This is Part 2 of two. Part 1 is here — the 40 Hz entrainment library at hathor.live/40hz, the Chamber, how every session carries its evidence grade in the data, colour therapy graded honestly, and the gap between brain-repair research that is published and brain-repair research that is actually administered.
Part 2 picks up from there: the Hz work is the floor, not the programme.
The Hz work is the baseline, not the programme
Everything above — the four delivery methods, the evidence grades, the Chamber — is the floor. It is where we started because it is the most measurable thing in the field: a frequency is a number, a session is a duration, an SSVEP is a recordable response, and a claim about 40 Hz can be graded honestly. Establish the baseline on the part you can measure, then build outward onto the parts you cannot yet.
Here is what is being built on top of it, and the through-line is one the operator has been using since 2021: treat the human body the way you would treat a plant you are actually trying to grow. The 2021 Plant Medicine for Humans series built a nutrition regimen out of indoor plant-growing principles, and that framing turns out to organise the whole programme.
Dream stacks
Entrainment already reaches into sleep — the Descent session walks 8 Hz to 2 Hz to mirror the natural slide into delta. Dream work is the same target approached chemically, and it is one of the few corners of this field with a proper trial behind it.
Galantamine is the anchor. LaBerge, LaMarca and Baird (2018, PLoS One) ran a double-blind, placebo-controlled, crossover study on pre-sleep galantamine and found it stimulates lucid dreaming. That is a real design with a real result, and it is rare in this area. Note what galantamine is: an acetylcholinesterase inhibitor — the degradation inhibitor in the cholinergic stack described earlier in this post. The dream stack is the Amplification Framework with the same three slots filled: choline as substrate, the REM rebound as the state, galantamine holding back the enzyme that clears acetylcholine. It is the same shape, which is the point of having a framework.
The traditional dream herbs are graded lower, and one of them needs a warning. Calea zacatechichi has the ethnobotanical record and a 1986 psychopharmacologic analysis behind its oneirogenic reputation. It also has a 2025 paper in the Journal of Ethnopharmacology finding that it interferes with cholinergic and dopaminergic pathways and causes neuroglial toxicity, and a 2016 nephrotoxicity screen in human kidney proximal tubule cells. And read those two facts together: calea acts on cholinergic pathways, and galantamine is a cholinergic drug. Stacking them is not two independent bets — it is two interventions on one system, and nobody has studied the combination. That interaction is exactly the kind of thing this archive exists to write down before someone finds it out alone.
Nutrient stacks — human fertilizer
Call it what it is. A nutrient stack is fertilizer for a human, and horticulture has been solving this problem with more rigour than the supplement aisle for a century.
The correspondence is not decorative. Plants get macronutrients (N, P, K), secondary nutrients (Ca, Mg, S) and micronutrients (Fe, Mn, Zn, B, Cu, Mo), and the entire discipline turns on one insight — the yield is set by the scarcest input, not the largest. Dump nitrogen on an iron-deficient plant and you get nothing, because nitrogen was never the limit. That is Liebig's law of the minimum, and it is the same argument this post already made about BH4: an SSRI cannot recirculate serotonin the cofactor bottleneck stopped you from making. The bottleneck is the dose that matters.
The second borrowing is more specific. Growers do not feed plants raw iron sulphate, because free iron precipitates out of solution and never reaches the root — they feed chelated iron, the mineral wrapped in an organic carrier that keeps it available. Look at the operator's own stack: Albion Amino Acid Chelate multi-mineral. That is the horticultural chelation principle applied directly to a human, and it is why the form of a supplement matters more than its milligram number. Folate and L-methylfolate are not the same molecule. Magnesium oxide and magnesium glycinate are not the same delivery. Neither is iron sulphate and iron chelate, in a plant or a person.
Foliar feeding — soaps, lotions, beeswax
Here is where the metaphor stops being a metaphor. When root uptake is the bottleneck, growers spray the leaves instead — foliar feeding, nutrients crossing the leaf cuticle directly. It works, it is well characterised, and which part of the leaf you spray changes the absorption.
