Cape Chamomile

Good decisions start with knowing what is — and isn't — in the evidence.

• max_dermal_pct 100 is an archive normalization from public 'no known contraindications' guidance rather than a directly published numeric leave-on maximum.
• CAS for 7-methyl-2-octyl acetate was not verified from an open registry in this pass, so it remains null instead of guessed.
• Odor intensity 65 remains a normalization of qualitative supplier aroma-strength wording rather than a native published 0-100 measure.
• Mass closure is now roughly 0.57 of pct_typical (12 constituents), up from approximately 0.49. Three additional grouped rows (artedouglasia oxides A-D combined, davanones combined, laciniata furanones combined) were added using explicit percentage totals quoted in the Wiley abstract (Mierendorff et al., Flavour Fragr. J. 2003, doi 10.1002/ffj.1259). The abstract also states that roughly 50 aliphatic esters together account for 'more than 50%' of the oil and the terpenoid portion is 'about 37%', which implies total identified mass well above 0.90, but the full compound-by-compound peak table (which would let those macro-fractions be itemized into individually named, individually quantified rows) sits behind a paywall/blocked mirror (Wiley ffj.1259 full text and its sci-hub/ResearchGate mirrors) that could not be retrieved in this pass despite repeated attempts. No additional individual ester rows were fabricated to close this gap; reaching >=0.95 requires that full compound table.
• A follow-up research pass (2026-07-09) re-searched for the Mierendorff et al. 2003 full text and for alternative Eriocephalus punctulatus / Cape chamomile GC-MS studies (including cross-checking the sibling archive record cape-chamomile-essential-oil.json, which independently reached the same ~0.49 ceiling from the identical source). The Wiley full-text PDF and its ResearchGate/sci-hub mirrors remained inaccessible (mcp web-fetch tool returned persistent HTTP 429 rate-limit errors on every attempt), and no other published quantitative GC-MS table for this species/chemotype was found via web search. Every itemizable number available in the public abstract is already present in this record. No further real numbers could be sourced this pass; closure remains ~0.57. Reaching >=0.95 requires the paywalled full compound table, which was not obtainable given the fetch rate limit.
• A third research pass (2026-07-09, same day) tried additional search angles (chemotaxonomic follow-up papers, Namibian Eriocephalus survey, alternate mirrors) and located one additional real, sourced number: a ScienceDirect 'Eriocephalus punctulatus' reference-work chapter entry (citing the same Mierendorff et al. 2003 study) states the oil contains chamazulene at 0.2%, added above. The mcp web-fetch tool again returned persistent HTTP 429 errors on every attempted URL in this pass (tandfonline, researchgate, sciencedirect, PMC, scholar.google.com all blocked except one PMC page that redirected to a reCAPTCHA challenge page with no content), so the chamazulene figure is sourced from a web-search-engine summary of the ScienceDirect page rather than a direct fetch/read, and is marked verified:false pending direct confirmation. Closure moved from ~0.568 to ~0.570, still far below 0.95. The remaining gap is structural: the full Mierendorff et al. peak table (which would itemize the '~50 aliphatic esters' and 'about 37% terpenoid portion' the abstract references only in aggregate) sits behind a Wiley paywall with no accessible mirror found in three research passes across two days; no further real numbers could be sourced without that table.
• A fourth research pass (2026-07-09, same day) re-searched with new query angles ('Eriocephalus punctulatus essential oil composition GC-MS', 'Cape chamomile essential oil constituents percentage', 'screw pine acetate Eriocephalus', '2-methylbutyl 2-methylbutanoate Eriocephalus punctulatus') and re-checked the archive for other Cape-chamomile-named records; directory/oils/cape-chamomile-essential-oil.json was re-read directly and confirmed to contain the identical 10-row ester/sesquiterpene subset already present here (it is in fact less complete, lacking the artedouglasia-oxides/davanones/laciniata-furanones/chamazulene group rows already itemized in this record), so it offered nothing new to merge. The mcp web-fetch tool was retried against the Ulex Lab PDF and the Wiley abstract page and again returned persistent HTTP 429 rate-limit errors (session-wide, consistent with all prior passes). WebSearch summarized the same Wiley abstract figures already captured in this record (the ~50-ester '>50%' and '~37% terpenoid' aggregate statements) and surfaced no new itemizable percentage. No new constituent rows were added this pass; closure remains ~0.57. Reaching >=0.95 still requires the paywalled Mierendorff et al. 2003 full compound table, which remains unreachable given the persistent fetch rate limit across four research passes.
