TENFLY Solfeggio Tuning Fork Set: An Independent Review

Daniel Mercer publishes books. This site sells no hardware, and it is affiliated with no tuning fork manufacturer — not TENFLY, and not any of the makers named further down for comparison. No manufacturer has paid for, seen, or influenced anything on this page, and nothing above a link changes because of it. Retailer links are marked where they appear. No unit was bought and none was tested here: every figure about this object comes from the seller's own listing, from buyers who published readings, or from the physics literature. Nothing here is a recommendation to buy, and nothing here is a claim about health.

Two things put this listing on the site. The extraordinary claim is not buried in a bullet — it is in the product title, on the set that led its search by rating count when that search was read on 2026-08-29. And one buyer did something almost nobody does: they measured all nine forks and published the numbers. So for once a seller's accuracy claim can be set against published readings rather than against opinion alone. Both halves of what follows are worth a reader's time, and they point in opposite directions. Several independent buyers who handled this set thought the metalwork, the mallets and the presentation were good. The claim printed above them, about 528 hertz and DNA, traces back through a single cell-culture paper to a 1999 book, and the loop closes there.

Volatile figures carry their date. The listing (ASIN B094Y23MGC, brand TENFLY, first available 19 May 2021) showed 4.6 out of 5 from 248 ratings when it was read on 2026-08-29, with a rating distribution that day of 81% five-star, 10% four-star, 4% three-star, 2% two-star and 3% one-star, and a Best Sellers Rank of #35,390 in Musical Instruments and #381 in Music Tuning Accessories. Treat every figure, quotation and engraving reported from this listing as of 2026-08-29 only. Listings, images and bullets are edited without notice, and none of it can be re-checked once the seller changes it. No price appears anywhere here: the browsing session that produced this research was localized to a non-US market, and converting or reprinting that figure would give a US reader a wrong number.

What Is Actually in the Box, and What the Engravings Say

Nine forks covering 174, 285, 396, 417, 528, 639, 741, 852 and 963 hertz, in aluminum alloy, offered in silver or gold. Total item weight for the whole package is 1.5 pounds. Buyers describe two mallets, a pouch, a cleaning cloth, and each fork individually sleeved. The listing recorded on 2026-08-29 does not say whether these forks are weighted or unweighted, which is one of the eight items chapter 3 of The Tuning Fork Healing Handbook asks a listing to state. A total of 1.5 pounds for nine forks plus accessories is consistent with unweighted forks, but that is an inference drawn here from a shipping figure, not a specification the seller published.

The praise is unusually consistent in what it is about, and what it is about is the object. The most-endorsed review on the listing (Mø, five stars, US, 28 March 2023, marked helpful by 18 people as of 2026-08-29) describes years in holistic practice and then spends its length on build, tone, the second mallet, the extra pouch, the cleaning cloth and the individual sleeves. Others say much the same: a beginner (TIMOTHY W OSHEI, five stars, 19 January 2025) calls the set easy to use and light; another (BreScha, five stars, 13 August 2026) calls it good to learn with; short positives from Australia, Germany and the US single out finish and presentation. One five-star review contains a breakage report — Piscearicorn (five stars, US, 22 May 2026) writes that the blue mallet had a compromised piece of wood and snapped in half within the first five minutes of use. A five-star rating carrying a failure is worth noticing, because it shows at least one hardware failure that the low-star counts do not capture. One instance is not a rate, and nothing here estimates one.

Two reports are testimony and are handled as testimony. One buyer (Shell Dawkins, five stars, 20 November 2025) makes a health claim for the set, in both physical and spiritual terms; the wording is not repeated here. Another (Mø, above) writes that even their cat is obsessed with them — which is the listing's own "playing with pets" bullet reappearing in the wild as a customer's words. We record that these buyers said it. Neither is restated here as something the forks do. A customer review is testimony, not evidence: it records what one person believed happened, establishes nothing about cause, and cannot support a health claim for this or any other object. The handbook's name for one person's before-and-after is personal experiment, and its rule is that your own notes guide your practice but do not establish a universal effect. On the pets question specifically: no peer-reviewed study of tuning forks used with companion animals could be located for this page. Everything a search returns is a retailer, a practitioner blog or a course provider. There is a real literature on music and auditory enrichment for shelter animals, but that is recorded music played to animals, not a struck fork, and importing it here would be exactly the category error this page criticizes elsewhere.

Now the engravings, because they answer a complaint and raise a new one. A four-star buyer (Amazon Customer, US, 14 November 2024, marked helpful by 10 people as of 2026-08-29) wrote that two forks were marked incorrectly: that 639 should be FA and 741 should be SOL. The seller's own listing image, read at full resolution on 2026-08-29, engraves exactly that — FA on the 639 fork and SOL on the 741. This page cannot know which unit that buyer received. The images may post-date a November 2024 purchase, or the forks in that particular box may have differed. Their honesty is not in question here, and the only useful thing to tell a reader is the checkable one: this specific complaint does not reproduce against the listing as it stood on 2026-08-29.

The oddity no reviewer in the recorded material flagged is a different one. Three forks — 174, 285 and 963 — all carry the same engraved letter, F. It is not a hexachord syllable, and it cannot distinguish three forks from one another. Those three are precisely the tones outside the classical six, the later additions to the set, which is also how both of this publisher's books treat them. F is a real note letter for exactly one of the three: computed here at A = 440, F3 is 174.614 hertz, so the 174 fork sits about 6 cents below a genuine F3 and is fairly called an F. The other two are nowhere near — 285 is about 48 cents above C♯4 and 963 about 44 cents below B5. A correct marking on one fork copied onto the other two is a plausible reading of that pattern, but it is inference from an engraving, not anything the seller states.

