Fraxin

Fraxin at a Glance

Fraxin is a coumarin glucoside — fraxetin with a sugar attached — found in horse-chestnut and ash bark, and grouped with esculin as the coumarin fraction of horse-chestnut seed.

Like the other coumarin glucosides here it functions largely as a poorly absorbed precursor to its aglycone. Its research is entirely preclinical.

What Belongs to Fraxin and What Doesn’t

Fraxin is a minor molecule surrounded by better-known neighbours, and the main risk on a page like this is that it inherits their properties. Three claims sit adjacent to fraxin and belong to something else.

Not esculin’s toxicity. Esculin is the coumarin named as a reason raw horse chestnut is not eaten and the one reduced during aescin standardisation. Fraxin shares the bark and the chemical class and none of that history: no isolated-toxicity signal appears in the reviewed literature, and in every injury model below it is protective rather than harmful ^{1Reference 1Ferdous J et al. · 2024ReviewPharmacological Activities of Plant-Derived Fraxin with Molecular Mechanisms: A Comprehensive Review — reviewView study →}.

Not the uricosuric effect. Cortex Fraxini has a long anti-gout reputation, and there is real rodent evidence for it — but the study that produced it used the whole coumarin fraction of ash bark against oxonate-induced hyperuricemia, not isolated fraxin ^{11Reference 11Li JM et al. · 2011AnimalProtective effects of Cortex Fraxini coumarines against oxonate-induced hyperuricemia and renal dysfunction in mice — mouse modelView study →}. Ash bark carries fraxin, fraxetin, esculin and esculetin together, and the fraction’s activity has never been apportioned between them. Fraxin should not be called uricosuric.

Not warfarin’s mechanism. Fraxin’s aglycone is a 7,8-dihydroxy-6-methoxycoumarin. The anticoagulant coumarins are 4-hydroxycoumarins, a different substitution pattern with a different target, so the “coumarins thin the blood” claim does not reach any molecule in this set — the same discipline that applies to scopoletin and esculetin.

What is left is fraxin’s own: a reproducible rodent NF-κB story, hepatoprotection in toxin models, and a newer ferroptosis result. And one honest complication that runs the other way — because fraxin is a glucoside, some of what is credited to it in vivo is probably its aglycone’s work, which makes the isolate question below unusually literal.

What It's Used For

Fraxin is a coumarin glucoside — fraxetin 8-O-glucoside — and one of the minor coumarins of horse-chestnut and ash bark, grouped with esculin as the seed’s coumarin fraction. Its research is entirely preclinical, and as a glucoside much of the in-vivo activity is plausibly its aglycone fraxetin’s.

Evidence by IndicationStrength of support · Beta

01Anti-inflammatoryShaky32%Rodent lung, gut and brain injury models ^{2,3,4,5Reference 2Li W et al. · 2019AnimalFraxin ameliorates lipopolysaccharide-induced acute lung injury in mice by inhibiting the NF-κB and NLRP3 signalling pathways — mouse modelView study →Reference 3Ma X et al. · 2019AnimalFraxin Alleviates LPS-Induced ARDS by Downregulating Inflammatory Responses and Oxidative Damages and Reducing Pulmonary Vascular Permeability — animal modelView study →Reference 4Sun W et al. · 2024AnimalFraxin Ameliorates Ulcerative Colitis by Modulating Oxidative Stress, Inflammation and TLR4/NF-κB and MAPK Signaling Pathways — animal modelView study →Reference 5Yao H et al. · 2022AnimalProtective effects of fraxin on cerebral ischemia-reperfusion injury by mediating neuroinflammation and oxidative stress through the PPAR-γ/NF-κB pathway — animal modelView study →}

02HepatoprotectiveShaky26%Two CCl₄/chemical hepatotoxicity models ^{6,7Reference 6Niu X et al. · 2017In vitroHepatoprotective effect of fraxin against carbon tetrachloride-induced hepatotoxicity in vitro and in vivo through regulating hepatic antioxidant, inflammation response and the MAPK-NF-κB signaling pathway — in vitro and animal modelView study →Reference 7Chang BY et al. · 2017AnimalFraxin Prevents Chemically Induced Hepatotoxicity by Reducing Oxidative Stress — animal modelView study →}

