Forty-six records have a notes field that says "do not use on any part of the body" — and bergamot oil and jasmine absolute are on the list

About 4 minutes.

Most materials keep their risk information in the GHS or toxicity columns, where you have to read a grade and interpret it.

But one field issues an instruction outright — and because it is free text, nobody searches it.

Sweeping the notes field

I swept the notes field across all 42,225 records for the various phrasings of "do not use on the body".

Forty-six records.

And there are only two reasons: phototoxic and sensitizing.

The phototoxic group

"This oil should not be used anywhere on the body because of its phototoxic effects."

Material	Suppliers
Cumin seed oil	65
Bergamot oil	58
Angelica root oil	46
Tagete oil (South Africa)	35
Cumin oleoresin	17
Tagete oil (Egypt)	14
Citrus paradisi peel extract	10
Marigold absolute	9
Verbena oil (France)	9
Fig juice concentrate ("extreme phototoxic")	6
Rue oil, pomelo essence oil and others	0–5 each

Most are Apiaceae and Rutaceae plants — they contain furocoumarins, molecules that absorb UVA and then react with DNA in skin. The citrus line is taken apart in Why citrus oils are phototoxic; this list extends it to cumin, angelica, tagete, verbena and fig.

The fig records are worth noting — their wording is "extreme phototoxic", and the three fig-related records are the only ones in the database that use "extreme".

The sensitizing group

"Do not use on any part of the body because of sensitizing potential."

Material	Suppliers
Jasmin absolute (India, from concrete)	34
Treemoss absolute	29
Costus root oil	21
Peru balsam resinoid	20
Jasmin oil	13
Oakmoss concrete	8
Treemoss concrete	7
Verbena absolute (France)	6
Inula helenium root extract	5
Opoponax resin	5
Costus root absolute / resinoid	3 each

Jasmin absolute has 34 suppliers, and the notes field tells you not to put it on skin.

Several in this group have their own articles here: oakmoss (the atranol regulatory story) and costus root oil, whose notes field is the bluntest I have read.

The sentence is blunt, and you should know its limits

To be clear: this is not "these materials cannot be used".

notes is free text, and it issues a binary instruction while the actual rules are graduated:

• IFRA restricts bergamot oil; it does not prohibit it. It is a lead material in a great many perfumes.
• Furocoumarin-removed versions exist (usually labelled FCF or bergapten-free), and their phototoxic risk is different.
• Limits vary by product category — leave-on versus rinse-off — and this field does not address that layer at all.
• Which records carry the sentence is inconsistent. All four regional bergamot oil records have it; most other citrus oils in the same database do not.

So the correct use is: treat it as a prompt to go and check IFRA, not as a verdict.

(How to read the database's regulatory fields, and what they cannot answer, is in Those 80 "not used anymore" materials.)

How dermatology sees it

De Groot and Schmidt opened a series in Dermatitis in 2016, and the first part is an introduction. Their opening gives the scale:

Essential oils are widely used in the flavor, food, fragrance, and cosmetic industries in many applications. Contact allergy to them is well known and has been described for 80 essential oils.

The same paragraph carries a line I find even more worth remembering:

The relevance of positive patch test reactions often remains unknown. Knowledge of the chemical composition of essential oils among dermatologists is suspected to be limited, as such data are published in journals not read by the dermatological community.

Neither side is reading the other's data. Composition data goes into fragrance-chemistry journals; allergy data goes into dermatology journals. What that series does is carry the first to the second.

A later part of the series covers sandalwood oil, ylang-ylang oil and jasmine absolute — which is exactly one of the entries above.

What to do

• Read the notes field before buying a natural. It is free text, so a search will not surface it. You have to open the record.
• When you see this sentence, look up the relevant IFRA standard. Limit, product category, and whether a treated version exists.
• Phototoxic and sensitizing are different things. Phototoxicity needs light; sensitisation is an immune response requiring prior exposure. They are handled differently.
• For citrus, look for FCF / bergapten-free grades. Other phototoxic materials do not necessarily have a treated equivalent.
• Leave-on products (perfume, lotion) and rinse-off products (soap, shampoo) have different standards.
• Do not skip this step because it is natural. All 46 are natural extracts.

What this doesn't establish

• This is a field sweep, not a safety assessment. I counted what the database says, not any material's actual risk.
• "Forty-six" depends on my match string. I used regular expressions for "not be used … on … body" and "do not use on any part of the body"; other phrasings may be missed.
• The absence of the sentence does not mean safe. This field's coverage is very low; most records' notes are empty or say something else.
• I did not check IFRA's text for any of them. What I say about bergamot above is industry common knowledge, not a check against the standards' current version and figures.
• De Groot's paper is an introduction, and I quote two of its statements. The "80 essential oils" figure is theirs; I did not go back to that list of 80.
• Patch-test prevalence figures are not discussed here. Those come from populations tested because contact dermatitis was suspected, not from the general population.
• I did not check record by record how many of the 46 also carry a GHS classification. The two fields are filled in independently.