Most bonsai content online is recycled folklore — repeated until it sounds true, but rarely sourced. We started this site to be the opposite of that. Below are the rules we hold ourselves to. They exist so you can verify our claims, push back where we’re wrong, and trust that an updated guide means we learned something — not that we changed the SEO.
§ 1The editorial promise
Everything else in this document is operational detail. The four promises below are the foundation that the rest exists to make true.
Promise 1 — Every factual claim has a named source
“Water bonsai once a week” is a claim. So is “junipers tolerate Zone 4 with siting”, “ASPCA lists ficus as toxic to dogs”, and “the chokkan form was codified in the Edo period”. Each of those needs a source we can name and you can verify. Where we couldn’t find a reliable source, we either ran the experiment ourselves and published the data — see /data/ — or we wrote nothing.
The threshold is conservative. “Many growers report” is not a source. “It’s commonly said” is not a source. “In our experience” is a source only when the experience is from a named contributor with verifiable credentials.
Promise 2 — Domain experts review their domain
An ISA-certified arborist reviews every page that touches tree health, pruning, or root work. A state-licensed horticulturist reviews soil, fertiliser, propagation, and our /data/ research. A DVM reviews every page where a species touches pet toxicity — cross-checked against the ASPCA poison-control database and the Pet Poison Helpline. Each reviewer is named, with credentials and a contact link on their bio page.
Promise 3 — Updates are visible, not silent
Every guide carries a “last reviewed” date. When we change a recommendation — when new research overturns a previous claim, when a reader catches an error, when a regulation changes — we update the page and log the change publicly on the correction page. We do not silently rewrite. Search results that link to our older language should still find the language, with a note explaining what changed and why.
Promise 4 — Money disclosed, never hidden
We accept no sponsored content, no paid placements, no link exchanges, and no “guest posts” from anyone we have not vetted. Where we use affiliate links to products we genuinely recommend, the affiliate relationship is disclosed at the top of the page (not in fine print), and the disclosure links to our full affiliate policy. If we ever break this rule, it will be by mistake — and we want you to tell us so we can fix it.
§ 2Source hierarchy
We classify sources into three tiers. The tier determines how a claim is cited (inline link vs end-of-article reference) and whether a claim can stand alone or requires corroboration. The full annotated list of 81 verified sources lives on the sources page.
A single Tier 1 citation is sufficient to make a claim. Examples we cite regularly:
- USDA — Plant Hardiness Zone Map (2023 revision), PLANTS Database, NRCS
- ASPCA Animal Poison Control toxic plant database; Pet Poison Helpline (855-764-7661)
- University extension publications — Cornell Cooperative Extension, Oregon State University Extension Service, UF/IFAS Extension, Missouri Botanical Garden Plant Finder, UC ANR, NC State Extension, Penn State Extension, Royal Horticultural Society (UK)
- Peer-reviewed journals — HortScience, Journal of the American Society for Horticultural Science (JASHS), HortTechnology, Scientia Horticulturae, Arboriculture & Urban Forestry (AUF), Urban Forestry & Urban Greening (UFUG), Journal of Environmental Horticulture (JEH), Arboricultural Journal
- International conservation bodies — Xerces Society (pollinators), Lady Bird Johnson Wildflower Center (natives)
A Tier 2 source can support most claims and is sufficient alone for claims of institutional fact (e.g. exhibition history, curator statements, collection composition). For empirical claims (light requirements, watering frequency), Tier 2 should accompany or be checked against a Tier 1 source. Examples:
- Museums with named curators we cite — National Bonsai & Penjing Museum (Michael James); Pacific Bonsai Museum (Aarin Packard); Brooklyn Botanic Garden / C.V. Starr (David Castro); Larz Anderson / Arnold Arboretum (Peter Del Tredici); North Carolina Arboretum (Arthur Joura, since 1992); Chicago Botanic Garden (Chris Baker); Longwood Gardens; Omiya Bonsai Art Museum (Dr. Fumiya Taguchi); Morton Arboretum; Morikami Museum