03Antioxidant / Nrf2Shaky22%Cell protection + coumarin Nrf2/ARE chemistry ^{1,8Reference 1Ferdous J et al. · 2024ReviewPharmacological Activities of Plant-Derived Fraxin with Molecular Mechanisms: A Comprehensive Review — reviewView study →Reference 8Whang WK et al. · 2005In vitroNatural compounds, fraxin and chemicals structurally related to fraxin, protect cells from oxidative stress — in vitroView study →}

04Bone protectionShaky18%An emerging rodent osteoporosis/OA cluster ^{9,10Reference 9Zheng X et al. · 2024AnimalDelay the progression of glucocorticoid-induced osteoporosis: Fraxin targets ferroptosis via the Nrf2/GPX4 pathway — animal modelView study →Reference 10Qian Z et al. · 2021AnimalFraxin Prevents Knee Osteoarthritis through Inhibiting Chondrocyte Apoptosis in an Experimental Rat Osteoarthritis Model — rat in vivo animal modelView study →}

Mechanisms

All mechanisms are preclinical (rodent/cell).

Pharmacokinetics

Fraxin is a coumarin O-glucoside, so it follows the pattern of its class: poorly absorbed intact, and hydrolysed by gut and microbial β-glucosidases to its aglycone fraxetin — the same arrangement as esculin → esculetin.

One experiment measures all four molecules at once, which makes the pattern hard to argue with. Beagle dogs given a single oral dose of a Ledum palustre extract reached peak plasma concentrations of 67.04 ± 12.09 ng/mL of fraxin against 369.7 ± 48.87 ng/mL of fraxetin — the aglycone five and a half times higher — while in the same animals esculin reached 46.75 ng/mL against esculetin’s 209.9 ng/mL ^{12Reference 12Wang Z et al. · 2018Simultaneous Determination of Aesculin, Aesculetin, Fraxetin, Fraxin and Polydatin in Beagle Dog Plasma by UPLC-ESI-MS/MS and Its Application in a Pharmacokinetic Study after Oral Administration Extracts of Ledum palustre L. — animal pharmacokinetic modelView study →}. Fraxin peaked earliest of the four, at 0.83 ± 0.18 hours.

The consequence is direct: an oral dose of fraxin delivers mostly fraxetin, so the rodent results above are, in exposure terms, partly fraxetin results, and no study in this literature separates the two. There is no human pharmacokinetic data for either.

Clinical Trials

There are no human clinical trials of isolated fraxin. Human exposure occurs only as an untargeted minor constituent of Cortex Fraxini and horse-chestnut preparations, where any activity belongs to the whole extract.

Last checked: September 2026.

Isolate vs. Plant Studies

The isolate/plant question has an extra layer here, because there are three things it could mean and the answers differ.

Isolate vs. whole extract. Every pharmacology result above used purified fraxin at a stated rodent dose ^{2,3,4,5,6,7,9,10Reference 2Li W et al. · 2019AnimalFraxin ameliorates lipopolysaccharide-induced acute lung injury in mice by inhibiting the NF-κB and NLRP3 signalling pathways — mouse modelView study →Reference 3Ma X et al. · 2019AnimalFraxin Alleviates LPS-Induced ARDS by Downregulating Inflammatory Responses and Oxidative Damages and Reducing Pulmonary Vascular Permeability — animal modelView study →Reference 4Sun W et al. · 2024AnimalFraxin Ameliorates Ulcerative Colitis by Modulating Oxidative Stress, Inflammation and TLR4/NF-κB and MAPK Signaling Pathways — animal modelView study →Reference 5Yao H et al. · 2022AnimalProtective effects of fraxin on cerebral ischemia-reperfusion injury by mediating neuroinflammation and oxidative stress through the PPAR-γ/NF-κB pathway — animal modelView study →Reference 6Niu X et al. · 2017In vitroHepatoprotective effect of fraxin against carbon tetrachloride-induced hepatotoxicity in vitro and in vivo through regulating hepatic antioxidant, inflammation response and the MAPK-NF-κB signaling pathway — in vitro and animal modelView study →Reference 7Chang BY et al. · 2017AnimalFraxin Prevents Chemically Induced Hepatotoxicity by Reducing Oxidative Stress — animal modelView study →Reference 9Zheng X et al. · 2024AnimalDelay the progression of glucocorticoid-induced osteoporosis: Fraxin targets ferroptosis via the Nrf2/GPX4 pathway — animal modelView study →Reference 10Qian Z et al. · 2021AnimalFraxin Prevents Knee Osteoarthritis through Inhibiting Chondrocyte Apoptosis in an Experimental Rat Osteoarthritis Model — rat in vivo animal modelView study →}. The one whole-extract result — the anti-hyperuricemia work — is explicitly a coumarin fraction study, and it stays off fraxin’s ledger ^{11Reference 11Li JM et al. · 2011AnimalProtective effects of Cortex Fraxini coumarines against oxonate-induced hyperuricemia and renal dysfunction in mice — mouse modelView study →}. So the attribution is clean in that direction.