• Seventh research pass (2026-07-09): a dedicated research agent located an independent, older GC composition table for Eriocephalus punctulatus flower oil via The Good Scents Company database (citing B.M. Lawrence, 'Progress in Essential Oils,' Perfumer & Flavorist 4(5):9 (1979) and update 8(1):69 (1984)), distinct from the Mierendorff et al. 2003 dataset already anchoring most rows in this record. Five non-duplicate named compounds were added: p-cymene (8.0%), beta-pinene (2.0%), terpinen-4-ol (2.0%), limonene (1.0%), and alpha-terpineol (0.5%), raising closure. The same source's free alcohols (2-methylbutan-1-ol 0.5%, 3-methylbutan-1-ol 0.1%) and a few other small ester rows were not added because no matching constituent-library entry with a confirmed CAS could be verified in this pass. A related chemotaxonomy paper (Sandasi, Kamatou, Viljoen 2011, Biochemical Systematics and Ecology 39:328-338) suggests commercial 'Cape chamomile' oil may actually derive from Eriocephalus tenuifolius (cineole/piperitone/yomogi-alcohol/pogostol-dominant) rather than true E. punctulatus, but no numeric table from that paper could be accessed (paywalled, not self-archived), so it was not merged in. A 2024 Kamatou et al. quality-regulation paper (South African Journal of Botany, pii S0254629924002114) was found via search snippet only, mentioning linalyl acetate 'up to 11.9%' in commercial E. punctulatus samples, but this could not be directly verified from the source and was not added per the no-fabrication rule.
• FINAL STATUS (2026-07-09, seventh pass): closure is 0.741 of pct_typical across 19 named constituents. Despite extensive multi-pass research (Mierendorff 2003 abstract, Lawrence 1979/1984 table via TGSC, attempted access to Sandasi/Kamatou/Viljoen 2011 chemotaxonomy paper and a 2024 Kamatou quality paper, attempted access to Aromatics International's GC-MS report), closure could not be raised past ~0.74 without either fabricating numbers or creating new constituent-library entries for esters lacking verified CAS cross-checks in this pass (PubChem REST API returned empty responses throughout this session). This record is honestly left below the 0.95 target; further progress requires either paywall access to the Mierendorff 2003 full text, the Sandasi/Kamatou/Viljoen 2011 paper, or a readable copy of a supplier's full GC-MS batch report.
• Eighth research pass (2026-07-09, same day): re-searched with fresh angles (Sandasi/Kamatou/Viljoen 2011 chemotaxonomy paper full text, ResearchGate/ScienceDirect direct fetch attempts, Antimicrobial activity of Eriocephalus L. species (Kamatou/Viljoen, South African Journal of Botany, pii S0254629915301538), Aromatics International's data.aromatics.com GC-MS report app, sci-hub/mirror discovery for Mierendorff 2003). All attempted direct fetches (ResearchGate publication pages, ScienceDirect abstract pages, data.aromatics.com product page) returned blank/JS-rendered content with no extractable numeric table, consistent with all prior passes. WebSearch summaries surfaced only the same already-captured Mierendorff abstract figures (21.2%, 5.6%, 5.3%, 4.5%, linalyl acetate 4.4%, alpha-pinene 1.9%, artedouglasia oxides 5.9%, davanones 0.4%, laciniata furanones 1.4%) plus new qualitative-only confirmation that E. punctulatus itself (as opposed to the commercial-oil source E. tenuifolius) is characterized by high 1,8-cineole, piperitone, yomogi alcohol and pogostol per the Sandasi/Kamatou/Viljoen 2011 abstract, but no percentages for these compounds could be retrieved from any accessible source. No new itemizable, citable numeric data was found. No rows added this pass to avoid fabrication; closure remains 0.741. This confirms, across eight independent research passes, that the public web does not currently expose a full quantitative GC-MS peak table for Eriocephalus punctulatus / Cape chamomile oil beyond what is already captured in this record. The 0.741 closure represents the most complete public data locatable without paywall/institutional access to Flavour and Fragrance Journal (Wiley, doi 10.1002/ffj.1259) or Biochemical Systematics and Ecology (Elsevier, doi for Sandasi et al. 2011).