One sentence on where this product is filed, and then the page moves on. A set whose own bullets frame it in medical terms is filed by that same seller under Musical Instruments, at #35,390 in that category and #381 in Music Tuning Accessories as read on 2026-08-29 — not under Health & Household. From the object's point of view that filing is the accurate one. This is a set of pitched acoustic resonators, and everything technically checkable about it — frequency, sustain, material, build — is a musical-instrument property.

One Buyer Measured All Nine Forks

Writing from the United Kingdom on 2 August 2022 at three stars, a buyer named Kas reported that some forks looked damaged on the tips and that the tone was not as advertised, and then did the thing that makes this page possible: they published a reading for every one of the nine forks. The arithmetic in the table below was recomputed independently here from those readings, as (measured − marked) ÷ marked and as 1200 × log₂(measured ÷ marked), and it reproduces to the last digit.

Three limits belong to those numbers before any conclusion is drawn from them, and they belong to every later mention of them on this page. This is one buyer, one set, one session. The readings are whole numbers, so each carries at least ±0.5 hertz of rounding. And the buyer never said what they measured with — which is strictly weaker than the comparable case on this site's ENSO review, where the reviewer named the app and its documented limits could be discussed. A second buyer offers a contrary verdict on the same question: writing from Mexico on 3 September 2025 at five stars, Ramon says some were very exact on the hertz and others varied a very little, and calls that tolerable. Both verdicts are on the record and both belong here.

What the numbers do support is close to what Kas said, but not identical to it. The reviewer wrote that the higher tones especially are off, and their own figures broadly bear that out: the three highest forks average 1.86% absolute deviation against 1.02% for the three lowest. Across all nine the mean absolute deviation is 1.23% and the median is 1.14%. But the deviation is not monotonic in frequency and must not be described as though it were. The 285 hertz fork read exactly on the mark. The 417 hertz fork read sharp, not flat. And the 174 hertz fork was the second-worst in the set at −2.30%. Seven read low, one was exact, one read high. That is a tendency with exceptions, not a law.

The headline consequence is unavoidable and is stated without gloating. The fork this entire product is sold on — 528 hertz, named in the title and in the DNA bullet — measured 522 hertz on this buyer's unit, about 20 cents flat, which is roughly a fifth of a semitone.

Against what should that be judged? Not against a figure invented here. Chapter 3 of the handbook is explicit: compare repeated readings with the manufacturer's stated tolerance, and do not invent your own universal tolerance when none was specified. TENFLY publishes no tolerance anywhere in the listing material recorded on 2026-08-29, and with no tolerance stated there is no reference temperature for one either, so the only available comparison is against what other makers publish. Riverbank Laboratories tunes to within 0.5% at 20 °C and Omnivos Therapeutics states 0.5% of indicated frequency at 20 °C, both established for this site's earlier review and read on 2026-08-29. American Diagnostic Corporation publishes two figures that do not agree with each other: 1% of the rated frequency on the product page for one of its tuning forks, and 5% of the rated frequency in its own instruments FAQ, both verified at the source on 2026-08-30. Both are printed here, because printing only the stricter one would be printing only the figure that makes this seller look worse. That gap inside one company's published material is also a fair measure of how loose this category is, and it matters here, because at 5% every fork in this set passes. Neither ADC figure is an industry standard; it is one manufacturer's own copy, published inconsistently on its own site. There appears to be no ISO or ANSI standard for tuning-fork frequency tolerance, which is precisely why the manufacturer's own figure is the thing to look for. All of these are other companies' numbers, not a standard this seller agreed to.

Carried through both edges of the ±0.5 hertz rounding band, rounding changes no verdict. Against 0.5%, only the 285 fork is inside; the other eight sit outside even at the most favorable edge of their band, the closest being 639 at 0.55%, 417 at 0.60% and 396 at 0.63%. Against 1%, four are inside — 285, 396, 417 and 639 — and five are outside at their favorable edge too: 174 at 2.01%, 528 at 1.04%, 741 at 1.15%, 963 at 1.61% and 852 at 2.64%. The 528 fork's band runs from −1.23% to −1.04%, which is outside 1% at every point in it. Against ADC's other published figure, 5%, all nine are inside, with the largest deviation in the set at 2.70%.

Now the part that makes this more than a complaint, because the standard explanations for a fork reading off do not work here, and one of them can be dismissed without any arithmetic at all. Temperature is the first. A fork's frequency falls as it warms, because Young's modulus falls, and a single ambient temperature acts on every fork in a set in the same direction. But the 417 fork read sharp and the 285 fork read exactly on the mark. No single room temperature produces that pattern alongside seven flat readings, whatever the coefficient turns out to be. Temperature cannot be the explanation, though it could be a small component sitting on top of one. That argument needs no number, which is fortunate: the published temperature coefficients for fork metals are thin and disagree with each other, which is itself part of why makers stamp a reference temperature on a spec sheet.

Striking force is the second, and here the honest answer is that direction is contested while scale is not. This site's earlier review established the relevant measurement: a 2020 Nanjing University group using double-grating Doppler interferometry found softening rather than hardening, pulling a roughly 511 hertz fork down by about 0.15% at twenty-five times their reference drive amplitude. That was a continuously driven resonance sweep, not a struck fork, and the authors note two things that matter here — their softening result is the opposite of what a cantilever model predicts, and the coils and gratings coupled to their fork could themselves alter the sign of the effect. Rossing, Russell and Brown had earlier reasoned from a cantilever model that some modes of a fork may sit higher than normal immediately after a hard blow while others sit lower. So nothing in the literature lets this page rule striking force out on direction. Scale rules it out instead. The largest deviation reported here is about eighteen times the Nanjing shift and the 528 fork's is about eight times it, so explaining these readings by how hard the forks were struck would require an effect an order of magnitude larger than the largest one anyone has measured. A hard blow does change which modes get excited, which is what confuses an analyzer app in the first instant, but the fundamental itself barely moves.

