Sources and Editorial Standards

How Tree TSAR Evaluates Evidence and Builds Taxonomic Treatments

A Tree TSAR page may draw on many kinds of sources, but citation count alone does not determine the strength of a treatment. A phylogenomic paper, nomenclatural index, protologue, regional flora, herbarium record, conservation assessment, and horticultural database can all be valuable while answering very different questions.

Tree TSAR is a curated synthesis. Sources are gathered, compared, and interpreted before they become part of an accepted treatment or public narrative. Important disagreements should remain visible rather than being concealed behind one authoritative-looking answer.

Start with the Question, Not the Database

Different sources are authoritative for different purposes. The International Plant Names Index, for example, supplies nomenclatural information such as spelling, authorship, types, and publication details, while explicitly distinguishing that function from the taxonomic judgments about accepted names and synonyms provided by resources such as Plants of the World Online.

A nomenclatural question may therefore require IPNI, the original publication, and the Madrid Code. A phylogenetic question may depend primarily on recent molecular studies. Conservation may require IUCN or a national authority; historical common names may be best documented in regional floras, linguistic works, or archival sources.

Tree TSAR does not assign one database the role of universal botanical authority.

Primary Scientific Literature Carries the Main Argument

When Tree TSAR makes claims about phylogenetic relationships, species boundaries, hybridization, morphology, ecology, biogeography, divergence history, or genomic evidence, primary scientific literature is preferred where available.

A database can show the outcome of a revision without revealing why it was proposed. The underlying paper can reveal sampling, analytical methods, support, conflicting datasets, morphological evidence, geographic coverage, and limitations.

Reviews, monographs, floras, and major syntheses remain valuable because they place individual studies into context. The objective is not to privilege primary papers mechanically, but to make important conclusions traceable to the evidence behind them.

Peer Review Is Important but Not Sufficient

Peer review substantially increases the evidentiary value of scientific literature, but it does not make every conclusion equally persuasive. Studies differ in taxon sampling, genomic coverage, marker choice, analytical assumptions, geographic representation, voucher documentation, and interpretation.

A technically strong phylogeny can also support more than one defensible rank treatment. Tree TSAR therefore evaluates both the evidence and the classification proposed from it.

A newly published treatment is not adopted merely because it is newer; an established treatment is not retained merely because it is familiar.

Authoritative Databases Are Evidence Sources, Not Master Records

Resources such as Plants of the World Online, World Flora Online, IPNI, regional floras, biodiversity portals, and institutional databases assemble information at scales that no single project could reproduce efficiently.

Tree TSAR depends on them while retaining editorial judgment. A database can lag behind recent research, adopt one of several defensible circumscriptions, inherit unresolved nomenclatural problems, or follow a taxonomic philosophy different from Tree TSAR’s.

This is especially important for synonyms. An external source can treat a Tree TSAR accepted taxon as synonymous with another name; that treatment is evidence to review, not permission for an automated process to erase the accepted Tree TSAR record.

The Code Governs Names, Not Taxonomic Opinion

The Madrid Code is the current international framework governing scientific names of algae, fungi, and plants. It addresses matters such as valid publication, priority, typification, legitimacy, conservation, rejection, and hybrid nomenclature (Turland et al. 2025).

The Code does not determine which competing biological circumscription Tree TSAR must accept. A nomenclaturally correct name can be applied to different taxonomic concepts by different authors.

Keeping nomenclature and taxonomy separate is therefore one of Tree TSAR’s fundamental editorial standards.

Historical Literature Still Matters

Taxonomy has a long memory. A nineteenth-century protologue may determine the application of a name still used today; an old flora may preserve a regional common name; a historical combination can become relevant again after a modern revision.

Historical literature should not be dismissed because it predates molecular systematics. Its role simply needs to be interpreted correctly. A historical source can be indispensable nomenclaturally while containing a classification no longer supported phylogenetically.

The age of a source changes the questions asked of it, not whether it has value.

A DOI Is an Identifier, Not a Seal of Approval

Tree TSAR includes DOIs when available because persistent identifiers make scholarly sources easier to locate and verify.

The presence of a DOI does not establish scientific quality. Crossref, one of the major DOI registration agencies, explicitly cautions that a Crossref DOI does not by itself certify peer review, authority, or content quality (Crossref 2019).

A reference therefore has to be checked for what it actually is and whether it supports the claim being cited.

How Conflicting Evidence Is Handled

Scientific disagreement is expected. Two studies may recover different phylogenetic relationships, or they may recover substantially the same tree and disagree only about rank. A dispute can also be nomenclatural rather than biological.

Tree TSAR first identifies what kind of disagreement exists. The narrative can then explain the principal alternatives and state why one treatment is adopted when the difference is consequential.

This is particularly important for genus and family boundaries, where several rank treatments can represent the same phylogenetic information.

Citations Should Support Claims, Not Decorate Paragraphs

Citation density follows the evidence rather than a fixed quota. Routine explanatory statements do not need a reference after every sentence, while consequential claims about phylogeny, nomenclature, conservation, distribution, or unusual biology should be traceable.

Tree TSAR also avoids turning narratives into a sequence of paper summaries. The biological question should lead the prose, with citations showing where the evidence comes from.

References follow the Tree TSAR Panax style: author-date citations in the text and detailed scientific references at the end of the page.

References Are Verified Before Publication

Bibliographic accuracy is part of scientific accuracy. Important references are checked for authors, title, year, journal or publisher, and DOI where one exists. The source must also be relevant to the sentence or concept it supports.

