Water Science — What the Physics Actually Supports
Water Science — What the Physics Actually Supports
Sub-article of Hydration — Wheel of Health. See also: Water — The Most Underestimated Medicine, Supplementation, Monitor, Purification.
Water is the only substance on earth whose solid form floats on its liquid. Every lake that freezes from the top down, every fish that survives a winter, every cell that does not rupture in the cold, depends on an anomaly that has no business existing. Water should behave like the other hydrides in its column of the periodic table. Hydrogen sulfide boils at minus sixty degrees. By the same logic water should boil at around minus eighty, which would leave the planet dry and this article unnecessary.
It does not, because of the hydrogen bond — a weak electrostatic attraction between the partial positive charge on hydrogen and the lone electron pairs on a neighbouring oxygen. Each water molecule can form four of them. The result is a liquid held together far more strongly than its molecular weight predicts, and every anomaly follows from that single structural fact.
This article carries the physics and the biochemistry that the Hydration pillar states as position and the Water article delivers as protocol. It exists because the water discourse is unusually corrupted in both directions — an industry selling purification devices on one side, an industry selling dissolved minerals on the other, and a research literature thin enough that both sides can quote it. The practitioner needs to know which claims the physics funds, which the epidemiology funds, which rest on a single untranslated document, and which remain genuinely open.
The Anomalies
Water’s peculiarities are not trivia. Each one is load-bearing for biology.
Maximum density occurs at 3.98 degrees Celsius rather than at the freezing point. Ice is roughly nine percent less dense than the liquid it forms from, because the hydrogen-bond network locks into an open hexagonal lattice with more empty space than the jostling liquid contains. Bodies of water therefore freeze from the surface downward, insulating what lies beneath.
The specific heat capacity is 4.18 joules per gram per kelvin — among the highest of any common substance. Heating water means breaking hydrogen bonds before kinetic energy can rise, so water absorbs enormous thermal input for modest temperature change. This is why an organism that is sixty percent water holds core temperature through a hot afternoon, and why the ocean governs climate.
Surface tension is the highest of any liquid except mercury. Capillary action follows from it, and with capillary action comes the movement of sap up a tree against gravity and the movement of fluid through the finest vessels of the body.
Water dissolves more substances than any other liquid. The molecule is bent rather than linear, which leaves it electrically lopsided — oxygen holding partial negative charge, the two hydrogens partial positive. That polarity lets water surround and separate ions, which is why blood carries what it carries and why every metabolic reaction happens in aqueous solution rather than in air.
These properties are not disputed by anyone. They are the ground the contested claims stand on, and they are the reason the Wheel of Health places Hydration where it does — not as one input among many but as the medium every other input operates within.
The Interface
The interesting question is what water does near a surface, because that is the only condition under which biology encounters it. There is very little bulk water in a living organism. There is water against membranes, against proteins, against the interior faces of organelles.
Gerald Pollack’s laboratory at the University of Washington documented that water adjacent to hydrophilic surfaces forms a layer, sometimes hundreds of micrometres deep, that excludes solutes — microspheres suspended in the water are pushed out of the zone and accumulate beyond its boundary. The layer carries net negative charge; the region beyond it carries the complementary positive. Pollack calls it the exclusion zone and argues it constitutes a fourth phase of water, structurally intermediate between liquid and ice, with the stoichiometry H₃O₂ rather than H₂O.
The observation is real and reproducible. Solute exclusion at hydrophilic interfaces happens; the charge separation is measurable; infrared input expands the zone. Where the interpretation is contested is whether this requires a new phase of matter or is explained by conventional interfacial chemistry — electrical double layers, the physics of which has been understood since Gouy and Chapman a century ago. That dispute is live and the article does not need to settle it, because the biologically consequential claim survives either reading: water at an interface is not the same water as water in a glass, and the interior of a cell is nothing but interface.
The objection raised against restructuring devices runs through a single number. In bulk liquid water the hydrogen-bond network reorganizes on a timescale of roughly one picosecond — a trillionth of a second — from which it is inferred that no imposed order can persist long enough to matter. The inference moves faster than the physics permits. The picosecond figure describes the lifetime of an individual hydrogen bond in undisturbed bulk liquid. It does not describe the persistence of mesoscale ordering, which is a different observable at a different length scale, and the exclusion zone itself is the standing counterexample: a structure that holds for hours against a network whose individual bonds turn over a trillion times a second. Order at the level of the ensemble does not require any particular bond to survive. A vortex holds its shape while every molecule in it is replaced.
That does not convert the device claim into an established one, and it should not. The honest statement is that the mechanism is unresolved in both directions — no one has demonstrated that imposed ordering survives storage and gastric transit, and no one has demonstrated that it cannot. What the field has is the interfacial physics, which is solid, and a set of claimed effects downstream of resonance and vortex treatment, which are reported but not yet mechanistically closed.