The damaged tips are the third candidate, and they are the fairest reading of this buyer's own evidence to get right. Daniel Russell warns that a fork should never be struck against a hard tabletop or hit with a metal object, because a dent can change its frequency, and Kas reported damage on tips in the same review as the readings. But the two observations do not join up in the direction reported. Removing metal from the tine tips raises pitch; it does not lower it. And a dent that deforms without removing mass has no clear reason to pull a fork one to three percent flat. Seven forks reading low across one box looks much more like tuning at manufacture than like damage in transit. That matters because it keeps the two complaints separate: conflating them would quietly let the seller off the second one.

What is left is the ordinary explanation, and it is worth spelling out because it makes the whole thing legible. Rossing, Russell and Brown's 1992 paper gives the governing equation, and the material properties enter it only through one square-root term; the geometry does everything else. Frequency goes as the inverse square of tine length, so a fractional frequency error corresponds to half that fractional length error in the opposite direction. The 852 fork reading 2.70% flat corresponds to tines about 1.35% too long; the 528 fork reading 1.14% flat, about 0.57% too long. The whole set sits within about 1.4% of correct tine length. Turning that into millimeters requires assuming a tine thickness the seller does not publish, so treat what follows as illustration only: using the 1992 paper's own 7 mm aluminum tines, the errors would be on the order of two millimeters on the 174 fork, one on the 852 and half a millimeter on the 528. These are final-tuning and finishing errors, not exotic ones. Forks are brought to pitch by grinding metal away at the last stage, which is exactly why they are cheap to get right and cheap to get wrong.

The honest verdict is three words long, and it is the same three the last review reached: credible, unverified, narrow. One observer, one set, nine forks, one session, on an unnamed instrument, against the 248 ratings recorded on 2026-08-29. It establishes nothing about the other sets shipped. But nothing about temperature, striking force or tip damage explains the pattern, and the seller published no figure anyone could have measured against. That omission is the fair criticism, and until it is fixed nobody outside the company can settle this. For contrast, a proper measurement looks like NIST's own procedure: a GPS-disciplined counter reading to 0.1 hertz, a clamping contact microphone, three strikes averaged with the fork flipped 180 degrees between each, temperature logged, two hours of equilibration beforehand, and an expanded uncertainty of ±0.02%. Nobody unboxing a set at home is doing that. The manufacturer is supposed to have done it, and to print the result.

Does Twenty Cents Matter? Three Different Answers

This is the question a buyer actually wants answered, and it has three answers rather than one, because it depends entirely on what the fork is used against. Neither of this publisher's books contains any material on cents, and neither treats acoustic beating between two struck forks — the handbook's beat material is about binaural beats delivered separately to the two ears, and its interval material is confined to octave arithmetic. Everything numerical in this section was therefore computed for this page and is not attributed to them.

Sounded alone, with nothing to compare it to, essentially nobody can tell. Identifying that a tone is 522 rather than 528 in isolation requires absolute pitch, which is rare. A listener with a fork in one hand and no reference has no way to detect a fifth of a semitone, and would not experience the tone as wrong.

Sounded against a true reference at the marked frequency — an app tone, or a correctly tuned second fork — the beat rate is exact arithmetic rather than psychoacoustics. Two tones six hertz apart beat six times a second, so a 522 fork against a true 528 source produces a six-per-second wobble. The other beat rates against a true reference, from this buyer's readings, are four per second on the 174 fork, three on the 396, three on the 417, four on the 639, nine on the 741, sixteen on the 963 and twenty-three on the 852. What a listener makes of the faster rates — countable pulses, or a roughness in the tone — depends on the listener and the level, and this page cites no psychoacoustic source for where that crossover sits. So the arithmetic is printed here and no threshold is.

Judged by comparing two successive tones with no simultaneous reference — the ordinary case of playing a fork and then an app tone — published pitch-discrimination thresholds vary widely with the listener, the level and the timbre, from a few cents for trained listeners under laboratory conditions to considerably more for untrained ones. Twenty cents sits inside that spread. The honest answer there is that it depends on the listener, and no single threshold number is printed here because the literature does not supply one.

Used the way the seller actually intends — two forks sounded together — the errors partly cancel, and this is the part that runs in the set's favor. The Solfeggio numbers happen to contain exact just intervals: 528 ÷ 396 and 852 ÷ 639 are both exactly 4/3, a just perfect fourth of 498.0 cents. Measured on this buyer's forks, 522 against 393 gives 491.4 cents, an error of only about seven cents, which is inaudible in practice. But 829 against 635 gives 461.5 cents, an error of about thirty-seven cents, because those two forks err in the same direction by different amounts. Of the seven pairs checked below, five land within ten cents of their intended interval and two are off by more than thirty.

One acoustic fact makes even the bad pairs less obvious than their cents figures suggest, and it belongs here in fairness to the seller. A softly struck fork produces a nearly pure tone with none of the integer harmonics that most musical instruments carry. Two forks sounding together therefore have no coinciding upper partials to beat against each other. The sharp, obvious roughness that gives away a mistuned interval on two guitars or two voices is simply absent with forks, which means a thirty-cent interval error between two of them is considerably harder to hear than the number implies.