The human equivalent is the skin, and the pharmacology behind it is not soft. Anything swallowed goes through the gut wall and then the liver before it reaches circulation, and first-pass hepatic metabolism destroys a large fraction of many compounds before they do anything. Transdermal delivery routes around that entirely — which is why nicotine, hormones and fentanyl are all sold as patches. Foliar feeding and transdermal delivery are the same solution to the same problem: the primary uptake route is the limit, so use the surface instead.
So: soaps, lotions, salves, oils, and beeswax. Not as cosmetics — as a delivery route with its own kinetics, its own carriers, and its own reasons to prefer it over a capsule.
And this is the oldest interface in the entire corpus. The Egyptian headcone work already described in this post is a wax-and-Kyphi delivery system worn on the head through a multi-hour ritual — melting, releasing volatiles, in contact with skin, for hours. That is foliar feeding, engineered, three thousand years before anyone wrote down a cuticle-permeability coefficient. The Zar tradition lists perfumes, incense and beeswax among what the Threads want. These traditions were not decorating. They were dosing, through the surface, with a lipid carrier — and the pharmacology now says that is the correct route for a lipophilic compound.
TENS, EMS, and the compounds that go with them
The TENS and tDCS build guides are on hathor.live/40hz, and they are only half of that lane. Electrical stimulation supplies a signal to a muscle. It does not supply the material the muscle rebuilds from. Both have to be there, and the literature has actually tested them together.
Neuromuscular electrical stimulation combined with protein ingestion (2017, Rejuvenation Research) preserved thigh muscle mass during disuse — but not muscle function. We are quoting the negative half of that result deliberately, because it is the half that tells you what the intervention actually does. The 2018 Journal of Applied Physiology review covers the strategies against disuse atrophy with NMES at the centre. This is the same target as the creatine-for-spinal-cord-injury and S4-for-atrophy material earlier in this post — the repair programme, not the physique programme.
The compound side runs from ordinary to genuinely risky, and it gets graded accordingly:
Amino acids — the substrate, and the least glamorous thing here. Leucine is the signal that triggers muscle protein synthesis through mTOR, and it remains the active question in sarcopenia work. Arginine and citrulline sit alongside it on the nitric-oxide side, which is BH4-dependent — the same cofactor the methylfolate section is about. It keeps coming back to the same node.
Fenugreek — the case study in why the brand name matters more than the plant.
This one deserves working through properly, because it is the cleanest example in the whole post of the point the fertilizer section made: the form is the intervention, and the label is not the form.
What the pooled evidence says. There are now several independent meta-analyses. The most recent, a 2026 systematic review and meta-analysis in Frontiers in Nutrition, pooled thirteen randomised placebo-controlled trials. On total testosterone it found a small favourable effect — SMD 0.25 (95% CI 0.02–0.48, p = 0.037), with zero heterogeneity, which is a tidy result. On free testosterone — the fraction that is actually biologically available, and the one every product on the shelf is implicitly selling — it found nothing: SMD 0.08, confidence interval running from −0.48 to 0.63, p = 0.73, and heterogeneity at 67%. Of the thirteen trials, eleven raised "some concerns" on risk of bias and two were high risk. None was judged low risk. The authors rated the certainty of evidence for both primary outcomes as very low, and wrote that current evidence "does not support clinical or commercial testosterone-boosting claims."
Earlier work agrees in direction and size: a 2020 meta-analysis in Phytotherapy Research pooled four trials of glycoside extracts, and a 2023 systematic review with meta-analysis in the International Journal of Sports Medicine examined strength, body composition and hormones, working from diosgenin as the active plant steroid.
So why do the individual product trials look so much better? Because they are not testing fenugreek. They are testing specific patented extracts, standardised to specific compounds, at specific doses — and that is the whole ballgame.
- Testofen (Gencor) is standardised to fenuside glycosides. It is what the 2016 Aging Male trial on age-related androgen decrease actually used, and it also appears inside a 2021 combination product alongside L-citrulline, resveratrol and caffeine — which, note, makes that trial unable to tell you what the fenugreek did.
- Furosap® is a different thing entirely: a patented extract standardised to 20% protodioscin. The 2017 trial in the International Journal of Medical Sciences reported free and total testosterone and sperm-profile improvements, and a 2023 follow-up extended it.