Isolate vs. its own aglycone. This is the one that is not clean. Fraxin is poorly absorbed and cleaved to fraxetin, which reaches roughly five times the plasma concentration from the same dose ^{12Reference 12Wang Z et al. · 2018Simultaneous Determination of Aesculin, Aesculetin, Fraxetin, Fraxin and Polydatin in Beagle Dog Plasma by UPLC-ESI-MS/MS and Its Application in a Pharmacokinetic Study after Oral Administration Extracts of Ledum palustre L. — animal pharmacokinetic modelView study →}, so “isolated fraxin, given orally” and “fraxetin, delivered slowly” are hard to tell apart. No study dosed both and compared. It is a real limit on every in-vivo claim here — although the in-vitro cell-protection work ^{8Reference 8Whang WK et al. · 2005In vitroNatural compounds, fraxin and chemicals structurally related to fraxin, protect cells from oxidative stress — in vitroView study →}, where no gut is involved, does belong to fraxin as such.

Isolate vs. the herb people actually use. Nobody takes fraxin. It reaches people as one minor coumarin in ash bark or horse-chestnut material, at unquantified levels, and no clinical trial of either preparation exists to transfer.

Its Role in Plants

Fraxin has the most interesting plant biology of any molecule in this set, and it is about iron.

Plants growing in neutral or alkaline soil face a supply problem: iron is abundant but locked up as insoluble Fe³⁺. Arabidopsis solves it by secreting coumarins from its roots, and the key one is fraxin’s aglycone. The enzyme scopoletin 8-hydroxylase converts scopoletin to fraxetin, and it is strongly induced in the roots of iron-starved plants; mutants that cannot make it secrete less fraxetin and grow badly on alkaline media with immobilised iron, while overexpressing lines do the opposite. Fraxetin reduces Fe³⁺ in vitro, and notably it does so better at neutral pH than at acidic pH — exactly where a plant needs the help. Across natural Arabidopsis accessions, the amount of fraxetin secreted correlates with growth, iron content and chlorophyll, making it a determinant of which plants can live on chalky soil at all ^{13Reference 13Tsai HH et al. · 2018Scopoletin 8-Hydroxylase-Mediated Fraxetin Production Is Crucial for Iron MobilizationView study →}. The same iron-deficiency response produces a family of related redox-active coumarins ^{14Reference 14Rajniak J et al. · 2018Biosynthesis of redox-active metabolites in response to iron deficiency in plantsView study →}.

Fraxin is the storage and transport form of that molecule: glucosylation at the 8-position makes it water-soluble, chemically quiet and safe to accumulate — the same logic that makes esculin the storage form of esculetin. The biosynthetic chain runs scopoletin → fraxetin → fraxin, which is why ash bark carries the whole series rather than any one of them.

That iron chemistry is not confined to plants, and it is worth noticing where it reappears. Fraxin’s most distinctive pharmacological result is that it restrains ferroptosis — iron-dependent cell death — through Nrf2/GPX4 in a glucocorticoid-induced osteoporosis model ^{9Reference 9Zheng X et al. · 2024AnimalDelay the progression of glucocorticoid-induced osteoporosis: Fraxin targets ferroptosis via the Nrf2/GPX4 pathway — animal modelView study →}. A molecule whose plant job is handling Fe³⁺ turning up in iron-dependent cell death is a connection to hold lightly. It has not been tested as a mechanism, and the two settings are not otherwise linked.

