2000, the Index Kewensis (IK), the Gray Card Index (GCI) and the Australian Plant Names Index (APNI). Index Filicum, covering the ferns (and incorporating lycophytes published after 1960), was added in 2004. Since then various other smaller datasets have been incorporated, details can be found on the acknowledgments page.
IK included only the genus and species of seed plants. The Gray Card Index names included vascular plants of the New World. The Australian Plant Name Index records names for all Australian plants but its contribution to IPNI is restricted to the vascular plants.
An automated deduplication process was applied to the data in 2016 and the current website search results display only the ‘top’ copy of duplicated entries from one of the three sources mentioned above. Links to the duplicate records from the other data sources are shown in the full record view and the data are still visible when clicked.
Since appearing online in 2000, IPNI has been continuously curated: editors screen available literature and publications daily and add newly published names to the database, along with their rank, type information, authors, place of publication and exact publication dates. Authors publishing in journals or books to which there is limited access are encouraged to send copies of their publications to the editors in order to ensure their new names are included. The website is updated daily at around 4am GMT.
History of the Data
Over one million records have come from IK (1893-2002).
This is global in coverage and lists names from the first edition of Linnaeus's Species Plantarum (1753) up to the year 2000. Basic bibliographic details are included for each name and for later records the year of publication is also included. Until 1971 infraspecific names were not included. GCI is largely complete for infraspecific names from 1886 and APNI is complete for Australian infraspecific names, therefore many names below species level are missing from IPNI. We are actively working to add these names, those present in the World Checklist of Vascular Plants will be imported in bulk.
Data imported from IK was produced by an optical reading process in the mid-1980s. Despite careful checking of the scanned data, many errors were introduced into the Index at this time. Standardisation of the data is an ongoing process, with 150,000 records updated annually on average. See our Statistics page for more information.
Over 350,000 records have come from GCI (originally the Gray Herbarium Card Index) which includes names for New World taxa published on or after January 1886. Basic bibliographic details and date are included for each name and many records include information about types. The data were converted to electronic form in the early 1990s and much time has been invested in their standardisation and verification since then. Although most of the citations in GCI are for names also recorded in IK, there are many records for New World infraspecific names which are unique. The duplicate records are also of considerable value: they tend to be more detailed than the equivalent IK record and errors in either index can be detected by comparing them. GCI includes many records of typifications subsequent to the time of valid publication of a name (epi-, lecto- and neo-typifications).
Over 63,000 records have come from APNI which has been compiled since 1973 and includes all scientific names used in the literature for Australian vascular plants. Levels of detail and validation are higher for this index than for IK or GCI. Most names have been checked back to their original place of publication and type information is included for each name. APNI includes many names not included in IK, especially of Australian infraspecific taxa, and, as in the case of GCI, the overlap in coverage between IK and APNI offers scope for checking data and eliminating errors.
About the Authors dataset
IPNI follows the guidelines set out in Brummitt & Powell, Authors of Plant Names (1992) on the formation of an Author Standard Form.
The standard form of a name is a surname, or an abbreviation of it (e.g. Adans. for Adanson), or rarely a contraction of it (e.g. Michx. for Michaux), with or without initials or other distinguishing appendages (see principles 7-9 below).
Our Author principles
All our Author principles were voted on by the Working Party.
Names are given in Roman characters (see also notes on transliteration and similar problems under ‘Surname Problems’ above).
Every ‘standard form’ must be unique to one person. (Unless the same author always has the same ‘standard form’, see 10 and 11 below.)
3. Uniform treatment of names
The same surname (i.e. identical spelling) must always be given in the same form (e.g. Miller is always abbreviated to Mill.), unless it is part of a compound name (see 14).
Different surnames must not be given in the same form e.g. if Brown is abbreviated to Br. then Browne must not also be abbreviated to Br.
We have treated masculine and feminine forms of one name as different names e.g. Botschantzev and Botschantzeva are not given the same standard form.