- Governing bodies & federations — Nippon Bonsai Association; World Bonsai Friendship Federation (WBFF); Bonsai Clubs International (BCI); American Bonsai Society (ABS, est. 1967); Federation of British Bonsai Societies (FoBBS); National Bonsai Foundation (NABF); European Bonsai Association (EBA); Potomac Bonsai Association; Golden State Bonsai Federation; Blue Ridge Bonsai Society
A Tier 3 source is sufficient for technique and aesthetic claims attributed to the master, where the master is the original source of the technique or interpretation. For empirical claims, Tier 3 supports but does not replace Tier 1. The 18 masters we cite by name include:
- Living masters — Walter Pall (naturalistic yamadori); Ryan Neil (6-year Kimura apprenticeship, conifers); Jonas Dupuich (Bonsai Tonight, pines); Michael Hagedorn (Crataegus Bonsai, deciduous); Bjorn Bjornholm (Kouka-en lineage); Suthin Sukosolvisit (tropicals); Marc Noelanders (European styling); Peter Chan (21 RHS Chelsea golds); Boon Manakitivipart (US-based Japanese training)
- Editor-historians — William N. Valavanis (Cornell-trained, International BONSAI magazine 1979–2020, 164 issues)
- Historical canon — John Y. Naka (Bonsai Techniques I & II, Goshin); Yuji Yoshimura (Longwood 1959, first widely-translated English-language manual); David De Groot (Pacific Rim)
- Specialist contributors we cite — Arthur Joura (NCA); Chris Baker (Chicago); Michael James (NBPM); Dr. Fumiya Taguchi (Omiya); Peter Del Tredici (Arnold Arboretum)
The site maintains a blocked-sources list of 57 domains — sites that have published demonstrably false bonsai advice, plagiarised primary sources without attribution, used AI-generated images presented as real specimens, or that are link-spam farms. We do not cite from them and we do not link to them.
§ 3How we cite
Citations exist for the reader, not for the writer. The rule of thumb: a curious reader should be able to verify a claim in two clicks or fewer. We use two citation modes — inline links and end-of-article references — chosen to keep the prose readable while making verification trivial.
Inline links — for direct factual claims
A factual claim with a single authoritative source uses an inline link directly under the claim. The link text describes the source, not just “study” or “research” — the reader knows where they’re going before they click.
Japanese maples (Acer palmatum) are hardy to USDA Zone 5 with reliable winter dormancy, per the 2023 USDA Plant Hardiness Zone Map.
End-of-article references — for synthesised or contested claims
When a claim draws on multiple sources, or when a topic has genuine expert disagreement, we use a numbered reference at the claim and a “References” block at the end of the article. The reference block includes author, title, publication, year, and a direct URL or DOI where available.
¹ Roberts, K. et al. (2019). “Substrate particle size and water retention in container production of Acer palmatum.” HortScience 54(6): 1042–1047. doi:10.21273/HORTSCI13892-19
When the source is one of our own contributors
Where an experienced contributor’s first-hand observation is the source, we say so. “In 12 years of training shimpaku junipers at our studio, [Contributor Name] has observed that…” is cite-able because the contributor is named and their experience documented on their bio page. This is Tier 3 — it supports but does not replace empirical sources.
When we couldn’t find a source
If a question has no published answer, we run the experiment ourselves and publish the dataset openly under CC BY 4.0. See our original research — eight datasets to date including akadama bulk density across six brands, watering frequency across 24 species over twelve months, and a pet-toxicity reference cross-checked against ASPCA. The full method is documented on the methodology page.
§ 4Who reviews what
Editorial responsibility is named, not “our team”. Every page on this site is owned by a specific named editor and reviewed by at least one named domain expert before publishing. The table below shows which reviewer is responsible for which content category, and what the review actually checks.