Prevalence in Nature

Fraxin is an Oleaceae molecule first, and a minor constituent everywhere it occurs. Its name and its principal source are the same: ash (Fraxinus), whose bark — Cortex Fraxini — is a herbal material in its own right and carries fraxin alongside fraxetin, esculin and esculetin ^{11,12Reference 11Li JM et al. · 2011AnimalProtective effects of Cortex Fraxini coumarines against oxonate-induced hyperuricemia and renal dysfunction in mice — mouse modelView study →Reference 12Wang Z et al. · 2018Simultaneous Determination of Aesculin, Aesculetin, Fraxetin, Fraxin and Polydatin in Beagle Dog Plasma by UPLC-ESI-MS/MS and Its Application in a Pharmacokinetic Study after Oral Administration Extracts of Ledum palustre L. — animal pharmacokinetic modelView study →}. Manna ash (F. ornus) is a named source.

It is also present in horse-chestnut bark and seed, where it was first identified, grouped with esculin as the seed’s coumarin fraction ^{15Reference 15Rochleder F · 1860Ueber das Vorkommen des Fraxin in der Rinde von Aesculus HippocastanumView study →}. Beyond those two genera it appears in marsh Labrador tea (Ledum palustre) — the plant used in the comparative dog pharmacokinetic study ^{12Reference 12Wang Z et al. · 2018Simultaneous Determination of Aesculin, Aesculetin, Fraxetin, Fraxin and Polydatin in Beagle Dog Plasma by UPLC-ESI-MS/MS and Its Application in a Pharmacokinetic Study after Oral Administration Extracts of Ledum palustre L. — animal pharmacokinetic modelView study →} — and, as fraxetin, in the root exudates of a wide range of plants under iron deficiency ^{13,14Reference 13Tsai HH et al. · 2018Scopoletin 8-Hydroxylase-Mediated Fraxetin Production Is Crucial for Iron MobilizationView study →Reference 14Rajniak J et al. · 2018Biosynthesis of redox-active metabolites in response to iron deficiency in plantsView study →}.

No quantified content figure was found for any source, so the table below records source and part rather than concentration.

Major Natural Sources

Source	Family	Part	Notes
Ash (Fraxinus spp., Cortex Fraxini)	Oleaceae	Bark	The principal source; with fraxetin, esculin and esculetin 11,12Reference 11Li JM et al. · 2011AnimalProtective effects of Cortex Fraxini coumarines against oxonate-induced hyperuricemia and renal dysfunction in mice — mouse modelView study →Reference 12Wang Z et al. · 2018Simultaneous Determination of Aesculin, Aesculetin, Fraxetin, Fraxin and Polydatin in Beagle Dog Plasma by UPLC-ESI-MS/MS and Its Application in a Pharmacokinetic Study after Oral Administration Extracts of Ledum palustre L. — animal pharmacokinetic modelView study →
Manna ash (Fraxinus ornus)	Oleaceae	Bark	Named source
Horse chestnut (Aesculus hippocastanum)	Sapindaceae	Bark, seed	Where fraxin was first identified 15Reference 15Rochleder F · 1860Ueber das Vorkommen des Fraxin in der Rinde von Aesculus HippocastanumView study →; grouped with esculin
Marsh Labrador tea (Ledum palustre)	Ericaceae	Aerial	The extract used in the comparative dog PK study 12Reference 12Wang Z et al. · 2018Simultaneous Determination of Aesculin, Aesculetin, Fraxetin, Fraxin and Polydatin in Beagle Dog Plasma by UPLC-ESI-MS/MS and Its Application in a Pharmacokinetic Study after Oral Administration Extracts of Ledum palustre L. — animal pharmacokinetic modelView study →

Discovery & Synthesis

Fraxin is named for the ash, Fraxinus, and like the other bark coumarins it was picked up in the mid-nineteenth century because it crystallised and fluoresced. The aglycone was described as “Fraxetin” in 1859, and in 1860 Friedrich Rochleder reported finding fraxin in the bark of Aesculus hippocastanum — establishing that the ash coumarin and the horse-chestnut coumarins were the same family of substances, which is why the two barks have been discussed together ever since ^{15Reference 15Rochleder F · 1860Ueber das Vorkommen des Fraxin in der Rinde von Aesculus HippocastanumView study →}.

The aglycone was synthesised in 1938 by Ernst Späth and E. Dobrovolny, who made fraxetin, fraxidin and iso-fraxidin in one paper — the forty-second in Späth’s series on natural coumarins, the same series that settled osthole and imperatorin ^{16Reference 16Späth E · 1938Synthese des Fraxetins, des Fraxidins und des Iso-fraxidins (XLIIView study →}. The glucoside itself is not usefully synthesised; extraction from bark is trivial by comparison.