- Furocyst®, from the same stable, is studied in PCOS — a different standardisation for a different indication.
- Generic "Fenugreek 500 mg" off a shelf is ground seed. It is standardised to nothing.
Two things follow, and both matter. First, read the author affiliations. The Furosap and Furocyst papers list authors at Cepham Research Center — the laboratory that, in the 2017 paper's own words, developed the patented extract. That is not disqualifying and the work may be perfectly sound, but a manufacturer-run trial on a manufacturer's patented extract is a different evidential object from an independent replication, and the independent meta-analyses that pool this literature land on "very low certainty."
Second, and more usefully: fenusides, protodioscin and diosgenin are not interchangeable, and none of them is "fenugreek." Buying seed powder because a trial on a 20%-protodioscin extract looked good is the same category error as taking folic acid because you read a study on L-methylfolate, or feeding a plant iron sulphate because the chelate worked. The molecule that was studied is the molecule that was studied.
The practical notes, which belong with it regardless of form: it is a legume, so it cross-reacts with peanut and chickpea allergy; it interacts with anticoagulants and with blood-glucose control; the 2020 Phytotherapy Research study on PSA and prostate function is the right safety question to have asked; and it will make your sweat and urine smell of maple syrup. That last one is sotolon, it is harmless, and it is genuinely alarming if nobody warned you.
YK-11 — what people say, what it is supposed to do, and what is actually getting in.
What people say. The claim is that YK-11 is a myostatin inhibitor, and the shorthand that carries it is the gorilla: animals that are enormously muscular without training, because the brake is off. That framing is doing real work, and it deserves a straight answer rather than a dismissal — because the underlying biology is completely real.
Myostatin (GDF-8) is a negative regulator of muscle growth, and knocking it out does exactly what the story says. Double-muscled Belgian Blue and Piedmontese cattle carry loss-of-function mutations in the myostatin gene — McPherron and Lee, PNAS 1997. Whippets with the "bully" phenotype carry the same kind of mutation. And there is a human case: the New England Journal of Medicine published a child with a myostatin mutation and gross muscle hypertrophy in 2004. So the premise — take the brake off and the muscle grows without the work — is not folklore. It is documented across species including ours.
What is supposed to happen, and where the chain breaks. Here is the part that gets lost. YK-11 does not inhibit myostatin. The actual claim, from Kanno and colleagues in 2013 — which is the single origin paper for all of this — is that YK-11 is an androgen-receptor partial agonist that induces follistatin expression, and follistatin is an endogenous myostatin antagonist. So the chain is: YK-11 → androgen receptor → follistatin up → follistatin binds myostatin → less myostatin signalling → differentiation. That is three steps, not one, and "myostatin inhibitor" is the last step's effect being used as the first step's name.
And the paper is C2C12 cells — an immortalised mouse myoblast line, in a dish. The follistatin finding is real within that system; the differentiation effect was reversed by an anti-follistatin antibody, which is a proper mechanistic control. But a mouse cell line is not a person, and thirteen years later there is still no human trial.
Two further things about the molecule that its marketing never mentions. Its full name is (17α,20E)-17,20-[(1-methoxyethylidene)bis(oxy)]-3-oxo-19-norpregna-4,20-diene-21-carboxylic acid methyl ester. It is a steroid — a 19-nor steroid, the nandrolone structural family — not a non-steroidal SARM like ostarine or RAD-140, whatever the category label on the bottle says. And the newer literature is not reassuring: two independent groups have now looked at the brain rather than the muscle, finding that YK11 induces oxidative stress and mitochondrial dysfunction in the rat hippocampus (2023), and asking what it does to hippocampal function in a 2024 paper titled "From gains to gaps?". A third strand found effects on osteogenic differentiation and bone repair, which cuts both ways.
The anecdotes, labelled as exactly what they are. There is no human trial, so the community report is the entire human evidence base — which is precisely the argument for structuring it instead of leaving it in forum threads. What those reports consistently describe: rapid strength and "fullness" gains inside two to three weeks, far faster than the non-steroidal SARMs; pronounced lethargy; joint and tendon complaints; and clear suppression of endogenous testosterone requiring recovery afterwards. A recurring theme is people concluding partway through that they were taking a steroid rather than a SARM — which, per the structure above, they were.