This is especially important when generative AI participates in research or drafting. AI systems can produce fluent but incorrect summaries and plausible-looking citations. Resnik & Hosseini (2026) discuss hallucinated citations as a scholarly integrity problem when fabricated references enter research workflows.

Tree TSAR therefore treats citation discovery and citation verification as separate tasks.

How Tree TSAR Uses Generative AI

Generative AI can assist with literature organization, comparison of competing treatments, extraction of structured information, consistency checking, drafting, and critique. It is not treated as an independent scientific source.

A public scientific claim should ultimately rest on traceable literature, specimens, authoritative databases, institutional resources, or another appropriate source. AI output does not become evidence because it sounds confident.

Editorial review also applies to the prose. Public narratives should read as coherent botanical writing rather than automatically assembled literature reviews. Repetitive sentence structures, boilerplate transitions, excessive rhetorical fragments, and long runs of one-line paragraphs are edited for clarity and natural flow.

Sources in Other Languages

Important botanical research and vernacular evidence are not confined to English. Tree TSAR considers non-English taxonomic revisions, floras, historical works, linguistic sources, and ethnobotanical literature when they can be identified and interpreted responsibly.

Translations and transliterations should preserve enough information about the source language and original term for the reasoning to be checked. A translation should not be presented as though it were the original vernacular expression.

Absence of Evidence Is Not Evidence of Absence

No literature search can guarantee complete coverage. Relevant research may be poorly indexed, published under an obsolete scientific name, contained in regional literature, or hidden in a paper whose primary subject is something else.

Tree TSAR therefore avoids claiming that no contrary evidence exists merely because none was found. Suggestions from specialists, regional botanists, horticulturists, herbarium workers, and knowledgeable readers can expose sources that ordinary searches miss.

Every taxon page provides a Suggest a Correction link for this reason.

Provisional Treatments Require Extra Transparency

Sometimes a phylogenetic placement is well supported before the necessary formal name or combination has been validly published. Tree TSAR can use [TSAR] to mark an explicitly provisional editorial placement when no suitable published name is available.

[TSAR] is not botanical authorship and does not constitute valid publication under the Madrid Code. Its purpose is to make the biological interpretation visible without misrepresenting its nomenclatural status.

Corrections Are Part of the Project

Tree TSAR is designed to be revisable. A correction may involve an authorship, overlooked synonym, newly published phylogeny, common-name interpretation, distribution statement, conservation assessment, broken link, or much larger reconsideration of taxonomic rank.

A submitted suggestion is evaluated rather than accepted automatically. When better evidence changes the strongest interpretation, the page should change with it.

The objective is not to maintain an appearance of never having been wrong. It is to maintain a reference capable of improving.

Commercial Support Does Not Determine Taxonomy

Tree TSAR may be supported through reader contributions, sponsorship, advertising, and affiliate revenue associated with Phytonomen resources. Those relationships remain separate from taxonomic and editorial decisions.

A sponsor, advertiser, publisher, nursery, retailer, or affiliate partner does not receive influence over accepted taxonomy, source selection, or scientific conclusions by supporting the project.

Likewise, inclusion of a book in the Phytonomen Bookstore does not imply that Tree TSAR accepts every taxonomic treatment contained in it.

What Tree TSAR Means by “Authoritative”

Authority is contextual. An institution can be authoritative for a type specimen, a nomenclatural record, a conservation assessment, a cultivated-plant registry, or a regional flora without becoming the final authority on every question involving that plant.

Tree TSAR’s role is to identify which source is appropriate to the question and then place that information alongside the other relevant evidence.

This avoids two opposite errors: dismissing major databases because they occasionally contain outdated treatments, and treating those databases as if they could never be questioned.

Traceability Is the Standard

Readers do not have to agree with every Tree TSAR treatment. They should be able to determine why it was adopted.

Significant claims should lead to evidence, major disagreements should be identified, provisional treatments should be labelled, and the distinction between Tree TSAR editorial judgment and formal nomenclatural action should remain clear.

Scientific classification changes because evidence changes and because researchers can interpret the same evidence differently. Tree TSAR’s editorial standard is therefore that a treatment should be well supported, clearly reasoned, traceable to its evidence, and capable of revision when better information becomes available.

Explore Further

About Tree TSAR - The project’s broader purpose and scope.
Taxonomy and Nomenclature - The distinction between biological classification and the rules governing names.
Synonyms, Hybrids, and Search - How alternative names remain connected to accepted records.
Suggestions and Corrections - How readers can submit evidence or report problems.

References and Further Reading

Crossref (2019) Myth busting in Mumbai: Crossref is not a mark of publisher or content quality. Crossref. https://www.crossref.org/blog/myth-busting-in-mumbai/

International Plant Names Index (n.d.) International Plant Names Index. Royal Botanic Gardens, Kew; Harvard University Herbaria; Australian National Herbarium. Accessed 17 August 2026. https://www.ipni.org/

Resnik DB, Hosseini M (2026) Hallucinated citations produced by generative artificial intelligence may constitute research misconduct when citations function as data in scholarly papers. Accountability in Research. https://doi.org/10.1080/08989621.2026.2645390

Turland NJ, Wiersema JH, Barrie FR, Gandhi KG, Gravendyck J, Greuter W, Hawksworth DL, Herendeen PS, Klopper RR, Knapp S, et al. (eds.) (2025) International Code of Nomenclature for algae, fungi, and plants (Madrid Code). Regnum Vegetabile 162. University of Chicago Press. https://www.iaptglobal.org/madrid-code-online