Peach Kernel Oil
Peach kernel oil is rated Caution, on a comedogenic score of 2 out of 5. The unusual thing about that number is who it belongs to. It is usually pinned to James Fulton's 1989 rabbit-ear survey — but Fulton has no peach kernel oil row. He tested apricot kernel oil. The material actually named "peach kernel oil" was tested by a different lab, Morris and Kwan, in 1983, who scored it 2 on their own 0 to 5 scale. So the number survives, for a reason almost nobody citing it gives.
Oleic-rich kernel oil; low-to-moderate legacy clog score, commonly cautioned for acne-prone skin.
What it is
The oil pressed from the kernel inside a peach stone — the seed, not the fruit. On a label it is Prunus Persica (Peach) Kernel Oil. It is one of a cluster of chemically related but materially distinct stone-fruit kernel oils, alongside apricot (Prunus armeniaca) and sweet almond (Prunus amygdalus), that sit in the same emollient role in a formula and are close enough in composition to be confused for one another — a confusion that matters here, because it is exactly what happened to this ingredient's score. Chemically it is dominated by oleic acid. Chamli and colleagues, characterising Tunisian peach varieties in 2017, report roughly 68 to 75% oleic acid, 16 to 22% linoleic acid, and 5 to 7% palmitic acid, with the exact split varying by variety and growing region. It contains no lauric or myristic acid to speak of — the short and mid-chain saturated fatty acids that sit at the top of the comedogenic legacy scores. A formulator reaches for it as a light, spreadable, oleic-rich carrier oil, much as they would for apricot or almond.
Why we rate it Caution
We rate it Caution at Low confidence. The score behind that rating is 2 out of 5, and the point of this page is where that 2 actually comes from. Search this ingredient and you will be told it scores 2, and the number is usually attributed, directly or by implication, to Fulton's 1989 rabbit-ear table — the source of most comedogenic numbers in circulation. We rendered that table at 400dpi and read it row by row. It has no peach kernel oil row. What it has, in Section VI, "Oils," is "Apricot kernel oil — 2(1)." Apricot, not peach. The two oils are close relatives and are routinely treated as one, but they are not the same material, and reading a peach score off an apricot row is precisely the near-miss substitution this site exists to catch. We had been doing it ourselves. So does any paper actually test peach kernel oil? One does. Morris and Kwan, in the Journal of the Society of Cosmetic Chemists in 1983 — six years before Fulton — ran a rabbit-ear survey on their own 0 to 5 scale, and their table includes a row named "Peach Kernel Oil," scored 2, applied undiluted. That is the exact material, named in full, and it is the source the number on this page rests on. Reported as it should be, it is 2 of 5, Morris and Kwan 1983. What that 2 is worth is harder to say than the lists imply, and we are not going to borrow an answer. Fulton published a key to his own grades; Morris and Kwan's verbal key is not something we hold, so we do not call their 2 borderline, moderate or slight. It is a 2 on their five-point scale, and that is the whole of what the number states. For position within their own table and nothing more: they scored olive oil and avocado oil 0, and sweet almond oil 3. The rating is Caution because of what the evidence is, not because a 2 falls in anyone's band. The only direct measurement of this oil we could trace is a single small historical rabbit-ear result, from one laboratory, on an oil the paper does not characterise — no variety, no origin, no refining state, no purity. Nothing we found has replicated it, and nothing has tested the oil on human skin. A result like that is not a finding of comedogenicity and not a clearance either, and Caution is the word this site uses for that position. The rest of the legacy record does not speak to it. Fulton's 1989 table has no peach kernel oil row, his earlier 1984 survey with Pay and Fulton has none, and the 2007 Nguyen, Dang and Maibach rabbit study did not test it.
What the evidence doesn’t tell you
This rests on a single rabbit-ear assay from 1983, and nothing else we could find measures the same thing. The rabbit ear is not a human face: in Morris and Kwan's method the material goes into the external ear canal of three rabbits, undiluted here, and follicles are graded microscopically for keratosis. A 2 in that model is a statement about rabbit follicles, not a prediction about any particular person's pores, and not a measurement of a finished product at the concentration a formula would actually use. We identified no human comedogenicity study of peach kernel oil — no cyanoacrylate follicular biopsy, no counted comedones on human skin, no clinical trial on acne-prone faces. The confident "peach kernel oil is non-comedogenic" that circulates on retailer and blog pages is not attached to any study we could trace, and we do not repeat it; an unsourced reassurance is the same failure as an unsourced alarm, pointed the other way. What we could trace is one 1983 rabbit score of 2 out of 5, which is neither a clearance nor a finding. The oil's composition invites a mechanism, and we are deliberately not building one. Peach kernel oil is oleic-rich, and oleic acid has been argued in the literature to disrupt the skin barrier; it would be easy to spin that into a story about why a 2 is "really" a concern. But Morris and Kwan measured this oil directly and got a 2, and a chain of reasoning about oleic acid cannot raise or lower a number that was actually read off a rabbit's ear. The rating rests on the measurement, not on the fatty-acid profile. The rabbit-ear assay itself is contested. Draelos and DiNardo's 2006 re-evaluation in the Journal of the American Academy of Dermatology tested finished products on small human panels and found that an ingredient's isolated rabbit score is a poor guide to how the finished product behaves — a finding about formulations, not a head-to-head human re-test of this oil. A 2025 review in JAAD Reviews, surveying the field from 1972 onward, found that no standardised comedogenicity test exists at all and that "non-comedogenic" is an unregulated label. Those caveats apply to this 2 exactly as they apply to every other number in the corpus.