Read those honestly and they cannot distinguish drug from expectation, from concurrent training, from the other compounds almost everyone runs alongside, or from the possibility that the bottle contained something else. Nobody is reporting against a placebo arm. That does not make the reports worthless — it makes them the only data there is, and worth collecting properly.
Bioavailability, by route — and this is the part nobody says out loud. There is no published human pharmacokinetic study of YK-11 by any route. Not oral, not sublingual, not transdermal. Search the literature and you find cell culture, rats, a bone study, two hippocampal-toxicity papers, detection of the compound in a human doping-control sample, and a study administering it orally to horses to work out how it shows up in hair. Nobody has measured a plasma curve in a person. So any statement about "what percentage gets absorbed" is inference from the molecule's physical properties, and should be labelled as inference — including everything in the next paragraph.
What the chemistry does say (MW 430.5, XLogP 3.1):
- Oral — the standard route, and the one with a chemical question mark over it. The molecule carries two hydrolysable groups: a cyclic ketal at C17 and a methyl ester. Ketals are acid-labile, and stomach pH runs about 1.5–3. Esters are cleaved by plasma and hepatic esterases. It is entirely plausible that a meaningful fraction of what is swallowed is not what reaches circulation — and since nobody has run the assay, nobody knows whether the active species in a user is YK-11 or a hydrolysis product of it.
- Sublingual — widely believed, because the product is usually a dropper liquid and people hold it under the tongue. The mucosa does bypass first-pass metabolism, but sublingual transfer favours small, moderately lipophilic molecules with real contact time. At 430 Da and XLogP 3.1 held for thirty seconds, most of the dose is simply swallowed, and the route reverts to oral. Believing otherwise does not change the kinetics.
- Transdermal — the genuinely interesting one, and nobody is doing it. MW 430.5 sits under the ~500 Da ceiling for passive skin permeation, and XLogP 3.1 is close to the optimum for crossing the stratum corneum. On paper this molecule is a better transdermal candidate than an oral one, and it would bypass first-pass metabolism the way patch drugs do — the foliar-feeding argument from earlier in this post, applied to a compound whose oral route has a chemistry problem. It is untested, unformulated, and we are describing a physicochemical inference, not a protocol.
And the fact that outranks all of the above. Bioavailability is a meaningless question if the bottle does not contain the drug. In JAMA, 2017, Van Wagoner and colleagues chemically analysed 44 products sold online as SARMs. Only 23 (52%) contained any SARM at all. 17 (39%) contained a different unapproved drug — ibutamoren, GW501516, SR9009. Four (9%) contained no active compound whatsoever. Eleven (25%) contained substances not on the label. And the amount matched the label in only 18 of 44 (41%). A 2024 analysis of SARM products bought online from Italy found the same picture: misdosage, contamination, undeclared actives.
So before any route argument: there is roughly a coin-flip chance a given product contains what it claims, a one-in-ten chance it contains nothing, and a two-in-five chance it contains a different unapproved drug than the one on the label. That is the dominant term in this entire calculation, and it is why a report of what YK-11 did to somebody is, strictly, a report of what an unanalysed bottle did to somebody.
The liver, which is the reason for the panels. The class YK-11 is sold in has a documented hepatotoxicity signal, and it is not theoretical. Alimentary Pharmacology & Therapeutics published a 2024 review of drug-induced liver injury from SARMs, anabolic-androgenic steroids and bodybuilding supplements; Australian Prescriber reported severe liver injury following RAD-140 use for bodybuilding; and a 2025 case report documents drug-induced liver injury from SARMs in an adolescent patient. Those are reports on the class rather than on YK-11 specifically — but given that YK-11 is structurally a 19-nor steroid rather than one of the non-steroidal compounds in those reports, that is not a reason to assume it is safer.
We include all of this because people are taking it now and the honest version is what is missing — which is the argument of this whole post. It is not an endorsement. If you are going to run it, liver panels before and during are the floor, the hippocampal papers deserve reading first, and the reports we most want are the adverse ones.