| Content category | Primary reviewer | What the review checks |
|---|---|---|
| Species hubs (botanical profile, ID, hardiness, varieties) | Licensed Horticulturist | Latin name validity, USDA zone accuracy against 2023 map, hardiness range, light tolerance, mature size, growth rate, propagation viability. |
| Species toxicity callouts (every species page) | DVM | ASPCA cross-check, Pet Poison Helpline cross-check, toxic compound identification, expected clinical signs, severity rating, emergency contact accuracy. |
| Care guides (watering, fertilising, repotting, placement, seasonal) | ISA Arborist + Horticulturist | Method accuracy against extension publications, species-specific deviations flagged, seasonal timing per zone, before-and-after expectations realistic. |
| Styling & design (forms, wiring, deadwood, pruning) | Senior Practitioner + Arborist | Form definitions cite master canon (Naka, Yoshimura, Valavanis); technique correctness; structural pruning safety; wire choice and removal timing. |
| Problem solving (diagnostics, pests, diseases, rescue) | Arborist + Horticulturist | Symptom-to-cause mapping accuracy, differential diagnosis completeness, treatment safety (chemicals, dilutions), recovery expectations honest. |
| Propagation & growing (seed, cuttings, layering, grafting) | Horticulturist | Success-rate claims sourced; rooting-hormone usage cited; timing per zone correct; growing-on transition steps realistic. |
| Tropical species (ficus, jade, fukien tea, etc.) | Tropical Specialist | Florida-based first-hand expertise; light claims for true indoor cultivation; humidity requirements honest; pest pressure realistic. |
| Yamadori & collected trees | Yamadori Collector | Legal collection requirements per region, after-care timeline honest, species suitability for collection, ethics of collection. |
| Original research (/data/ pages) | Horticulturist + Senior Practitioner | Method documented per methodology page, sample size justified, statistical claims honest, raw dataset published under CC BY 4.0. |
| Inspiration & culture (masters, history, gardens, museums) | Senior Practitioner | Names and dates verified against museum/governing-body records; quotes cited; gallery photography credited. |
The full workflow — from brief to published page — is documented on the review process page. Briefly: an author drafts → a fact-checker cross-checks every cited claim against the linked source → a domain reviewer signs off → a copy editor reviews voice and clarity → the page is published with a visible “last reviewed” date and the names of everyone who reviewed.
§ 5What we don’t do
Naming what we will not do is at least as important as naming what we will. Most of the bonsai content problem on the internet comes from one of the practices below. If you see us doing any of these, please tell us.
§ 6Review cadence
Bonsai evidence is not static. The 2023 USDA Plant Hardiness Zone Map moved roughly half of US counties up by half a zone. The Pet Poison Helpline updates its toxicity database continuously. New peer-reviewed substrate research is published most years. Our review cycle reflects this.
Annual full review — Cornerstone guides
The site’s cornerstone guides — first bonsai, watering, soil, styles, repotting, diagnostic — get a complete editorial review at least once every twelve months. The review re-validates every external citation, checks for newer research that changes a recommendation, and verifies that all internal links still resolve. The review date is visible on the page.
Quarterly refresh — Problem and seasonal pages
Diagnostic and seasonal-calendar pages are refreshed quarterly. Seasonal pages are reviewed before each relevant season (a winter dormancy page is reviewed in October, not in March). Diagnostic pages are refreshed when symptom patterns shift — typically after the spring growth flush and again after summer pest pressure.
Event-triggered review — Everything else
Any page can be re-reviewed at any time if any of the following happens:
- A cited source publishes new findings that change a recommendation
- A reader flags an apparent error and the flag survives editorial check
- A regulation changes (USDA zone update, ASPCA listing change, IUCN status change)
- A named reviewer revises their view based on new field experience
- A linked external resource moves or disappears (broken-link audit, monthly)
Every page on the site exposes its current review status in the page metadata — last review date, next scheduled review date, and a link to the relevant correction-log entry if any. If a page has not been reviewed in over twelve months, that fact is shown prominently at the top of the page.
§ 7Correction policy
We will make mistakes. The question is what happens when one is caught. Our policy is the opposite of the bonsai industry norm — we log the change publicly, we keep the original wording cite-able, and we don’t pretend it was always this way.
What gets logged
Three categories of change trigger a public log entry on the correction page:
- Factual corrections — a previous claim was wrong; the page now says something different. The old wording, the new wording, the source that prompted the change, and the date of correction are all logged.
- Substantive updates — a previous claim was correct as of the last review but is now outdated. The change is logged with the date and the source that necessitates the update.
- Recommendation changes — we previously recommended X, we now recommend Y. The log explains why and when. Examples include changes to our preferred akadama brand, repotting timing for specific species, or first-bonsai species recommendations as we learn what actually survives.
What does not get logged
Routine copy editing — fixing a typo, improving sentence flow, adding a paragraph break — does not get a log entry. The goal is a useful change history, not a noisy one. If you ever can’t tell whether a change was substantive, our default is to err on the side of logging it.