For on-body contact use, a deviation of this size sits below anything skin has been shown to resolve — comfortably so against the typical figures, and only narrowly so against the most favorable one in the literature. Published Weber fractions for vibrotactile frequency discrimination sit around twenty percent in the flutter range and around thirty percent higher up; the most favorable figure anywhere in that literature is about three percent, and it comes from a special stimulus — trains of half-sinusoid pulses delivered to the finger — rather than from a struck fork. The largest deviation in this set is 2.70%, which is inside even that best case but not by much. Two further limits belong to the conclusion. That most favorable study tops out at 384 hertz while this set reaches 963, so nothing in the figure quoted here speaks to the upper forks at all. And the listing material recorded on 2026-08-29 never says whether these forks are weighted, which is the design feature that matters for contact use in the first place. So: if the intended use is vibration felt on the body, these deviations are very probably irrelevant. If the intended use is listening against a reference, they are not.

Two closing observations, one in the seller's favor and one against. In the seller's favor: the handbook's own decision order for choosing a tone puts safety and comfort first, then delivery route, then clarity of signal, then personal preference, then goal and ritual meaning, and the exact number last. By that order, twenty cents is a long way down the list of things that should decide a purchase. Against: accuracy is the thing this particular listing sells on, and accuracy is what this buyer measured. The gap between the claim and the delivery stands on its own regardless of whether anyone could hear it. And one thing no buyer should expect from this set in any case: none of these nine numbers lands on an equal-tempered note. Four of the nine — 285, 639, 852 and 963 — sit more than forty cents from the nearest one, close to a quarter-tone. This set will not agree with a piano, and it was never meant to.

Where the 528 Hz DNA Claim Actually Came From

The claim on this listing is not hedged and it is not buried. The product title attaches DNA to the set as its selling purpose. A bullet calls 528 hertz "the frequency of love" and then attaches DNA repair to it; the same bullet frames the set in medical terms. Both the DNA wording and the medical wording are described here rather than reproduced. Those are the seller's claims and are not restated anywhere on this page as things the object does. What follows is where those words came from, traced honestly, because tracing them is more useful to a buyer than arguing with them.

The nine numbers are modern. The six-tone core — 396, 417, 528, 639, 741 and 852 hertz — was published and popularized in the 1999 book Healing Codes for the Biological Apocalypse, co-authored by Joseph Puleo, a naturopath and herbalist who in the 1990s described a digit-reduction procedure applied to the verses of Numbers 7:12–83 and read the resulting integers as hertz values, and Leonard Horowitz, a former dentist turned writer on alternative health. That book is the primary source for the modern Solfeggio claim set. The three now-standard additions — 174, 285 and 963 — belong to a later nine-tone expansion, and chapter 6 of The Tuning Fork Healing Handbook declines to assign those three to a single verified original source. No prior text — medieval, Renaissance, early modern or nineteenth-century — contains this specific set of nine frequencies. This site already publishes the full account at the guide on Solfeggio frequencies, and this page links there rather than re-running it.

But the arithmetic is worth three lines here, because a reader can check it without trusting anyone. Lay the nine numbers out in ascending order as a three-by-three grid — 174, 285, 396 across the top; 417, 528, 639 in the middle; 741, 852, 963 at the bottom. Every row steps by exactly 111. Every column steps by 243 and then by 324. Every one of the nine has a digital root of 3, 6 or 9. And each of the nine is a cyclic rotation of one of just three seed numbers: 174 rotates to 741 to 417 and back; 285 to 852 to 528; 396 to 963 to 639. That is not a hidden pattern discovered in scripture. The six core tones are the output of a procedure that selected for threes, sixes and nines, and finding threes, sixes and nines in them confirms only that the procedure ran; the three later additions continue the same arithmetic, whoever added them. This is exact arithmetic, not an argument, which is why it is safe to print.

"The frequency of love" is Horowitz's phrase and his commercial brand, not a traditional or a scientific descriptor. He built later output around it — LOVE the Real Da Vinci CODE, and The Book of 528: Prosperity Key of LOVE in 2011 — along with a 528 love revolution campaign and a radio venture selling music transposed so that C falls at 528 hertz, which is to say a tuning reference of A = 444. When a listing calls 528 the frequency of love, it is quoting a book title, and readers should know that.

The medieval costume deserves precision in both directions, because the syllables engraved on these forks are genuinely old and the numbers attached to them are not. Guido d'Arezzo, born around 991, drew ut, re, mi, fa, sol and la from the opening syllables of successive lines of the hymn Ut queant laxis. That is a real and consequential system, and it is a ladder of relative intervals: a re is a whole step above an ut wherever the singer starts. Guido never wrote a number in hertz, because the unit did not exist — Hertz demonstrated electromagnetic waves in the 1880s, the name was attached to the unit by the International Electrotechnical Commission around 1930, and it replaced cycles per second in the international system of units only in 1960. So TENFLY's UT, RE, MI, FA, SOL and LA engravings are conventionally correct as syllables. What is modern is bolting a cycles-per-second value onto each one. The handbook is careful about the opposite overstatement too: older cultures were not imprecise about pitch, they used intervals, reference instruments, pipes, strings and local standards. The modern element is the universal numerical label, not the existence of musical precision.

One more strand, because it is the one usually offered as evidence. The DNA claim is typically sourced to a purported John Hutchison or Glen Rein experiment. Both were checked. Glen Rein's work, published between 1993 and 1996 mostly in conference proceedings, with one peer-reviewed paper in a specialist journal in 1995, measured ultraviolet absorption in DNA solutions before and after exposure to human intention. Not sound. Not 528 hertz. The most-cited item is a paper in the proceedings of a 1996 conference in Denver, and it names no acoustic frequency at all: no hertz value appears in it, and neither does 528, nor sound. The wavelengths it does name are ultraviolet, which is a different region of the spectrum from audible sound rather than a neighboring one. The Hutchison and Lazaryan claim to have cleared oil-spill water with Solfeggio tones in 2010 circulates through blogs, a slide document and a commercial lab certificate; no peer-reviewed publication of it could be found, so it is described here as circulating rather than as having been examined. That is the whole lineage behind the sound-and-DNA link: a proceedings paper about conscious intention, and an unpublished remediation story.