Where you’ll see it
On a label it appears as Prunus Persica (Peach) Kernel Oil, most often as an emollient carrier oil in facial oils, massage and body oils, balms and moisturisers, where it stands in for apricot or sweet almond kernel oil. Those are the categories it is marketed into; we have not audited its prevalence, and position on an ingredient list does not tell you the concentration. Keep it separate from its relatives, because conflating them is how its score went wrong. Apricot Kernel Oil (Prunus Armeniaca Kernel Oil) is the material Fulton actually tested at 2(1), and it has its own entry. Sweet Almond Oil (Prunus Amygdalus Dulcis Oil) is a third relative, scored 3 by Morris and Kwan, and also has its own entry. Three related oils and three numbers — but not three rows in one paper: peach and sweet almond are Morris and Kwan rows, apricot is a Fulton row, and no single paper tested all three.
Sources
Each source says what it actually is. Several widely-cited “sources” for comedogenic ratings are republishing the same 1989 assay, and counting them as independent agreement is how a thin evidence base gets made to look thick.
- Morris WE, Kwan SC. "Use of the rabbit ear model in evaluating the comedogenic potential of cosmetic ingredients." J Soc Cosmet Chem 34:215–225 (1983) The primary source of this page's number, and the only paper we could find that tested the exact material. An independent rabbit-ear assay on its own 0 to 5 scale, six years before Fulton 1989 and not a republisher of him — its authors contradict Fulton in print, in their discussion, on other materials. Method: rabbit ear canal, three rabbits, materials applied undiluted unless a dilution is stated, fourteen applications, microscopic grading. Its table has a row named "Peach Kernel Oil," scored 2, tested undiluted. Cited for that row. Its oils table is printed rotated in the journal and returns nothing under OCR, which is one reason so few pages quoting a peach number have actually read it.
- Fulton JE. "Comedogenicity and irritancy of commonly used ingredients in skin care products." J Soc Cosmet Chem 40:321–333 (1989) The table most comedogenic scores descend from, cited here for what it does NOT contain. Read off a 400dpi render of the journal scan. There is no peach kernel oil row anywhere in it. Section VI, "Oils," prints "Apricot kernel oil — 2(1)": a different material, where the "(1)" is Fulton's notation for a result obtained "using refined oil," and where the section heading itself carries his asterisk, "Results depend on source of raw material." So the row the peach number is usually read off is not peach, and even for apricot it is a qualified figure. Method: roughly 10% in propylene glycol, inner ear of three New Zealand albino rabbits, follicles scored for keratosis. Cited as the source of a misattribution, not as evidence for this rating.
- Chamli D, Bootello MA, Bouali I, Jouhri S, Boukhchina S, Martínez-Force E. "Chemical characterization and thermal properties of kernel oils from Tunisian peach and nectarine varieties of Prunus persica." Grasas y Aceites 68(3):e211 (2017) Peer-reviewed lipid analysis, and the source of the fatty-acid composition on this page — oleic roughly 68 to 75%, linoleic 16 to 22%, palmitic 5 to 7%, across Tunisian peach and nectarine varieties. Nothing to do with pores; cited because a composition claim needs a composition source, and this is one we can name. The ranges vary by variety and origin, which is why they are given as ranges.
- Draelos ZD, DiNardo JC. "A re-evaluation of the comedogenicity concept." J Am Acad Dermatol 54:507–512 (2006) A Current Issues and Opinion piece in JAAD that tested finished products on small human panels and found products testing non-comedogenic despite containing ingredients the rabbit assay had scored comedogenic. Cited narrowly, because it cuts against a reassuring reading of a low score as hard as it cuts against an alarming reading of a high one: it is evidence that isolated rabbit-ear ingredient scores transfer poorly to finished formulas. It is not a human re-test of peach kernel oil, and we do not present it as one.
- "Comedogenicity in cosmeceuticals: a review of clinical relevance, regulatory gaps, and future directions." JAAD Reviews (2025) A recent peer-reviewed review of the whole field, covering the literature from 1972 onward. It is a synthesis of the legacy literature, not a new measurement, and independent of Fulton only in authorship, since the corpus it surveys includes his. Finds that no standardised comedogenicity test exists and that "non-comedogenic" is an unregulated label. Cited for the field-wide caveat that sits under every number on this site, this 2 included.
Others in the same family
Oils behave similarly enough that the evidence for one is often wrongly read across to the others. These are its structural relatives, not a guess. What the evidence says about oils as a family ›
Last reviewed 2026-08-12 · How we decide