Manufacture today is extraction, from ash or horse-chestnut bark, for research and analytical use. Fraxin is a reference standard in the quality control of Cortex Fraxini rather than a product in its own right, and there is no licensed drug or supplement containing it.

Patents: not yet researched.

Toxicity & Safety

Fraxin is the benign co-coumarin of horse chestnut, not the toxic one — it should not inherit esculin’s reputation, and it is not a blood thinner. What it lacks is any human data at all.

Safety at a Glance

✅ Low / no toxicity

• No isolated-toxicity signal. Nothing appears in the reviewed literature, and in the injury models it is uniformly protective rather than toxic.
• No safety study has been done. No repeat-dose, genotoxicity, interaction or human data — the absence of a signal is not a clean result.
• Not a warfarin-type anticoagulant. A 7,8-dihydroxy/6-methoxy coumarin via fraxetin, not a 4-hydroxycoumarin — the “coumarins thin the blood” claim does not apply.
• Avoid in pregnancy and lactation. No reproductive-safety data on the isolate.
• Do not inherit esculin’s framing. The “toxic component removed during standardization” story is esculin-specific and not fraxin’s.
• Not uricosuric on its own. The classic Cortex Fraxini anti-gout use rests on the whole coumarin fraction, not isolated fraxin.

Full Safety & Interactions Detail

Fraxin is a minor dietary and medicinal coumarin glucoside with low apparent toxicity — no isolated-toxicity signal appears in the reviewed literature ^{1Reference 1Ferdous J et al. · 2024ReviewPharmacological Activities of Plant-Derived Fraxin with Molecular Mechanisms: A Comprehensive Review — reviewView study →}, and across every injury model above it is protective rather than harmful. The three inheritance traps are handled above: it is not esculin, it is not uricosuric on its own, and it is not a warfarin-type anticoagulant, so any coumarin-bleeding caution here is theoretical.

What has not been assessed. There is no repeat-dose toxicology, genotoxicity package, drug-interaction study or human exposure data for isolated fraxin. Its low flag reflects the absence of any signal in studies that were not looking for one, which is weaker evidence than a clean toxicology package — and weaker again given that an oral dose is largely delivering fraxetin, whose own safety profile is a separate question ^{12Reference 12Wang Z et al. · 2018Simultaneous Determination of Aesculin, Aesculetin, Fraxetin, Fraxin and Polydatin in Beagle Dog Plasma by UPLC-ESI-MS/MS and Its Application in a Pharmacokinetic Study after Oral Administration Extracts of Ledum palustre L. — animal pharmacokinetic modelView study →}.

Dosage

There is no established human dose — fraxin has never been given to a person as an isolate, and it is not sold as a supplement. The figures below are the exposures used in the research described above.

Two things make these harder than usual to translate. The published reports give rodent milligrams per kilogram without a consistent oral/parenteral split, and the oral route delivers mostly fraxetin rather than fraxin ^{12Reference 12Wang Z et al. · 2018Simultaneous Determination of Aesculin, Aesculetin, Fraxetin, Fraxin and Polydatin in Beagle Dog Plasma by UPLC-ESI-MS/MS and Its Application in a Pharmacokinetic Study after Oral Administration Extracts of Ledum palustre L. — animal pharmacokinetic modelView study →} — so a “fraxin dose” and a fraxin exposure are not the same quantity. Herbal exposure is indirect and unquantified, through ash bark or horse-chestnut material. These are doses studied in research and are not a personal recommendation.