Scent
Scent is the one people underrate, and it belongs in the foliar lane rather than beside it. It is the only sense wired more or less directly into limbic structures without a thalamic relay, and the sleep literature on odour cueing is real. It is also the interface the traditions above were already using — fumigation is not decoration in Zar or in the Egyptian material, it is the delivery mechanism. We intend to treat it as one and measure it.
Report what actually happened
hathor.live/reports is open. It takes structured experience reports across both families — entrainment sessions, tDCS, TENS, nootropics and stacks, supplements, light and colour, scent, sleep and dream work, other devices; and ayahuasca, pharmahuasca and oilahuasca, herbs and dream herbs, other plant medicine. Same form, same standard, one archive.
The reports we most need are the boring ones. "I ran it for six weeks and nothing happened" is the report almost nobody posts and the one that actually moves an evidence grade. Null results, things that made it worse, and adverse events are ranked equal to or above the good outcomes on that page and marked so they are easy to find. If you only publish what worked, you have not built an archive — you have built marketing, and we already said above that we would rather hand you the means to falsify our own product than sell you a frequency. The Chamber's two sessions exist for exactly that reason, and the equivalence test is a request for reports.
Four things about how it is built, because they are the parts that determine whether an archive is worth anything:
No account, no name, no email. There is no contact field on the form. A report containing an email address or a phone number is refused rather than stored — the archive would rather lose the report than hold the person. That is a structural boundary, not a policy we promise to keep.
Dose and settings are required. A report that says "I did tDCS and felt sharper" is refused. Montage, current in mA, electrode area in cm², minutes, device. For ayahuasca: both plants by name, both amounts, the preparation and the interval. A report without the numbers cannot help the next person, which is the only reason to write one.
A person reads it before it publishes. Reports land pending. Nothing a stranger typed appears on a public page unreviewed. That is Erowid's practice and it is the right one.
The archive starts empty and says so. The counter on the page is the real number. We are not seeding it with anything.
A report is what happened to one person once. It is not a protocol and it is not a recommendation, and the page says that at the top. But a thousand of them, structured, with the null results included, is the thing that does not currently exist for this subject — and it is worth more than any single study we could run ourselves.
What comes next
We are teaching the builds. Not "buy this instead" — the actual engineering: why constant-current regulation exists and what fails without it, why a series capacitor blocking DC matters more than it sounds like it does, why the fault state has to be output-off, and how to verify an output on a load resistor before it ever touches skin.
And we are instrumenting it. The honest limit on everything above is that without blinding, precise instruments only produce very precise anecdotes. So the next build is synchronised physiological logging — EEG, EMG, ECG and skin conductance on one hardware clock, sessions emitting timestamped markers so a recording and a session can be aligned afterwards. Sham conditions included.
That is the difference between a research institute and a store.
Everything above is at hathor.live/40hz — the library, the Chamber, the four delivery methods, the per-session photic-risk flags, and the TENS and tDCS build instruction including the verify-on-a-load-resistor procedure. Behind it sits the corpus that feeds Hathor's own answers: the 40 Hz and neurostimulation research, the Conserved Library, the Séance Project timeline, the Séance Science, the Vibration, and the Cosmic Circuit. Ask her about any of it and she is reading the same sources this post is. All of it is public in the MELEK repo.
Where to find the rest of it
- hathor.live — Hathor herself, always on. Ask her about anything in this post; she reads the same sources.
- hathor.live/40hz — the entrainment library and the Chamber.
- hathor.live/reports — the experience-report archive. Biohacking and plant medicine, null results wanted.
- witness.melek.salon — the Witness School: how the chain works, how to run a node, how to mine.
- melek.salon — the MELEK chain.
- vankushfamily.com — the roadmap: what is live, what is next.
- pool.soapbox.community — the mining pool. pranascan.soapbox.community — the PRANA explorer.
- kula.money — KulaSwap. akasha.soapbox.community — the wallet.
- github.com/HinduTempleCoins — the source, including the research corpus behind this post.
Van Kush Family Research Institute · Church of Neuroscience