Old language stays cite-able
If you arrived at one of our pages from a year-old article that cited specific language we have since changed, the correction log lets you find the original wording. We don’t try to erase what we used to say. Errors are how a reference grows.
Two species hubs listed Juniperus chinensis “Shimpaku” as USDA Zone 5. The 2023 USDA Plant Hardiness Zone Map confirms it is reliably Zone 4 with appropriate winter siting. The shimpaku and procumbens pages now show Zone 4–9. Source: USDA PHZM 2023 revision, cross-checked against Missouri Botanical Garden Plant Finder.
§ 8Style & voice
This section is more practical than philosophical. These are the conventions every page on the site follows, so the site reads as one consistent reference rather than a collection of freelance contributions.
Voice
Authoritative but not pompous. We are writing for a curious adult reader — someone who can handle a Latin name, a USDA zone, and a citation, but who does not want to feel quizzed. Where we use a technical term, we define it inline the first time. Where we have an opinion, we say it is an opinion. Where the evidence is thin, we say it is thin.
Language
US English. “Color” not “colour”, “fertilizer” not “fertiliser”, “favorite” not “favourite”. This document is one of the few exceptions and uses UK English for editorial consistency with the original spec — the live editorial guides on the site are US English. Date format is Month DD, YYYY (May 28, 2026).
Latin names & species nomenclature
First mention of any species uses the binomial Latin name in italics, with the common name first: “Japanese maple (Acer palmatum)”. Subsequent mentions use the common name. Cultivars are in single quotes after the species: Juniperus chinensis ‘Shimpaku’. Where a common name is ambiguous (there are several “jade bonsai” species), we use the Latin name on every mention.
Numbers & units
Temperatures: Fahrenheit first, Celsius in parentheses where the original source uses Celsius. “75°F (24°C)”. Lengths: imperial first for hobbyist measurements (pot dimensions, tree height), metric first for scientific measurements (substrate particle size, soil pH). USDA hardiness zones are written as “Zone 5” not “USDA 5”. Watering frequency: never as days-per-week (“once a week”). Always as a description of soil state (“when the top of the substrate is dry to a depth of one centimetre”).
Headings, lists, and tables
Headings are sentence-case, not title-case (“How we cite”, not “How We Cite”). Lists are used when content is genuinely enumerable; not when prose would read better. Tables are used for comparison content where the reader will scan across rows — species comparisons, zone tolerances, brand rankings.
Image & caption standards
Every photograph carries an alt text describing what is visible to a screen-reader user. Photographs of named specimens carry a caption with the species, cultivar, approximate age, and photographer. Stock photography is not used. Illustrations are labelled as illustrations.
Honest framing rules
Three phrases that signal honest framing the site uses regularly:
- “In our experience” — used only when the named contributor’s experience is the source. Not used as a hedge.
- “The evidence is thin” — used when a topic has no good source. Better than pretending one exists.
- “This will not work indoors” — used directly, not as a hedged possibility. Most bonsai die from being told they are indoor plants when they are not.
§ 9Flag an error
If you find an error on this site — a fact we got wrong, a citation that doesn’t support the claim, a broken link, a recommendation that contradicts current research, an attribution we missed — please tell us. Corrections from readers are how this reference improves.
The fastest route is the contact form with “Correction” in the subject. If you’d rather email directly, the editorial team reads editorial@artofbonsai.org. We aim to acknowledge correction submissions within 72 hours and to publish the result within 14 days — sooner for anything safety-related (pet toxicity, pesticide dilutions, structural pruning that could kill a tree).
The clearest correction submissions include: (1) the URL of the page in question, (2) a quotation of the wording you think is wrong, (3) what you believe the correct claim is, and (4) a source we can verify. None of those are required — even just “I think this is wrong, please check” is welcome — but the more you can give us, the faster the correction gets reviewed.
We credit substantive corrections on the correction-log entry. If you’d rather remain anonymous, we honour that — just say so in the submission.
Reviewed by the Lead Editor with the full editorial team. Next scheduled review: May 2027. Annual cycle. Substantive changes from v3.1: added explicit AI-image policy under § 5, expanded the source-tier examples in § 2 with named museums and curators, added the broken-link audit cadence under § 6.
Earlier versions: v3.1 (Nov 2025) · v3.0 (May 2025) · Full version history