Four registers, kept apart. Harmonism holds that water is not inert to its environment — it receives form, holds form, releases form, and participates in Logos as a medium rather than as a substance acted upon. Restructuring is therefore a real intervention on a real property, and the doctrine does not wait on the mechanism. The evidence establishes exclusion-zone formation at hydrophilic surfaces, its charge separation, and its expansion under infrared input. The traditions witness the same thing from the other side, and unanimously — the spring over the cistern, moving water over standing, the vortex the river makes on its own. What remains open is the mechanism: whether the exclusion zone requires Pollack’s fourth phase or is conventional double-layer chemistry, and by what route imposed ordering persists to the cell.
The system commits to the claim and holds the mechanism open. Those are different acts, and collapsing them in either direction is the error — asserting the mechanism as settled overreaches, and withholding the claim until the mechanism arrives is permission-seeking.
Hydrogen enrichment stands on a different kind of ground, and the difference is worth naming rather than blurring. Molecular hydrogen is a dissolved gas at a concentration you can measure with a reagent kit — the target range is 0.5 to 1.6 milligrams per litre — and its selectivity for hydroxyl radicals over the reactive oxygen species the body uses for signalling is mechanistically specified. A 2023 review identifies 81 clinical trials and 64 human-study publications; the larger figures in circulation, six hundred and a thousand, come from advocacy sources and count preclinical animal and cell work, which is where the bulk of the literature still sits. Structuring is verified at the level of the property and open at the level of the mechanism. Hydrogen is verified at the level of the mechanism and still accumulating at the level of the clinical outcome. Neither is the other’s proof, and a practitioner who knows which is which spends better.
What Dissolved Solids Measure
Total dissolved solids is the combined mass of everything in solution — inorganic salts, dissolved organic matter, whatever the water has met on its way. A handheld meter does not measure it directly. The meter measures electrical conductivity and multiplies by a conversion factor, typically around 0.5 to 0.7, calibrated against a standard salt solution.
This matters more than it sounds. Conductivity responds to charged species. It is nearly blind to uncharged contaminants, which is a category containing most pharmaceutical residues, most pesticides, most volatile organic compounds, and every microplastic particle. A reading of 12 parts per million tells you the water is low in dissolved salts. It tells you nothing about whether the water contains atrazine.
The reverse confusion is equally common. A spring reading 280 ppm is not thereby contaminated; it is mineralized. TDS is a measure of quantity, not of quality, and the two get conflated in both directions constantly — by people who treat a low number as proof of purity and by people who treat a high number as proof of goodness.
Regulators have been consistent about this. The United States sets 500 mg/L as a National Secondary Drinking Water Regulation — the secondary category is explicitly aesthetic and non-enforceable, concerning taste and scaling rather than health. Canada’s guideline matches it on the same reasoning. The World Health Organization’s Guidelines for Drinking-water Quality, fourth edition incorporating the first and second addenda, proposes no health-based guideline value for TDS at all, and discusses palatability and scale rather than physiology. No national authority anywhere has set a minimum.
Three Claims Under One Name
The low-TDS controversy is not one argument. It is three, and they have three different evidentiary statuses, and nearly all of the confusion in the popular discourse comes from treating a refutation of one as a refutation of the others.
The leaching claim
The first claim is that pure water strips minerals from body tissue — that drinking it dissolves you from the inside, pulling calcium from bone the way it would pull limescale from a kettle.
Trace it back and the whole edifice rests on one document. A 1980 report submitted to the World Health Organization carried an annex reviewing Russian-language research from the A.N. Sysin Institute in Moscow, describing experiments in rats, dogs, and human volunteers, and concluding that water below 100 mg/L disturbs water-salt balance and promotes ion loss. No official translation of the cited articles has surfaced. Two reviewers went through the annex in detail: Lee Rozelle, a membrane chemist, and Ronald Wathen, a nephrologist. Their reviews are worth reading in full because they are unusually blunt.
Rozelle noted that the dog data showed the same physiological changes at 50 mg/L and at 1,000 mg/L — a twentyfold difference in exposure producing no difference in effect, which is what a null result looks like when it is being reported as a finding. Volume consumed was not stated. Duration for the human subjects was not stated. In several passages it was not determinable whether the data described animals or people.
Wathen went at the mechanism. Salts are not leached from the human body; they are preferentially retained or excreted according to whether the organism is surfeit in water or salt. His formulation is the one to keep: the human body is not a lead or copper pipe.