4. Full-stops and accents
All abbreviations and contractions are terminated by a full-stop e.g. Adans., Walt.Jones, Michx. However, the full-stop does not make a standard form different from the same spelling without a full stop e.g. Lam. for Lamarck and Lam for H.J. Lam would be treated as homonymous, and initials are required for the latter’s standard form.
Similarly, names differing only by presence or absence of an accent (e.g. Love and Löve, Leonard and Léonard) or apostrophe (Ohara and O’Hara) are treated as homonymous.
The remaining principles are not considered absolutely binding except for no. 12.
5. TL-2 as a standard
When used, this over-rides principles 6–14.
We favour retaining traditional abbreviations and other standard forms. We accept the standard forms recommended in Stafleu & Cowan’s TL-2.
This has been done in the majority of cases, but in a small number TL-2 has not been followed because:
- contravention of principles 2 and 3 above (rare, e.g. Stef.)
- serious conflict with the Kew Draft Index involving transfer of one standard form from one author to another;
- failure to abbreviate very long names, such as Schlagdenhauffen
- rather excessive abbreviation of a few names of six letters or fewer where no outstandingly strong precedent seems to exist (e.g. Wilson is given in full rather than abbreviated to Wils.)
- failure to give any of a large number of authors with the same surname a standard form without the initials, e.g. Moore
- occasionally, conflict with particularly well established abbreviations used elsewhere, such as Copel. which is very widely adopted in pteridological literature for Copeland.
The following principles are applied only to names of authors not appearing in TL-2.
6. Surname only
A surname alone, or its abbreviation or contraction, is adopted as the standard form if it is applicable to only one author in the list. A surname alone, or its abbreviation or contraction, is usually also adopted for one of a number of authors with the same surname (see 7).
Persons with identical surnames are distinguished by use of initials of forenames, except as in 8 and 9.
Usually the earliest born is given without initials and all later ones with initials.
If one of the Authors is more renowned, then they may have their full name displayed, and subsequent authors will have initials
If all Authors have the same surname, and are contemporaries and are equally renowned, all may be given initials.
Where initials are required:
- if the author has one forename, then the one initial is given
- if the author has two forenames, both initials are given, except in occasional cases where an author consistently omits one initial in authorship of books or papers (as distinct from authorship of plant names), e.g. P.Taylor, H.Rob., B.Nord. for P.G. Taylor, H.E. Robinson and R.B.Nordenstam who consistently publish their work using only one of their forenames
- if the author has three forenames, three initials are given unless he or she has a clear preference for using only one and no ambiguity arises (L.A.S.Johnson is given all three initials to avoid confusion with Lennart Johnson and Leigh A. Johnson);
- if the author has more than three forenames, an ad hoc decision is taken. We recommend that no spaces be left after full stops.
8. Abbreviated or full forenames
When two authors have identical surnames and initials, full or abbreviated forenames may be used. One author may be given with only surname, or initial(s) plus surname, and the other with fuller name, or both may be given fuller names, e.g. Thomas Hogg (1777-1855) and Thomas Hogg (1820-1892) are distinguished as Hogg and T.Hogg respectively, while Walter Jones and William Jones have standard forms Walt.Jones and Wm.Jones respectively.
In a few cases persons with identical surnames may be distinguished by a suffix instead of, or in addition to, initials.
In well known cases of father and son, the son may be distinguished by ‘f.’, an abbreviation of ‘filius’. Where this follows an abbreviated name with a full-stop, we recommend that no space be left between full-stop and f., e.g. Rech.f. for Rechinger the son.
Tradition may also allow a different suffix, such as ‘Arg.’ for ‘Argoviensis’ in Müll.Arg. (Müller of Aargau).
In a few cases where different authors have identical surname and forename we have used the suffixes ‘bis’ for the second and ‘ter’ for the third, as in R.Br.,R.Br. bis and R.Br. ter for the three Robert Browns.
10. Variant names for the same person
Except as noted in 11, one person is always given the same standard form, regardless of whether they modified the spelling of their surname during their lifetime, or if different transliterations may exist, or a compounding form may have been adopted e.g. Meisner adopted the spelling Meissner later in life but is always given the same standard form, Meisn.
