Material guide · 116
Material guide: Peru balsam — its constituent list is a roll-call of molecules this series has already written up
· 18 min read
CAS 8007-00-9, shared by eight records: oil at 41 suppliers, the balsam itself at 29, absolute at 21, resinoid at 20. De Groot's 2025 analysis of the resin used to make patch-test allergens found the main volatiles to be benzyl benzoate, (E)-nerolidol, benzoic acid, benzyl alcohol, (E)-cinnamic acid, vanillin and (E)-benzyl cinnamate — four of which already have their own articles here. In dermatology it sits in the baseline patch-test series: Guarneri 2021 tested 12,030 patients with both Peru balsam and fragrance mix 1, each returning 3.6% positive, with only 1.0% overlapping — each catching about 2.5% the other missed.
About 6 minutes.
Short version
- Name: Peru balsam, from Myroxylon pereirae (Leguminosae), CAS 8007-00-9
- Odour type: balsamic — sweet, balsamic, cinnamyl, vanilla, amber, spicy, powdery, resinous
- Strength: medium, rank 2; substantivity 400 hours
- Suppliers: oil 41 (FEMA 2117, CoE 298), the balsam itself 29 (FEMA 2116), absolute 21, resinoid 20
- The resinoid record's notes field: "Do not use on any part of the body because of sensitizing potential"
- Oral rat LD50 > 5000 mg/kg, dermal rabbit > 10000 mg/kg
- It sits in dermatology's baseline patch-test series
Its constituent list is one you have already read
De Groot and colleagues did something practical in Contact Dermatitis in 2025: they analysed what is actually in the Myroxylon pereirae resin used to prepare commercial patch-test allergens, by headspace SPME with GC-MS and flame ionisation detection.
The main constituents are these seven:
| Constituent | Written up here? |
|---|---|
| Benzyl benzoate | Yes |
| (E)-nerolidol | Not yet |
| Benzoic acid | Not yet |
| Benzyl alcohol | Yes |
| (E)-cinnamic acid | Related: cinnamic alcohol, cinnamaldehyde |
| Vanillin | Yes |
| (E)-benzyl cinnamate | Not yet |
Four of the seven are molecules this series has covered one by one.
That explains why its odour field reads cinnamyl and vanilla — not as metaphor, but as composition. It also explains why it is used as a marker for fragrance allergy: it packages several known fragrance sensitisers inside one natural material.
The paper itself is chasing a different question. Peru balsam, colophonium and propolis are known to co-react in patch testing, and one candidate explanation is shared allergenic constituents. Their answer is negative:
The major ingredients of the volatile fractions of M. pereirae resin and colophonium are quite different; common haptens in volatile ingredients cannot readily explain co-reactivity.
A clean negative result. Two materials often go positive together, and the most intuitive explanation — overlapping composition — is ruled out by their own data.
Two tests, each catching 2.5% the other misses
In dermatology, Peru balsam is not a fragrance material but a reagent — long placed in the baseline patch-test series as a marker for fragrance allergy.
That position was recently questioned: since it is now barely used "as such", and fragrance mix 1 (FM1) appears more sensitive at detecting fragrance allergy, is testing it still worthwhile?
Guarneri and colleagues answered in Contact Dermatitis in 2021, retrospectively analysing 12,030 patients patch tested for contact dermatitis across 13 Italian dermatology clinics between January 2018 and December 2019:
| Positive | Share | |
|---|---|---|
| Peru balsam (MP) | 439 | 3.6% |
| Fragrance mix 1 (FM1) | 437 | 3.6% |
| Positive to both | 119 | 1.0% |
| MP only | 310 | 2.6% |
| FM1 only | 304 | 2.5% |
Each test catches 3.6%, and they overlap on only 1.0%.
Put another way: each test catches around 2.5% the other misses. Use one alone and you lose more than half the positives.
The paper also reports clinical relevance: among single sensitisations, 75.2% of MP reactions were clinically relevant (62.9% current, 12.3% past), against 76.3% for FM1. The authors conclude that Peru balsam is still worth testing alongside FM1 in the baseline series.
The implication for formulation is indirect but real: if you make something that stays on skin, a few percent of the population reacts to this class of molecule, and no single test finds all of them.
(The problem with "naturals are gentler" is measured in Synthetics give me headaches, so I only use naturals. Peru balsam is that article's strongest counterexample.)
The tomato thing
When Paulsen and colleagues wrote about tomato contact dermatitis in Contact Dermatitis in 2012, they noted something in passing:
The fruit contains fragrance compounds that are also present in Myroxylon pereirae (balsam of Peru), possibly accounting for cross-reactivity.
They patch tested an ether extract of tomato plants, and 8 of 93 patients (9%) reacted.
This is not a reason to worry about tomatoes. It is a reminder that these molecules are distributed more widely than you would guess, and that "fragrance constituent" and "food constituent" have no chemical boundary between them.
Eight records, two FEMA numbers
Under one CAS, 8007-00-9:
| Record | Suppliers | FEMA | Warning in notes? |
|---|---|---|---|
| Peru balsam oil | 41 | 2117 | No |
| Peru balsam (itself) | 29 | 2116 | No |
| Peru balsam absolute | 21 | — | No |
| Peru balsam resinoid | 20 | — | Yes |
| Resinoid replacer | 6 | — | — |
| Peru balsam specialty | 3 | — | — |
| Peru balsam replacer | 2 | — | — |
| Peru balsam oil replacer | 1 | — | — |
Only the resinoid carries the "do not use on any part of the body" warning, while the oil, the balsam and the absolute do not — despite sharing a CAS, a tree and nearly the same odour field.
Another instance of that field being inconsistent. The 46 records carrying the warning are collected in Forty-six records say keep it off your skin, whose whole point is that the presence or absence of that sentence cannot be used to judge safety.
(35.6% of records sharing a CAS disagree even on odour type — A third of the records sharing a CAS. These eight Peru balsam records are all balsamic, which is unusually consistent.)
How to use it
- A base note with 400 hours of substantivity. The sweet, vanilla and cinnamyl in the odour field make it easy to join gourmand and oriental structures; the database's blending suggestions list ylang ylang, patchouli, petitgrain, sandalwood, rose and spices.
- It is dark. The oil is a yellow amber-brown, the resinoid a dark brown viscous liquid. Factor the colour into a finished product.
- Check IFRA before putting it in anything leave-on. This is a restricted material, and the warning sentence appears on one of eight records.
- Do not infer gentleness from its being natural. Its constituent list is an assembly of known sensitisers.
- The flavour fields have content too: bitter, warm, balsamic, cinnamon, woody, vanilla, leathery, evaluated at 40 ppm. It carries FEMA and CoE numbers.
What this doesn't establish
- I have not smelled it. The whole article is fields, database-wide counts and three papers.
- De Groot's team analysed resin samples used to make patch-test allergens, not commercial fragrance-grade Peru balsam oil or absolute. The proportions may differ, and that is the biggest mismatch in this article.
- That paper measured the volatile fraction (headspace SPME). The non-volatile portion is not in that constituent list, and most of a resinoid's mass is non-volatile resin.
- Guarneri's 3.6% is a proportion of a patch-tested population — people tested because contact dermatitis was suspected — not a general-population prevalence.
- It is retrospective data from 13 Italian clinics. Exposure sources differ by region, so the figures need not extrapolate.
- I did not check the current IFRA standard for this material.
- The tomato paragraph is one sentence from Paulsen's paper. I did not check the primary data behind that cross-reactivity hypothesis.
- The eight-record table was pulled by name keyword and other naming may not have been caught.