What the Published Research on 528 Hz Actually Measured

Here is the sentence to keep, and it survives checking in both directions: no published experiment could be located that measured DNA — its structure, its damage or its repair — after exposure to a 528 hertz sound, and this page searched for one specifically. The nearest thing in the record is a rodent study that measured gene expression, which is a different quantity from DNA structure, damage or repair. That absence is not a gap waiting to be filled. This site's guide on what frequency therapy cannot do files the claim as contradicted rather than merely unproven, and it does so on a scale argument: a 528 hertz sound wave in air is about 65 centimeters long, computed here from the speed of sound divided by the frequency, against a DNA double helix about two nanometers across. The argument is set out in full there. There is research in this space, some of it decent, and it is described below. Every study found measured something else.

The paper the whole edifice rests on is a 2017 cell-culture study from a Tehran laboratory, published in an open-access addiction-therapy journal, on human astrocyte cultures treated with ethanol. Its full text was read for this page. What it measured was cell viability, membrane damage, reactive oxygen species and cell appearance under a microscope. It ran no DNA assay of any kind. DNA nevertheless appears in its abstract, in its introduction and in its discussion — the discussion being the authors' own interpretive section rather than background — but never as something the study measured. Every one of those mentions reports a claim carried in from elsewhere, and each traces to the same entry in the paper's own reference list: number 17, Horowitz's Healing Codes for the Biological Apocalypse, which the paper dates 2000 and which first appeared in 1999. That is the single most useful fact on this page. The study everyone cites for 528 hertz and DNA did not measure DNA, and cites the origin book as its authority for the DNA claim. The loop closes on itself.

The study's own findings deserve reporting straight, not sneered at and not enlarged. Its statistics rest on three independent determinations. The sound alone, at three different levels, produced no change in cell death. With ethanol present, the quietest exposure reduced cell death, by 18 to 20% depending on which part of the paper is read. Above that the paper stops agreeing with itself: the text reports no significant change at 100 dB and increased cell death at 120, while Figure 2's caption reports improved viability at 100 dB and no change at 120. A result a paper cannot describe consistently across its own text and figures is not a dose-response; it is a result asking for replication. The authors' own conclusion is modest and is about reducing the toxic effects of ethanol on cells in culture. That is the honest ceiling: one lab, cultured cells, three determinations. As for the venue, one fact can be stated flatly because a court found it: in the Federal Trade Commission's case against that journal's publisher, the US District Court for the District of Nevada entered summary judgment in March 2019 and imposed a $50.1 million judgment, finding deceptive claims about peer review and about database indexing.

The most-circulated human study is a 2018 paper in the journal Health. Its full text was read for this page. Nine healthy adults — one man and eight women, aged 26 to 37 — five minutes of soothing piano music, each hearing both conditions on separate days without being told which. Two things about it are decisive and neither is a matter of interpretation. First, the paper states its own method: every comparison it ran was a within-condition test between time points. No 528-versus-440 comparison was ever performed. Its headline — that one tuning has an especially strong stress-reducing effect — is inferred entirely from a result reaching significance in one arm and not in the other, and a difference in significance is not a significant difference. Second, and more fundamental: the study did not test a 528 hertz tone at all. Its introduction says plainly that there is no 528 hertz note in the ordinary scale, and that setting the reference tone to A = 444 puts one there. So the two arms were two tunings of the same piano recording. Computed here, A = 444 against A = 440 is about sixteen cents, and 444 raised by three equal-tempered semitones gives 528.01 hertz. The entire difference between the arms of the flagship human study on 528 hertz is about sixteen cents of tuning on a piece of music — smaller, as an observation about scale rather than an argument, than the twenty-cent gap this buyer measured on their own 528 fork. Worth noting too: that paper's own introduction says the effects ascribed to solfeggio frequencies have no scientific basis.

One properly designed, adequately sized, between-group randomized trial could be located, and it is the study on this page that deserves the most respect. Published in Brain and Behavior in 2026, it randomized 162 healthy university students in permuted blocks to three parallel arms of fifty-four — silence, 528 hertz music, and 432 hertz music — for twenty minutes through headphones during an exam period. Saliva was collected only after exposure, with no baseline, so every inference rests on the randomization. It found between-group differences in three salivary markers. On CREB the 528 arm was significantly higher than both the silence arm and the 432 arm. On BDNF the 432 arm was significantly lower than both of the others. On GRP78 the 432 arm was significantly higher than both. The paper is not consistent with itself about the 528 arm's BDNF: its abstract describes 528 hertz music as significantly increasing salivary BDNF and CREB compared with control, while its results section describes the 528 arm's BDNF as comparable to control and only the 432 arm as significantly lower. The results section is followed here, and the discrepancy is recorded rather than passed over, because the abstract is the version most citing sources will carry and it is the version that favors the claim this page is examining. The one functional measure, a Stroop task, showed no between-group difference at all. The authors' stated limits are that participants were aged eighteen to twenty-five, the exposure was twenty minutes, musical preference and prior experience were uncontrolled, and saliva may not reflect what happens in the central nervous system. No spectral definition of the two musical stimuli could be found in the paper, so it cannot be assumed they were tuning references in the sense the 2018 study used.

That description completes rather than contradicts what chapter 6 of The Tuning Fork Healing Handbook says about the same trial. The book reports the 432 arm's three markers against silence and the Stroop null, and treats the salivary findings as preliminary and hypothesis-generating, which is right. The trial's own between-group finding on CREB in the 528 arm belongs beside them. It is a real signal in salivary markers between two versions of a piece of music. A salivary marker is not a health outcome: none of the three is a symptom, a diagnosis or a clinical endpoint, and no benefit to anyone follows from the result. It is not DNA, it is not a fork, and the only functional outcome measured came out flat.