References

1. Ferdous J, Bhuia MS, Chowdhury R, et al. (2024). Pharmacological Activities of Plant-Derived Fraxin with Molecular Mechanisms: A Comprehensive Review — review. Chemistry & Biodiversity. https://pubmed.ncbi.nlm.nih.gov/38506600/
2. Li W, Li W, Zang L, et al. (2019). Fraxin ameliorates lipopolysaccharide-induced acute lung injury in mice by inhibiting the NF-κB and NLRP3 signalling pathways — mouse model. International Immunopharmacology. https://pubmed.ncbi.nlm.nih.gov/30530164/
3. Ma X, Liu X, Feng J, et al. (2019). Fraxin Alleviates LPS-Induced ARDS by Downregulating Inflammatory Responses and Oxidative Damages and Reducing Pulmonary Vascular Permeability — animal model. Inflammation. https://pubmed.ncbi.nlm.nih.gov/31273573/
4. Sun W, Hang L, Chen L, et al. (2024). Fraxin Ameliorates Ulcerative Colitis by Modulating Oxidative Stress, Inflammation and TLR4/NF-κB and MAPK Signaling Pathways — animal model. Alternative Therapies in Health and Medicine. https://pubmed.ncbi.nlm.nih.gov/38518135/
5. Yao H, Zhao J, Song X (2022). Protective effects of fraxin on cerebral ischemia-reperfusion injury by mediating neuroinflammation and oxidative stress through the PPAR-γ/NF-κB pathway — animal model. Brain Research Bulletin. https://pubmed.ncbi.nlm.nih.gov/35772607/
6. Niu X, Liu F, Li W, et al. (2017). Hepatoprotective effect of fraxin against carbon tetrachloride-induced hepatotoxicity in vitro and in vivo through regulating hepatic antioxidant, inflammation response and the MAPK-NF-κB signaling pathway — in vitro and animal model. Biomedicine & Pharmacotherapy. https://pubmed.ncbi.nlm.nih.gov/28922728/
7. Chang BY, Jung YS, Yoon CS, et al. (2017). Fraxin Prevents Chemically Induced Hepatotoxicity by Reducing Oxidative Stress — animal model. Molecules. https://pubmed.ncbi.nlm.nih.gov/28383514/
8. Whang WK, Park HS, Ham I, et al. (2005). Natural compounds, fraxin and chemicals structurally related to fraxin, protect cells from oxidative stress — in vitro. Experimental & Molecular Medicine. https://pubmed.ncbi.nlm.nih.gov/16264268/
9. Zheng X, Ye FC, Sun T, et al. (2024). Delay the progression of glucocorticoid-induced osteoporosis: Fraxin targets ferroptosis via the Nrf2/GPX4 pathway — animal model. Phytotherapy Research. https://pubmed.ncbi.nlm.nih.gov/39192711/
10. Qian Z, Ru X, Liu C, et al. (2021). Fraxin Prevents Knee Osteoarthritis through Inhibiting Chondrocyte Apoptosis in an Experimental Rat Osteoarthritis Model — rat in vivo animal model. Protein and Peptide Letters. https://pubmed.ncbi.nlm.nih.gov/34719360/
11. Li JM, Zhang X, Wang X, et al. (2011). Protective effects of Cortex Fraxini coumarines against oxonate-induced hyperuricemia and renal dysfunction in mice — mouse model. European Journal of Pharmacology. https://pubmed.ncbi.nlm.nih.gov/21620826/
12. Wang Z, Zhu W, Liu H, et al. (2018). Simultaneous Determination of Aesculin, Aesculetin, Fraxetin, Fraxin and Polydatin in Beagle Dog Plasma by UPLC-ESI-MS/MS and Its Application in a Pharmacokinetic Study after Oral Administration Extracts of Ledum palustre L. — animal pharmacokinetic model. Molecules. https://pubmed.ncbi.nlm.nih.gov/30205426/
13. Tsai HH, Rodríguez-Celma J, Lan P, et al. (2018). Scopoletin 8-Hydroxylase-Mediated Fraxetin Production Is Crucial for Iron Mobilization. Plant Physiology. https://pubmed.ncbi.nlm.nih.gov/29559590/
14. Rajniak J, Giehl RFH, Chang E, et al. (2018). Biosynthesis of redox-active metabolites in response to iron deficiency in plants. Nature Chemical Biology. https://pubmed.ncbi.nlm.nih.gov/29581584/
15. Rochleder F (1860). Ueber das Vorkommen des Fraxin in der Rinde von Aesculus Hippocastanum. Journal für Praktische Chemie. https://doi.org/10.1002/prac.18600800117
16. Späth E, Dobrovolny E (1938). Synthese des Fraxetins, des Fraxidins und des Iso-fraxidins (XLII. Mitteil. über natürliche Cumarine). Berichte der deutschen chemischen Gesellschaft. https://doi.org/10.1002/cber.19380710910

View all 16 references▾