The physiology backs him. Homeostasis holds extracellular osmolality at roughly 300 milliosmoles per litre — about 9,000 mg/L expressed as dissolved solids, which is to say the internal environment is two orders of magnitude more concentrated than any drinking water under discussion. The kidneys filter around 180 litres of plasma daily and reabsorb over ninety-nine percent of it, adjusting output through antidiuretic hormone, aldosterone, and thirst, with a rise in osmolality of about one percent sufficient to trigger the last of these. Osmotic gradients across cell membranes correct within a minute because water crosses membranes freely.
And the water never reaches the intestine at the concentration it left the glass. Saliva runs around 15 milliequivalents per litre of sodium chloride and about 30 of potassium; a single millilitre of it raises the dissolved-solids content of a 250 millilitre glass by roughly 10 mg/L. Gastric and intestinal secretions add far more. By the time absorption happens, the distinction between 5 ppm and 100 ppm has been erased by the body’s own fluids.
Then there is the field record, which is the part of this that is difficult to argue with. The United States Navy has distilled seawater for crew consumption for four decades, with reported values below 3 mg/L and submarine deployments running months at a time. NASA has flown crews on water at roughly 0.05 mg/L. The Environmental Protection Agency ran a supply at San Ysidro, New Mexico from 800 mg/L down to 40–70 mg/L. A field trial in Boulder, Colorado put around fifty households on water at 0.05 ppm. Thousands of private wells and small municipal systems deliver naturally low-TDS water to populations that have been drinking it for generations.
No mineral-depletion syndrome has emerged from any of them.
The verdict is that the leaching claim fails. It fails on provenance, on mechanism, and on the field record, and it fails without needing the trade association that most loudly refutes it — more on that shortly.
The contribution claim
The second claim is different and it is standing. It does not say that pure water takes minerals out. It says that mineralized water puts some in, and that removing the water removes the contribution.
Run the arithmetic before accepting the claim at the weight its advocates give it, because the arithmetic is decisive and almost nobody does it.
Calcium requirement runs at 1,000 milligrams daily for adults and 1,200 for women past fifty. Hard tap water at 80 mg/L — the upper end of what Kozisek calls optimum — would require 12.5 litres a day to meet it, and 15 litres for the older woman. Soft to moderate water at 40 mg/L requires 25 litres. At a realistic 2.5 litres daily, hard water delivers 200 milligrams: one fifth of requirement, from water at the hard end of the range. A tin of sardines eaten with the bones carries 382 milligrams, which is 4.8 litres of that same water in a single sitting. A cup of cooked kale beats two litres of it.
Magnesium runs at 420 milligrams daily for men and 320 for women. Water at 30 mg/L — the figure Israel’s remineralization pilot targets, and roughly what the country’s aquifer supply carried before desalination — requires 14 litres a day for a man. At the 10 mg/L most Israeli cities actually run, 42 litres. At 2.5 litres daily, the 30 mg/L water contributes 75 milligrams, under a fifth of requirement; the 10 mg/L water contributes 25 milligrams, six percent. Thirty grams of pumpkin seeds is 156 milligrams — 5.2 litres of the good water, in a handful.
Only bottled mineral water reaches interesting numbers, and it does so by being a different product: a high-magnesium bottled source at around 108 mg/L would cover roughly two thirds of male requirement at 2.5 litres daily. That is a supplement sold as a beverage, not what runs through a municipal main.
The absorption is not the issue and should not be claimed as one. Magnesium arrives in water as a free ion with no food matrix competing for it, and uptake is at least comparable to food. The issue is quantity. Water is an insignificant mineral source at every concentration a population actually drinks, and the mineral case has to be made on the margin rather than on the contribution.
The strongest evidence is not a trial. It is a country. Israel moved a large share of its municipal supply onto desalinated seawater, and reverse-osmosis desalination produces water with essentially no magnesium; post-treatment adds calcium for corrosion control and leaves magnesium out. Most Israeli cities now run under 10 mg/L against a WHO-discussed range of 25 to 50. Yona Amitai and colleagues at Bar-Ilan followed roughly 178,000 members of Clalit Health Services and reported approximately six percent higher incidence of heart disease among those on desalinated supply. A 2016 Sheba Medical Center analysis reported that cardiac patients in desalinated-supply areas died at about twice the rate of those on aquifer supply. An interministerial committee recommended magnesium reintroduction in 2010. The first pilot — raising magnesium from 5 to 30 mg/L at a single plant, privately funded — began in July 2026, sixteen years later.
This is association, not demonstrated causation, and supply zones differ in more than their water. But the finding does not need to be disputed, because the arithmetic above explains it without granting water any status as a mineral source. Roughly half of the adult population in industrialized countries falls below the estimated average requirement for magnesium on diet alone. For a population sitting in that deficit, water at 30 mg/L was supplying the last fifth of requirement — the fifth that carried them over the line. Remove it and the marginal cases fall under. A six percent shift in cardiac incidence is what the tail of a distribution looks like when its floor is withdrawn, and it is entirely compatible with water being a trivial source for anyone not standing on that floor.