Two further studies exist and neither rescues the title claim. A 2019 rodent study in Genes & Genomics, from the same Tehran laboratory, reported effects of prolonged loud 528 hertz exposure on brain testosterone and on anxiety-like behavior in rats. Only the abstract-level record was read for this page, so no figures from it are printed here. It is a rodent study in a legitimate journal; it measured gene expression rather than DNA structure, damage or repair, and continuous loud exposure is nothing like a struck fork. A 2024 human anxiety study, published in a veterinary journal, randomized 48 people to three minutes of 528 hertz against a neutral reading task and measured a standard anxiety inventory. Because the comparator was reading rather than music at another tuning, the design cannot separate 528 hertz from listening to music at all, and its own abstract is internally inconsistent about whether a pure tone or music was administered.

The mechanism question is handled here in words, and the numerical version is left where it already lives. DNA absorbs energy in the ultraviolet, near 260 nanometers of wavelength — a different region of the spectrum from audible sound, not a neighboring one. This site's guide on what frequency therapy cannot do carries the full argument, and a reader who wants it should go there rather than take a compressed version from a product review.

One framing note, offered as description and not as a legal conclusion about this seller. The FTC's Health Products Compliance Guidance, issued 20 December 2022, extends to health-related equipment and devices, sets the bar at competent and reliable scientific evidence framed as randomized controlled human clinical testing, and states that animal studies, in vitro studies and anecdotal evidence are not generally sufficient. Most of the 528 hertz literature falls into exactly those categories: cell culture, rodents and testimony. The one clear exception is the 2026 trial above, which is randomized, controlled and human — and which measured salivary markers rather than any health outcome, and found nothing on the single functional task it ran. It is cited here from this site's own earlier reading on 2026-08-29, because ftc.gov returned an access error when the wording was re-checked for this page and it was not verified again today.

So where does the claim sit on the handbook's five-level evidence ladder? The tone itself is direct evidence: a fork produces a measurable acoustic and mechanical signal, its pitch can be measured, people can hear and feel it. That is the whole of what is established. "The frequency of love" and the goal words attached to the other eight tones are modern convention, dated to 1999 and after. There is no traditional-use rung available, because the tradition invoked — Guido, the hymn, Gregorian chant — is relative-interval pedagogy that carried no hertz values at all. And DNA repair is not on the ladder. Chapter 6 states flatly that the claim is not supported by clinical evidence, and this site's own guide files it separately from the merely unproven labels on the other eight tones, as the one claim in the set that fails on physical scale rather than merely lacking support. It is worth adding that chapter 3's list of listing warning signs names promises to detox, repair DNA, destroy pathogens, balance hormones, or replace treatment in a single bullet, and the second item in that list is this listing's title claim. This publisher's book identified it as a warning sign before this listing was ever read.

Which leaves one clean logical point, and it is about the seller's own two bullets rather than about biology. This listing sells the set on one exact number, and separately advertises "more accurate frequency." The two bullets are not in conflict — they reinforce each other, which is precisely the problem. Advertising a specific hertz value as carrying a specific power makes accuracy load-bearing rather than cosmetic. Having made accuracy the point, the seller then declines to publish any figure it could be measured against, while the one buyer who did measure found 522. And on the ladder, the stated frequency is the one thing on this entire listing that is direct evidence — checkable in principle, as chapter 3 says plainly, because the stated frequency can be measured. That is exactly why the missing tolerance is the most substantive omission on the page.

The Seller's Claims, One at a Time

Every claim below is quoted or summarized from the listing as recorded on 2026-08-29 and then labeled. The labels are the same ones used on this site's earlier review: documented means it can be checked and holds; convention means it is a modern association rather than a finding; not supported means no evidence for this setup and this outcome could be found; not checkable means the claim states nothing that could be tested; and plausible but unverified means what it says.

Two of them are worth a sentence outside the table. The first is the accuracy bullet, which is the one this whole page turns on. It promises "more accurate frequency," which is a comparative with nothing named to compare against — more accurate than what, measured how, to what tolerance, at what temperature? None of that appears anywhere in the listing material recorded on 2026-08-29. And the same sentence promises "frequency up to 25 seconds," which conflates frequency with sustain: a duration is not a frequency. The one bullet on this listing that promises precision is imprecise about what it is measuring.

The second is the satisfaction bullet, and it counts in the seller's favor. Chapter 3 tells a buyer to look for a return route for a damaged, loose or badly out-of-tolerance instrument, and lists the absence of one as a warning sign. TENFLY offers twelve months of satisfaction service. That box is ticked, on the listing's own say-so, while the stated tolerance and the weighted-or-unweighted question are not.

Aluminum Versus Steel, and Twenty-Five Seconds

The listing makes an explicit superiority claim about its metal, so it gets checked in both directions. The handbook's own position is a refusal to take either side: aluminum-alloy and steel forks are both common, their weight, sustain, timbre, durability and price can differ, and an exotic metal name does not establish a wellness effect. What it tells a buyer to prioritize instead is stable construction, secure weights, smooth edges, a usable contact base, stated quality control, manufacturer support, and a sound you can use comfortably. That is the frame. What follows fills in the four things the seller's sentence actually asserts.

On pitch, the two metals are interchangeable and it is not close. A fork's frequency depends on the material only through the square root of stiffness divided by density. Computed here from published materials-reference values (MakeItFrom.com, read 2026-08-30), 6061 aluminum comes to about 5,055 meters per second and 304 stainless to about 5,064 — a difference of 0.17%. Steel's roughly threefold greater stiffness is very nearly cancelled by its roughly threefold greater density. For a fork of given dimensions, swapping one metal for the other barely moves the pitch, and the length and thickness of the tines do essentially all the work. Neither metal is inherently more accurate than the other, which means "more accurate frequency" cannot be earned by choosing aluminum. It is worth adding that aluminum is unremarkable as a laboratory fork material: the 1992 acoustics paper describes aluminum forks as what its own authors' general physics laboratories used, and Russell's 2020 survey photographs a steel set and an aluminum set side by side as standard apparatus. Neither metal is the outsider here, and nothing on this page says aluminum is a bad choice. It says the superiority claim is not established.

