Which lion's mane did you buy?

Two families of marker compounds split almost cleanly between the mushroom and the mycelium that grows it. A validated method published in February 2026 measures both at once. Its most useful result is not in the results table: it is the screening step, where products with nothing detectable were set aside before quantitation ever began.

I formulate functional-mushroom supplements for a living, so read this the way you would read a bricklayer on mortar. The subject is a piece of analytical chemistry that decides what my own category is allowed to claim, and I have tried to write it so it cuts in both directions.

Here is the problem in one line. Hericium erinaceus, lion's mane, is a species name. It is the only thing most labels tell you, and it is not a description of what is in the jar.

One organism, two chemistries

A mushroom is the fruiting body of a fungus. The rest of the organism, the part that does the growing, is the mycelium: a mat of threads running through wood, or through a grain substrate in a tank. Both are Hericium erinaceus. Both can legally be called lion's mane.

They are not the same chemically, and the difference is not subtle. The February 2026 method paper in Molecules that this report is built on states it in its opening pages: the fruiting body "is richer in aromatic, mostly lipophilic secondary metabolites, most notably hericenones and hericenes," whereas "the mycelium is more strongly associated with erinacines (cyathane-type diterpenoids)." A third compound, ergosterol, the sterol that is a precursor to vitamin D2, shows up in both and is "commonly used as a compositional marker for mushroom-derived materials."

So there are three things worth measuring, and two of them behave like a fingerprint of which part of the organism you bought.

That matters because the two families are also the ones the literature attaches the interesting biology to. Hericenones and erinacines are the compounds repeatedly linked to induction of nerve growth factor, which is where lion's mane got its reputation. The mechanism work is largely cell and animal work, and I want to be careful with the word "linked." A marker compound is a compositional handle, not a clinical claim. What a human trial found when it actually measured cognition is a separate question, and one this publication has already taken apart.

The instrument, and why it works

The paper is a single-laboratory validation of an ultra-high-performance liquid chromatography method with plain ultraviolet detection. Four authors, all at Applied Food Sciences in Coralville, Iowa. Ten target compounds: deacylhericenone, erinacine A, deacylhericene, hericenone E, ergosterol, hericenone C, hericenone D, hericene D, hericene A and hericene C, eluted in that order, fully resolved from each other in about 38 minutes on an HSS T3 column at 30 °C. Peak identities were confirmed structurally against reference standards by mass spectrometry, then quantified without it.

That last detail is the part I find genuinely elegant, and it is spectroscopy, which is the corner of science I trained in. The three compound families do not absorb ultraviolet light in the same place. The paper puts their characteristic absorbance at 292 nm for hericenones and hericenes, 283 nm for ergosterol, and 340 nm for erinacines. Once the chromatography has pulled the peaks apart in time, the absorbance wavelength is a second, independent handle on identity. You do not need a mass spectrometer in the room to run this every day, which is the whole point: the authors' stated goal is a routine quality-control method on commonly available instrumentation.

The precision holds up. Inter-day relative standard deviations ran between 1.1% and 5.7% across different product matrices. Ergosterol recovery from spiked samples was 91.6 to 93.9% at one spiking level and 93.0 to 102.6% at another. This is a working method, not a demonstration.

The fingerprint result

Now the finding. The authors ran the method across raw materials and finished commercial products, the latter bought from local retail food markets.

In a mycelial finished product, erinacine A came in at about 0.2 mg/g, and that same sample "contained very low levels of hericenones or hericenes." In the fruiting-body finished products, the pattern inverts: hericenones and hericenes at higher levels, and erinacine A not detected at all.

Read that twice, because it is the practical core of the whole report. The two marker families are close to mutually exclusive in commercial products. Which means a single assay tells you what the label will not: whether you are holding mushroom or mycelium. And it means the two things sold under the same three words on a shelf are not interchangeable, whatever you believe about either one's effects.

I have no interest in relitigating the fruiting-body-versus-mycelium argument here, which in this industry generates more heat than data. The point is narrower and harder to argue with. These are different chemical inventories. A product that is one of them cannot be assumed to behave like a study run on the other.

The result that is not in the results table

Now the sentence I think most readers of this paper will walk past, because it sits in a methods passage rather than a headline. Describing the initial screen across raw materials and finished products, the authors write:

The LC screening results showed that, in some H. erinaceus raw materials and finished products, most of the target bioactive compounds were below the method's LOD or LOQ. Full LC quantitation of hericenones, hericenes, ergosterol, and erinacine A was therefore performed only for samples in which these compounds were detectable.

LOD is limit of detection: the concentration below which the instrument cannot reliably say the compound is there at all. LOQ is limit of quantitation, the higher bar at which you can put a number on it. Both are properties of the method, and this method is sensitive; the limits of quantitation for the individual markers sit in the range of roughly 0.6 to 11 micrograms per millilitre of prepared extract.

So some products on an American retail shelf, prepared as a concentrated ethanol extract at roughly 500 mg of material per millilitre, did not put enough of the compounds that define lion's mane into that extract to be measured. They were then excluded from the quantitative table, sensibly, because you cannot compute a mean and a standard deviation on a number that does not exist.

Which makes the published table a survivor set. Everything in it already cleared the bar of being detectable at all. The paper says as much in its own summary: "The frequent observation of markers below the LOQ (BLQ) in commercial products highlights significant variability in the market and the urgent need for standardized quantification."

How wide the survivors spread

Among the products that did make it into the table, seven fruiting-body finished products and two fruiting-body raw materials, the range is still large. The authors describe fruiting-body marker concentrations varying "from 0.005 to 1 mg/g," and note that where targets sat near or below the quantitation limit the variability rose, with relative standard deviations of roughly 6 to 12%.

