Material guide · 151
Material guide: nutmeg oil — same tree as mace, and the human poisoning record is filed under the other one
· 16 min read
CAS 8008-45-5, 117 suppliers. It comes from the same tree as the mace oil I wrote up earlier, Myristica fragrans, from two different tissues: nutmeg is the kernel, mace is the red aril wrapped around the seed. Put the two records side by side and the most important difference is not in the odour but in the toxicology field. Nutmeg oil carries only an oral rat LD50 of 2,620 mg/kg. Mace oil carries a rat figure of 3,640 mg/kg plus a human one: oral TDLo 400 mg/kg, with hallucinations, distorted perceptions, toxic psychosis and cardiac arrhythmias. The famous poisoning cases involve nutmeg, yet the human data sits on the 34-supplier mace record rather than the 117-supplier nutmeg one. Both records also give occurrence as nutmeg seed, while mace's own definition field says it is the layer wrapped around the seed: two fields of one record disagreeing. Its optical rotation is registered at +8 to +30, a range wide enough to work as a fingerprint of origin.
About a 5 minute read.
The short version
- Name: nutmeg oil, CAS 8008-45-5, 117 suppliers
- Distilled from the dried kernel of the ripe seed of Myristica fragrans
- Odour type spicy; odour: spicy, woody, nutmeg, powdery, terpenic, pine, old wood, resinous
- Medium strength (rank 2), substantivity 52 hours; FEMA 2793, and FEMA GRAS is its only clearance
- Appears in 20 of 954 formulas at a median dose of 1.05% (quartiles 0.80–2.00%)
- The human poisoning record is not on this material. It is on mace oil
- Both records give
occurrenceasnutmeg seed, and one of them is not the seed
One tree, two tissues, two records
I wrote up mace oil earlier. This is its twin, and the database keeps them as two entirely separate records:
| Nutmeg oil #27772 | Mace oil #23088 | |
|---|---|---|
| CAS | 8008-45-5 | 8007-12-3 |
| Suppliers | 117 | 34 |
| Part | dried kernel of the ripe seed | dried red aril around the seed |
| Substantivity | 52 hours | 64 hours |
| FEMA | 2793 | 2653 |
| Refractive index @20 °C | 1.47500–1.48800 | 1.47400–1.48800 |
| Formulas in the corpus | 20 | 1 |
The refractive indices are all but identical. Two tissues, one tree, and physical constants that overlap too far to tell them apart.
The odour fields do separate them. Nutmeg oil adds pine and resinous; mace oil adds mastic
and incense. Ashokkumar and colleagues compared the leaf, mace, kernel and seed oils of the same
trees directly in Nat Prod Res in 2022 (PMID 32515616).
The tissue difference is real; it simply does not show up in the refractive index.
The human poisoning record is filed under the other one
This is the thing I most want to get across this round.
Nutmeg oil's toxicology field:
Oral rat LD50 2,620 mg/kg. Behavioral: somnolence (general depressed activity).
Mace oil's toxicology field:
Oral rat LD50 3,640 mg/kg. Behavioral: somnolence.
Oral human TDLo 400 mg/kg. Behavioral: hallucinations, distorted perceptions. Behavioral: toxic psychosis. Cardiac: arrhythmias.
That human entry appears only on the mace record.
And the well-known poisoning cases involve nutmeg. Ichibayashi and colleagues published a case report of nutmeg poisoning with electrolyte abnormalities in Cureus in 2024 (PMID 39559662). The constituent responsible, myristicin, is present in both tissues.
If you are looking up the 117-supplier material, you never see that paragraph.
I do not know why the source data splits it this way. The practical reading is clear enough: treat the safety information for these two as one body, not as whatever happens to be on the record in front of you.
(TDLo means the lowest published toxic dose. It is neither a safe limit nor a lethal dose,
and the figure applies to the essential oil rather than to ground spice. Do not convert it into
anything in your kitchen.)
Two fields of one record disagreeing
Both records carry the same single line in occurrence:
nutmeg seed
For nutmeg oil that is right. For mace oil it is not, and its own definition field says so:
Essential oil steam-distilled from the dried red case which surrounds the ripe seeds of the nutmeg.
The tissue around the seed is not the seed.
I have spent several rounds on the exercise of checking one record's fields against each other, using boiling points and melting points. This time it is a text field, and it needs no arithmetic: the definition field says aril and the occurrence field says seed.
The likely cause is that the occurrence field was filled from the parent material, nutmeg, and never adjusted for the tissue.
Optical rotation as a fingerprint of origin
The optical_rotation field reads +8.00 to +30.00.
That is an unusually wide range. A single compound's optical rotation usually spans a few degrees; this spans twenty-two.
The reason is that this is a natural oil rather than one molecule. Optical rotation is the weighted sum of every chiral constituent, and East Indian and West Indian nutmeg differ considerably in composition. This field is not a specification; it is a signal about origin and batch.
In practice: if your lot measures +9 and the next measures +28, both are within spec and they will not smell the same. The specific gravity (0.88–0.91) and refractive index (1.475–1.488) ranges are far too narrow to catch that.
The production route is in the notes field
Most materials' notes fields hold one sentence or None found. This one holds a real process:
Nutmeg oil is produced by steam distillation or steam and water distillation of freshly comminuted, dried nutmegs. The nutmegs should preferably be free from most of their fixed oil prior to distillation. The fixed oil can be alcohol extracted…
"Free from most of their fixed oil prior to distillation" is a working detail: nutmeg kernels carry a lot of fat (nutmeg butter), and leaving it in hurts both the distillation and the product. That kind of thing rarely makes it into a database field.
Dose and company
Across 20 formulas:
| Dose | |
|---|---|
| Median | 1.05% |
| Quartiles | 0.80% – 2.00% |
| Max | 5.7% |
The overall median I measured is 2.11%, so this one runs clearly below the general level. Spice oils usually do; they cover other things too easily.
Most frequent company: geranyl acetate (10), clove bud oil (9), hedione (8), bergaptene-reduced bergamot (8), coumarin (8), iso e super (8). Clove bud oil in second place makes this a cologne and spicy oriental roster.
What this doesn't establish
- I did not compare the two oils' chemical compositions. For the tissue difference I cite Ashokkumar and colleagues, not my own measurement and not this database's fields (neither record carries a composition table).
- I did not go back to the source for that human TDLo. The field cites Food and Cosmetics Toxicology from 1964; I am citing the field.
- "Read the two safety records together" is my advice, not a regulatory requirement. How IFRA and national regulations treat these two is in their own documents.
- The optical rotation explanation (origin variation) is a general principle, not a measurement of this material. The field gives only a range, with no origins attached.
- The 20-formula merge covers four spellings. The corpus also holds one
nutmeg essenceand onenutmeg oil terpeneless, which are different things and are not counted.