Castor Oil
Castor oil is one of the few ingredients where the number circulating online is real, checkable, and low. Fulton's 1989 table prints it as a 1 — a grade he himself called "not considered significant" — and a second, independent rabbit assay from a different lab scored it 0 to 1. Two legacy assays agreeing is rare here. It is still two rabbits' ears, and we can find no test of this oil on a human face.
Viscous but low-comedogenic oil widely used in cleansing balms and lip products.
What it is
The oil pressed from the seed of Ricinus communis. It is chemically unlike every other oil in this database: roughly ninety percent of its fatty acids are ricinoleic acid, an eighteen-carbon monounsaturate carrying a hydroxyl group mid-chain. That hydroxyl is why castor oil is thick, why it is far more polar than a typical vegetable oil, and why it is the only commercial oil built on a hydroxylated fatty acid. Formulators use it as an occlusive skin-conditioning agent, as a solvent and wetting agent for pigment — which is why it turns up in lip colour — and, in its salt and ester forms, as a surfactant. Hydrogenated castor oil, a separate ingredient, is the hardened version, used to thicken. One thing worth clearing away first, because it is the fear people actually arrive with — and we have to be more careful with it than we were. Ricin is water-soluble, so it does not partition into the oil the way the fat-soluble components do; it stays with the protein-rich seed meal. That is the cosmetics industry's account, and we previously wrote it as a flat "castor oil does not contain ricin." It is not that clean. McKeon, Patfield and He tested actual castor oil samples by ELISA in 2016 and found no measurable toxin in the well-processed grades — but they DID find ricin in the COLD-PRESSED sample, at 35 ± 13 µg/L. They are emphatic that this is far below anything toxic, and we are not raising an alarm: the point is that "properly refined castor oil showed no detectable ricin in this study" and "castor oil does not contain ricin" are different sentences, and only one of them survives contact with the measurement. Cold-pressed castor oil is sold directly to consumers. All of which is a question about poison, not about pores, and it has no bearing on our rating.
Why we rate it Clear
We rate it Clear. The evidence is two rabbit-ear assays, run six years and one laboratory apart, which agree. (We previously said twelve years. Morris and Kwan is 1983 and Fulton is 1989. It is an embarrassing thing to get wrong on a page about people not checking numbers.) Fulton's 1989 table — the source nearly every comedogenic number online descends from — prints Castor oil at 1. We read the row off a render of the journal scan rather than off a list quoting it. It matters what a 1 means in Fulton's own words, because the lists that publish his numbers leave his key out: "a minimal grade of 0 to 1 is not considered significant. Grade 2 to 3 is borderline. However, a grade of 4 to 5 is uniformally reproduceable and considered positive." A 1 is not a small amount of clogging. In the paper's own terms, it is nothing. The neighbouring rows hold: hydrogenated castor oil 1, PEG-8 castor oil 1, PEG-40 castor oil 0. Morris and Kwan, publishing in the Journal of the Society of Cosmetic Chemists in 1983, ran their own rabbit-ear assay on 32 materials at a different laboratory, and scored Castor Oil at 0 to 1. This is worth dwelling on, because it almost never happens on this site. Most of what looks like corroboration in the comedogenic-rating world is one Fulton number quoted by four websites. Here there are genuinely two measurements, made independently, and they land in the same place. Morris and Kwan were not deferential to Fulton — the same paper openly contradicts him on sodium lauryl sulfate and says so in print — so their agreement on castor oil is agreement, not repetition. That is the whole case, and it is stronger than most ratings here have behind them. It is also two rabbits' ears.