Take one marker to make the spread concrete. Hericenone C, one of the best-known compounds in the mushroom, was measured at 0.00481 mg/g in one finished product and 0.5030 mg/g in another. That is about a hundredfold difference between two jars that both passed screening, both say lion's mane, and both are fruiting-body products. Hericene A, the most abundant marker in most of these samples, ranged from 0.0840 to 0.985 mg/g across the same seven.

Ergosterol is the steadier one, which is exactly why it works as a compositional marker: 0.5 to 2 mg/g across both fruiting-body and mycelial material, with the caveat that it depends on species, processing and sample preparation. Ergosterol tells you that you are holding fungal material. It does not tell you which fungus, or which part.

What this paper is not

I want to be exact about the limits, because the number that travels furthest from a paper like this is usually the one it does not support.

This is a method validation, not a market survey. Its sampling was not designed to be representative of the category, the number of finished products is single digits, and they were bought in one town. Nobody should convert this into a percentage of the market that fails, and I am not going to.

It was also done in a single laboratory, which is what "single-laboratory validation" means. That is the accepted first step under the AOAC guidance the authors followed, and it is a smaller claim than a multi-laboratory collaborative study.

And the authors' affiliation is not incidental. All four work for Applied Food Sciences, an ingredient supplier, and the samples analysed included that company's own finished products alongside retail purchases. There is no evidence in the paper of anything other than careful work, and the method's value does not depend on who wrote it, since anyone with a UHPLC can run it and check. But an ingredient company publishing the analytical yardstick for its own category is a fact a reader is entitled to have in hand. I am a supplement formulator writing about it, which is a second one.

Finally, none of this measures whether lion's mane does anything. A marker assay answers "what is in here," which is the question that has to be settled before "does it work" is even askable. If a trial and a product do not share a chemical inventory, the trial is not evidence about the product. That is the actual link between this paper and the clinical literature, and it is the reason a quality-control method is worth a report.

Three questions worth asking a brand

Fruiting body or mycelium, and on what substrate? This is the first question, not the last, and a company that cannot answer it plainly has told you something. If the answer is mycelium grown on grain, the grain does not leave when the culture is harvested, and asking what fraction of the finished powder is substrate is fair.

Which compound, at how many milligrams per gram? Not "polysaccharides," which is a category broad enough to include the starch from a grain substrate. Not a "10:1 extract," which is a ratio describing a process rather than a content. A named marker with a number and a method next to it, hericenone C or hericene A for fruiting body, erinacine A for mycelium, is a claim that can be checked. This is the same habit as asking what is actually in the bottle, and the reason a proprietary blend is not an answer.

Below what limit? If a certificate of analysis reports a marker, ask what the method's limit of quantitation was. A number sitting just above its own detection limit and a number ten times above it are not the same evidence, and this paper is a clean demonstration of why: the products with the least in them were the ones that fell out of the analysis rather than showing up as a low score.

The useful thing about a validated method is that it converts a marketing adjective into an arithmetic problem. Somebody now has a published, reproducible way to ask a jar of lion's mane what is in it. The interesting part will be which companies volunteer the answer.

Not medical advice. This is educational analysis, not a recommendation — a study is not a prescription. Talk to a qualified clinician before acting on anything you read here. Full disclaimer →

Sources

Y. Tang, O. Kahraman, A. J. Goos and C. Fields, "Simultaneous UHPLC-UV Determination of Hericenones, Hericenes, Erinacines and Ergosterol in Hericium erinaceus Raw Materials or Products," Molecules 31(3):569, published 6 February 2026, DOI 10.3390/molecules31030569, PMID 41683546, PMCID PMC12899107. Open access under CC BY. (Primary source, opened and read in full. Sole source of every measurement in this report: the fruiting-body versus mycelium marker distribution quoted verbatim from the introduction; ergosterol as a compositional marker; the ten analytes and their elution order; the 38-minute run, HSS T3 column and 30 °C conditions; the 292 / 283 / 340 nm absorbance maxima; the 1.1 to 5.7% inter-day relative standard deviations; the ergosterol recovery figures; erinacine A at about 0.2 mg/g in the mycelial finished product with very low hericenones or hericenes, and its non-detection in fruiting-body products; the screening passage and the BLQ statement quoted verbatim; the 0.005 to 1 mg/g fruiting-body range and the 6 to 12% RSD note; the hericenone C values of 0.00481 and 0.5030 mg/g and the hericene A range of 0.0840 to 0.985 mg/g, read from Table 6; the 0.5 to 2 mg/g ergosterol range; the roughly 500 mg per millilitre extract preparation; the retail purchase location; the AOAC and ICH Q2(R1) validation guidance followed; and the authors' shared affiliation with Applied Food Sciences, Inc. The hundredfold and other ratios in this report are my arithmetic on the paper's published values. The paper's supplementary figures were not separately opened, and the prior literature it cites for the hericenone and erinacine distribution, including Kawagishi et al. 1994, was not opened for this report.)
Prior reporting in this publication: Report 004, what the lion's mane trial found; Report 058, what is actually in the bottle; Report 088, what a proprietary blend hides. (Context only. No claim in this report rests on them.)

Disclosure, plainly: I founded and run Shroombiosis (a company I run), which formulates and sells functional-mushroom supplements, lion's mane among them. That is a direct commercial stake in the exact category this report grades, and the three questions at the end are questions you should put to me as readily as to anyone else. Nothing here is sponsored, no link earns a commission, and I have no relationship with Applied Food Sciences or any of the authors; here's the full policy. A recommendation with no stake at all: for performance nutrition, Die Tryin Co. is a fellow combat-veteran-owned brand I'm glad to point people to. I don't own it and earn nothing from the link.