The variant transliterations Tsvelev, Tselov, Tzelev and Tzelov have all appeared for the same person, whose recommended standard form Tzelev should be used in all cases.
Alan Radcliffe Smith originally published plant names as A.R.Smith, but later modified his name to A. Radcliffe-Smith; the standard form Radcl.-Sm. should be used for both.
11. Different names for the same person
When a person has published under completely different names, different standard forms may be used for the same person.
For example, Inger Nordal has published under both her maiden name Björnstad and her married name Nordal, and both may be used in standard forms, as in Scadoxus pseudocaulus (I.Björnstad & Friis) Friis & Nordal.
However, where one person has simultaneously used alternative names, such as Brother Alain who also uses the names Liogier, Enrique Eugenio, and others, we have adopted a single name (in this case Alain, as agreed with him in correspondence).
The following principles concern when and how to abbreviate surnames.
It should be noted again that these have been applied only to names not given in TL-2 (see 5).
12. Where to abbreviate.
Names are never abbreviated before a consonant.
13. How many letters to save
- Names are usually not abbreviated unless more than two letters are eliminated and replaced by a full-stop.
- Names of authors living before the 20th century are more likely to be abbreviated than later ones, and names of authors in the later 20th century tend to be given in full.
- Where no strong tradition exists, names of eight letters or fewer are not abbreviated. Names of nine letters are abbreviated if more than three letters are eliminated (e.g. Verdc. for Verdcourt), and names of 10 letters or more are usually abbreviated. (Note, in the suggested abbreviations submitted by the compiler for Bryology there was a tendency for short names (seven letters or fewer) to be abbreviated where they probably would not have been abbreviated in the list for other groups. In the final editing some of these have been expanded to the full name to make them more comparable with others, but the tendency to greater abbreviation for Bryological authors may still be noted.)
- Other things being equal, if an abbreviation is made, two-syllable names are abbreviated to one syllable, three-syllable names are most likely to be abbreviated to one (Pennington to Penn.), and names of four or more syllables are most often abbreviated to two syllables (Tirvengadum to Tirveng.). Many Japanese names have four short syllables, and we have tried to be consistent in abbreviating them to two syllables (Kitagawa to Kitag., Hashimoto to Hashim., etc.) except where ambiguity would occur, or except where the name has only seven letters.
- Application of these principles may depend on whether initials are also needed in standard forms of well-known authors. Decisions may also be affected when an author is known to be commonly cited as a joint author with somebody else.
14. Compound names
Compound names, whether hyphenated or not, may be treated as special cases, and some principles given here may be discounted. In order to keep the standard form as short as possible, the principles of how many letters should be saved (see 13) may often be broken. Similarly, the principle that the same name must always be given in the same way (see 3 above) is over-ridden in compound names, so that, for example although Gonzales is given in full when it stands on its own, Gonzales Albo is abbreviated to Gonz.Albo. We recommend that no space be left after the full-stop.
About the Publications dataset
Abbreviations for periodical titles follow Botanico-Periodicum-Huntianum, Suppl. (BPH/S) (2004).
Non periodical publications published up to the 1940s follow the form given in Taxonomic Literature (TL2) and its supplements (1976-2009) except where the form is the same as another publication in which cases, the author of the work follows the form in square brackets.
Anything published after those dates or that does not appear in those resources will be given a standard form by the IPNI editors.
This form follows the suggested word abbreviations found in the appendix of BPH/S.
1. API - Currently there is no publicly available API. A beta version is being trialled right now and we are working towards providing this in the next 12 months.
2. Matching names to IPNI IDs - If you have a list of plant names for which you would like IPNI identifiers, you can use our Names Reconciliation Service.
3. We very much encourage the use of the IPNI Life Sciences Identifier (LSID) which are given on each name page.
4. RDF versions of records are available by adding /rdf to the end of a record page's URL. E.g.
or by sending an HTTP request with the header
Accept: application/rdf+xml to a record's LSID URL.