What the evidence doesn’t tell you
We can find no human study of castor oil's comedogenicity. Not on a face, not on a back, not with a comedone count. Everything above happened inside the ear of a rabbit, at roughly 10% in propylene glycol in Fulton's case, and Fulton's own oils section carries a footnote — "Results depend on source of raw material" — attached to the section heading, meaning it applies to castor oil as much as to the butters. A 1 from a batch of castor oil in 1989 is not a promise about the bottle in your hand. The thing that makes castor oil chemically interesting is also the thing nobody appears to have followed up. Its ricinoleic acid is hydroxylated, which makes it unusually polar for an oil, and Fulton's own generalisation from his 220-row survey was that materials soluble in both oil and water tend to be less comedogenic than materials soluble in only one. Those two facts sit next to each other and invite a tidy mechanistic story. Nobody we can find has tested that story on castor oil, and we are not going to write it as though somebody had, and a mechanism that explains a result is not additional evidence for it. There are two randomised human trials of topical castor oil on skin that grows sebaceous glands, and both point the opposite way from "clogging" — which is exactly why we are going to be careful with them rather than enthusiastic. Goto and colleagues, in Ophthalmology in 2002, ran a randomised, double-masked, placebo-controlled crossover trial of eye drops containing 2% castor oil in twenty patients with non-inflamed obstructive meibomian gland dysfunction. Meibomian glands are sebaceous glands, and the condition is, loosely, blocked gland orifices. Castor oil drops improved the obstruction. Now the caveats, all of which are load-bearing: this was the eyelid margin, not facial skin; 2% in an emulsion, not neat oil; the endpoints were tear-film measures and gland orifice appearance, not comedones; there were twenty people; and the most likely explanation is lipid and lubrication at the gland opening, which has little to do with follicular keratinisation. It is a real human result on a sebaceous structure and it is not a comedogenicity study. Both halves of that matter. The second trial we had missed, and it is the stronger of the two on exposure. Muntz and colleagues, in The Ocular Surface in 2021, ran a prospective, investigator-masked, randomised paired-eye trial in twenty-six people with blepharitis, using 100% COLD-PRESSED CASTOR OIL applied to the periocular skin and lid margin. Neat oil, not a 2% emulsion, on skin, in a randomised trial. Blepharitis improved. The same caveats apply — eyelid margin rather than face, the endpoints are lid and tear measures rather than comedone counts, twenty-six people — but it removes the objection we had leaned on hardest against Goto, which was that nobody had put the undiluted oil on skin. Somebody has, in a randomised trial, and the glands got better rather than worse. And the field-wide caveat, which applies here as everywhere: a 2025 review in JAAD Reviews, surveying the literature from 1972 onward, found that no standardised comedogenicity test exists and that "non-comedogenic" is an unregulated label. Castor oil's rating is well-supported by the standards of this subject. That is a remark about the standards of this subject. Separately, and not a pore question: castor oil and its derivatives do appear in the contact-allergy literature — the Cosmetic Ingredient Review's 2025 re-review cites case reports of allergic contact cheilitis traced to ricinoleic acid in lipsticks, and positive patch tests to castor oil and hydrogenated castor oil. Rare, and a different mechanism entirely. It belongs on the page because a reader who reacts to a lip product deserves to know the literature exists, not because it bears on the rating.
Where you’ll see it
Lipsticks and lip balms above all — castor oil wets and disperses pigment, which is why it is a standard base for lip colour — along with skin-care products and bath soaps. It appears on labels as Ricinus Communis (Castor) Seed Oil. Its relatives are separate ingredients with separate entries and separate scores: Hydrogenated Castor Oil (a thickener, Fulton 1), PEG-8 and PEG-40 castor oil (solubilisers, Fulton 1 and 0), and the ricinoleate salts and esters, which act as surfactants and emulsifiers. The Cosmetic Ingredient Review has assessed the whole group and concluded it safe as used — a statement about safety, not about pores, and the two are not the same question.
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.
- Fulton JE. "Comedogenicity and irritancy of commonly used ingredients in skin care products." J Soc Cosmet Chem 40:321–333 (1989) The primary source of our legacy score and of nearly every comedogenic number in circulation. Method: the ingredient at roughly 10% in propylene glycol, applied to the inner ear of New Zealand albino rabbits, three rabbits per assay, follicles scored for keratosis. Its Table I prints Castor oil 1, Hydrogenated castor oil 1, PEG 8 castor oil 1, PEG 40 castor oil 0. Read off a render of the journal scan, not off a list quoting it. It also supplies the grade key the lists omit — 0 to 1 "is not considered significant," 2 to 3 is "borderline," 4 to 5 is "uniformally reproduceable and considered positive" — and the section footnote on all oils and butters: "Results depend on source of raw material."
- 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) A second, independent rabbit-ear assay from a different laboratory, and the reason this page can say "two studies agree" rather than "one study, quoted often." Three rabbits, ear canal, 14 applications, microscopic grading, 0–5 scale. It scores Castor Oil at 0 to 1. It is not a deferential republication of Fulton — the same paper flatly contradicts him on sodium lauryl sulfate and prints the disagreement — which is what makes its agreement here worth something. Still a rabbit's ear, still not a person.