On weight, the seller is straightforwardly right, and this is a real ergonomic difference rather than a quality one. Aluminum runs about 2.7 grams per cubic centimeter against roughly 7.8 for stainless, so an aluminum fork of the same shape is about a third of the mass. For a nine-fork set that a beginner will hold repeatedly, that matters.

On corrosion, the claim holds against one kind of steel and fails against another. Bare aluminum relies on a natural oxide film and does well indoors and in dry conditions, though light pitting can appear outdoors within a few years without anodizing. Stainless steel carries at least 10.5% chromium and forms a self-healing passive layer with excellent resistance across a wide range of atmospheres. So a corrosion advantage over plain carbon steel is real, and over stainless it is not. The listing says only "the steel tuning forks on the market" without naming carbon or stainless, which leaves the comparator undefined — the same defect as "more accurate."

On durability, the claim runs backwards, and on a tuning fork that is not a cosmetic point. Aluminum alloys are markedly softer than fork steels: the comparison source used here lists tensile strength for 6061 at 130 to 410 megapascals against 580 to 1180 for 304 stainless. Russell's warning is that striking a fork on a hard surface or with a metal object can dent it and change its frequency. Softness on a fork is therefore a pitch-stability issue. That sits directly beside the same buyer's report that some forks in their box looked damaged on the tips, and it is the one place where the seller's own comparison points at a genuine weakness rather than a strength.

Only one peer-reviewed clinical head-to-head comparison of the two metals in forks could be located, and it does not favor aluminum. A 2013 study in Otolaryngology–Head and Neck Surgery, prospective and observational with fifty participants, compared 512 hertz steel and aluminum forks in a hearing examination, with bench testing on a sound-level meter, microphone and artificial mastoid. It found steel had more balanced air and bone conduction efficiencies, with aluminum relatively lower on bone conduction. That must be labeled for what it is: a hearing-examination study, not a sound-therapy study. It establishes only that fork material changes how efficiently vibratory energy couples into a body — in that test, in steel's favor — and it says nothing about whether either metal is better for anything else, because nobody appears to have tested that. It is cited here to stop the seller's comparative standing unexamined, not to invert it. And chapter 3's Safety First box is explicit that these instruments are for personal wellness practice, not home diagnosis — not for testing yourself for hearing loss, neuropathy, fracture, circulation problems or any disease.

On ring time there is one independent matched-geometry comparison, and it is academic rather than commercial. Burleigh and Fuierer machined tuning forks from seventeen alloys and one polymer to identical dimensions, within tens of micrometres, and compared pitch, harmonics and damping. Their aluminum-and-steel observation is about pitch rather than sustain: they report that steel and 6061 aluminum sound similar, because steel's roughly threefold advantage in stiffness is offset by its roughly threefold greater density — the same cancellation computed above, reached independently, which is the strongest confirmation on this page that the pitch argument is right. On duration they report only that some alloys damp out within seconds while others ring loud and long, with no figures for the aluminum-and-steel pair. So the sustain question is genuinely open, but it is not true that nobody outside the industry has compared the two metals. A correction to this site's own record belongs here too: the earlier review recorded the primary acoustics reference on fork modes — Rossing, Russell and Brown, 1992 — as paywalled and unread. The authors' own copy is freely available, and all seven pages were read for this page on 2026-08-30. The paper contains no decay times and no quality factors for any mode, so it cannot settle the sustain question either.

Which leaves the twenty-five seconds. A ring of that length is not implausible for an unweighted fork, but no unit was examined here and nothing on this page verifies it. And it is not a specification. Three problems, in order. The bullet is malformed on its face — a duration stated as if it were a frequency. Sustain judged by ear depends on the listening threshold you choose, the ambient noise, how hard the fork was struck, whether the stem is touching anything, and even orientation: Russell reports the sound is roughly 5 decibels louder with the tines aligned with the ear than perpendicular to it. With none of that stated, the figure cannot be falsified. And one number is being asked to cover nine forks of nine different sizes, with no fork named; nothing in the sources read for this page gives a decay time or a quality factor for any of them, so how much the ring should differ across the set cannot be computed here. What the physics does explain is why forks ring long at all. In its principal mode a struck fork radiates as a linear quadrupole, a famously inefficient radiator, so very little of its energy leaves as sound and most of it stays in the metal. The clang mode radiates far more efficiently, which is why that component — at roughly six times the fundamental — is loud for only an instant and then gone while the fundamental persists, and why placing the stem on a sounding board makes a fork much louder and correspondingly shortens the ring. Chapter 3 puts the buying consequence in one line: a long ring is pleasant but does not by itself prove pitch accuracy.

Measured Against the Book's Own Listing Checklist

Chapter 3 of The Tuning Fork Healing Handbook contains a section called "How to Read a Product Listing," written long before this product was examined. That makes it the fairest available rubric, because the criteria were not chosen with this listing in front of anyone. The table below runs the eight items against what TENFLY actually provides.

On the seven warning signs the same section lists, this set trips three outright. The book names promises to repair DNA explicitly, and the product title contains exactly that. It names claims that one metal or number guarantees a biological outcome, and this listing makes both — the aluminum-beats-steel bullet and the 528 bullet. And it names no explanation of whether the fork is weighted, which the recorded material does not give. One clarification about the count, because it would be easy to inflate: the same bullet in the book that names promises to repair DNA also names promises to replace treatment, so those are one warning sign and not two. This listing trips it on the first clause outright; on the second it goes as far as a general medical framing rather than a named condition or a promise to replace anything, and readers can weigh that for themselves. Against all this, two warning signs are clearly avoided: the frequencies are stated, and a return route exists in the form of the twelve-month satisfaction service.

On set size, the book's position is specific and it costs sellers money. Its skip list opens with large chakra, planetary or Solfeggio collections bought before you know which tones you enjoy — which is this exact product, for a first-time buyer. Its three sanctioned starter kits contain one fork, one fork and two forks; the minimum-cost one does not require a fork at all, accepting a phone tone in place of one; and none of the three contains a Solfeggio frequency. Its Myth versus Fact entry is blunt: a larger set offers more pitches, not more established benefit, and you should add a tone only when repeated use, rather than a marketing chart, gives you a reason. Chapter 3 says in as many words that there is no reason for a beginner to buy all nine. None of that makes a nine-fork set a bad object. It makes it the wrong first purchase for someone who has not yet decided whether they want contact vibration or listening.

Two smaller things, and one of them goes the seller's way. The book's storage rule is that forks should be stored separately or in a divided pouch so that prongs and weights do not strike each other; a buyer reports each fork individually sleeved, which is what that rule asks for. Nothing here compares that packing to other sets in the category, because nothing in the record supports such a comparison. On the other side, the recorded listing material contains no cleaning or care instructions, which is one of the eight items, and the book's own guidance is not to soak a fork or use alcohol, bleach, essential oils, abrasives or harsh cleaners unless the manufacturer confirms compatibility. A listing that says nothing leaves that unanswered.

The chapter also asks a buyer to keep a short equipment record for every fork they own: the fork, its frequency, its design, where it was bought, its stated tolerance, the date it was checked and its preferred use. The stated purpose is to prevent labels and marketing stories from replacing the facts about the instrument you actually own. Anyone buying this set can fill in every column but one, and the missing column is the one this whole page has been circling. The chapter's listing section closes with the sentence that is the right sentence for this product: buy the instrument, not the medical story wrapped around it.

The Category Norm, and Who This Set Fits

The most useful thing a buyer can be told before searching this term is that the DNA claim is not one rogue seller. Of the nine top listings returned by the "solfeggio tuning fork set" search read on 2026-08-29, at least four put DNA in the product title itself. This site's earlier review had already flagged a competing Solfeggio listing making a DNA claim in a completely different search on that same date. So the norm was visible then and is confirmed now at the top of this category. A buyer comparison-shopping in this niche will not escape the claim by switching sellers, which means the claim carries no information about which product is better made. It has to be set aside entirely, and the objects compared on the things that can be checked.

On the leadership question, precision matters. This set led that particular search by rating count on 2026-08-29, with 248 ratings. Rating count measures accumulated reviews, not sales. It is not the best-selling tuning fork set in any general sense: its Best Sellers Rank that day was #35,390 in Musical Instruments, and a different set reviewed earlier on this site carried 622 ratings on the same date, in a different search and under a different Amazon category tree. Best Sellers Ranks in different category trees are not comparable numbers and are not compared here.

One caution before the table, in the same terms the last review used. On 2026-08-29, a 4.8 average from 40 ratings is not a stronger result than a 4.6 from 248; it is a much smaller sample, and small samples move. All of these figures come from one browsing session on one date. None of the alternatives below has been examined the way this one has, and their listings have not been checked for claims.

So who is this set for? It fits a reader who already knows they want the Solfeggio series as a symbolic or ritual framework, who wants nine pitches rather than one, and who is buying an object rather than a promise. On that reading the reports are consistent and favorable across more than three years of reviews and several countries: good metal, good finish, individually sleeved forks, two mallets, a pouch and a cloth. It also fits someone who simply wants nine mid-range pitches to explore, provided they understand these are not equal-tempered notes and will not agree with a piano.

It does not fit a beginner who has not yet chosen between contact vibration and listening. That reader needs one fork, an activator and a pouch, and the book's own minimum-cost starting point does not require a fork at all — it accepts a phone tone in place of one. Nine forks bought before that comparison are nine forks chosen from a chart. It does not fit anyone buying because of the title, for the reasons set out at length above. And it does not fit a reader for whom a published tolerance is a requirement — that reader should ask any seller in this category for the figure and the reference temperature, because the answer is a fact about the product rather than a matter of taste. Riverbank's own caution belongs beside that advice: the maker warns buyers against paying more for claims of tighter tuning that too often disappoint. Chasing the smallest percentage is not the point. Whether a figure exists at all is.

One safety note, because the listing uses medical language and this page will not. These are acoustic instruments, and nothing here treats, relieves or diagnoses anything. Anyone considering on-body use should read chapter 5's P1–P8 placement rules and its per-point exclusions first — and note that the recorded listing material does not say whether these forks are weighted, which is precisely the design feature that determines whether contact use is appropriate at all. Chapter 3's Safety First box is explicit that these instruments are for personal wellness practice and not for home diagnosis of hearing loss, neuropathy, fracture, circulation problems or any disease, and the resemblance between a wellness fork and a clinical one does not license a clinical use. If something needs attention, it needs a clinician.

The last word belongs to the object, and the only witnesses to it are the buyers. Nine pieces of aluminium that several independent buyers thought were well made, in a case, with mallets, at a price this page deliberately does not print. No unit was examined here, so that is their verdict and not this page’s. One buyer measured them and found them mostly a little flat, with no published figure to measure against. The story printed on the box goes back to a 1999 numerology book by way of a cell-culture paper that measured no DNA. Buy the instrument, not the medical story wrapped around it.