- McKeon TA, Patfield SA, He X. "Evaluation of Castor Oil Samples for Potential Toxin Contamination." J Am Oil Chem Soc 93:299–301 (2016) Cited to correct ourselves, and it has nothing to do with pores. We had written flatly that "castor oil does not contain ricin." These authors measured it — ELISA on cold-pressed, USP-grade and neutralised oils — and found no measurable toxin in the well-processed grades but 35 ± 13 µg/L of ricin in the COLD-PRESSED sample. They stress it is far below any toxic dose, and so do we. It is here because a confident absolute we could not support is exactly the thing this site exists to object to, including when we are the ones saying it.
- Muntz A, Sandford E, Claassen M, et al. "Randomized trial of topical periocular castor oil treatment for blepharitis." Ocul Surf 19:145–150 (2021) The trial we had missed, and on exposure it is the stronger of the two: 100% COLD-PRESSED CASTOR OIL — neat oil, not a 2% emulsion — applied to the periocular skin and lid margin of 26 people with blepharitis, in a prospective, investigator-masked, randomised paired-eye design. It improved. This matters because our strongest reservation about Goto was that nobody had put undiluted castor oil on skin in a controlled trial. Somebody had. It is still the eyelid margin rather than the face, and its endpoints are lid and tear measures, not comedone counts — it is not a comedogenicity study, and we will not promote it into one.
- Goto E, Shimazaki J, Monden Y, et al. "Low-concentration homogenized castor oil eye drops for noninflamed obstructive meibomian gland dysfunction." Ophthalmology 109(11):2030–2035 (2002) ONE of two randomised human trials of topical castor oil on a sebaceous-gland-bearing site — we previously called it the only one, and missed Muntz 2021 below. Double-masked, placebo-controlled, crossover; 20 patients, 40 eyes; eye drops containing 2% castor oil with a polyoxyethylene castor oil emulsifier, six times daily, in two two-week periods. Meibomian gland orifice obstruction improved. Cited as a loose end and not as evidence for the rating: meibomian glands are sebaceous, but the eyelid margin is not facial skin, 2% in an emulsion is not neat oil, the endpoints were tear-film and orifice measures rather than comedone counts, and the plausible mechanism has nothing to do with follicular keratinisation.
- Cosmetic Ingredient Review (Fiume MM, Bergfeld WF, Belsito DV, et al.). "Re-Review Summary of Ricinus Communis (Castor) Seed Oil and Ricinoleates as Used in Cosmetics." Int J Toxicol 44(3 Suppl):117S–122S (2025); original final report Int J Toxicol 26(S3):31–77 (2007) An independent expert-panel safety review, not a comedogenic list and not downstream of Fulton. It concludes castor oil and the ricinoleates are safe as used in cosmetics, describes their functions as skin-conditioning agents, emulsion stabilisers and surfactants, and is the source of the contact-allergy case literature noted on this page. It says nothing at all about comedogenicity — which is itself worth knowing. Full text paywalled; we read the abstract and the panel's published conclusion, and we are not quoting figures from inside a document we have not opened.
- Cosmetics Info (Personal Care Products Council), "Ricinus Communis (Castor) Seed Oil" The cosmetics industry's own public ingredient database, run by the Personal Care Products Council — an industry body, and we treat it as one. Cited for two things a trade source is actually a reasonable authority on, and which we could not source from the paywalled CIR report: the composition (roughly 90% of the fatty acids in castor oil are ricinoleic acid, a monounsaturated 18-carbon fatty acid) and the product categories it is formulated into (lipstick, skin care, bath soaps). Also the source of the plain statement that castor oil contains no ricin, because ricin is water-soluble and does not carry into the pressed oil. Not cited for any claim about safety or comedogenicity, where its interests are not neutral.
- "Comedogenicity in cosmeceuticals: A review of clinical relevance, regulatory gaps, and future directions." JAAD Reviews (2025) 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.
- CosDNA and the comedogenic-rating charts generally Republishers of the legacy assay scores, not independent measurements — and on this ingredient they happen to be passing along a number that is real and that checks out against the paper. That is worth saying plainly, because the point of this site is not that the charts are always wrong. It is that they do not show their work, so a reader has no way to tell the times they are right from the times they are not. Castor oil is one of the times they are right.
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 2026-07-13 · How we decide