Two claims that sound opposed are both true. Water contributes an insignificant fraction of mineral requirement. Removing that insignificant fraction from an already-deficient population produces a measurable population effect. The second follows from the first rather than contradicting it, and any reading that has to reject one to hold the other has misunderstood where the effect lives.
And it is messy. A 2017 systematic review with meta-analysis pooled seven case-control studies across roughly 44,000 subjects and found protective effect sizes of 0.82 for calcium and 0.75 for magnesium — but with heterogeneity above seventy percent for both and publication bias the authors named themselves. Against it, the British Regional Heart Study reported in 2008, on cohort data rather than ecological correlation, that hard water does not protect against cardiovascular disease. Germany’s Federal Institute for Risk Assessment, which sells nothing, assessed water at or below 50 mg/L as posing no risk to the general population on a balanced diet, while flagging specific groups: people on restricted diets, those avoiding whole food categories, pregnant and breastfeeding women, and anyone in strenuous activity beyond about ninety minutes.
So the resolution is a distinction between two populations rather than a compromise between two positions. Municipal mineral content matters at civilizational scale precisely because populations do not supplement — the public does not track intake, does not test, does not correct, and a passive delivery mechanism reaching everyone regardless of attention is the only floor most people have. Withdraw it and the least attentive fall first. For the practitioner who tests and repletes deliberately at three or four hundred milligrams, the same withdrawal is arithmetically invisible; seventy-five milligrams was never load-bearing for someone already above requirement.
That is Monitor doing structural work rather than decorative work. The Harmonist protocol does not escape the finding by disputing it. It escapes by occupying the exception the finding defines — and the exception is not a loophole but the entire point of a monitored practice.
The corrosivity claim
The third claim is real, well-understood, and about plumbing rather than physiology — which is why it gets missed by both camps.
Water low in dissolved solids and low in alkalinity is chemically aggressive toward metal. With little buffering capacity and no scale-forming tendency, it dissolves what it flows through: lead from solder and older service lines, copper from pipe, zinc from brass fittings. Municipal systems manage this deliberately, which is what the WHO guidelines mean when they discuss dissolved solids in terms of distribution-main passivation rather than health — the scale layer is protective infrastructure.
The practical consequence is a design rule rather than a health warning. Point-of-use purification, where the membrane sits at the tap and the product water travels centimetres of inert tubing to a glass vessel, has no corrosivity exposure worth discussing. Whole-house reverse osmosis feeding an existing metal plumbing system does, and needs remineralization or pH correction after the membrane, not for the drinker’s sake but for the pipe’s.
The same chemistry has one consequence at the vessel. Ultra-pure water is a more aggressive solvent than mineralized water, so it extracts more from whatever holds it. Stored in plastic, it takes up more of the plastic. This is the strongest argument for glass in the entire hydration protocol, and it is an argument that only exists because the water is pure.
Reading the Discourse
The evidence on this question is not contested in a vacuum. Both sides sell something, and the practitioner reads better knowing what.
The most thorough refutation of the leaching claim is a technical bulletin published by the Water Quality Association — the trade body for the water treatment industry, whose members manufacture and sell the reverse-osmosis and distillation equipment the bulletin defends. Its authors are named and several are credentialed chemists. Its provenance analysis is correct, its appendix reviews are substantive, and its field record is verifiable. It is also, straightforwardly, an interested party producing a document that concludes its members’ products are safe.
The most cited case for minimum mineral content is a chapter by Frantisek Kozisek in a WHO-published symposium volume, Nutrients in Drinking Water, recommending a TDS minimum of 100 mg/L with an optimum of 250 to 500, calcium at 40 to 80, and magnesium at 20 to 30. Kozisek sits at the Czech National Institute of Public Health and is not commercially interested. His chapter is regularly cited as the WHO position, which it is not — the symposium’s own conclusions were silent on the matter, and the current Guidelines propose no TDS value in either direction. And the evidence base beneath his physiological claims is substantially the same Soviet material the trade association dismantles. The two documents are not independent readings of a common record. They are two interested readings of one thin record, one of them state-employed and one of them industry-funded.
Meanwhile the bottled mineral water sector markets high dissolved-solids content as a health property, and has every incentive for the contribution claim to be larger than the arithmetic supports.
Harmonism holds that a captured discourse is read by mapping the interests before weighing the claims, and that mapping interests is not the same as dismissing findings. The trade association is right about leaching. Kozisek is right that water-borne minerals are bioavailable. Neither is right because of who funded them, and neither is wrong because of it.
The Form Question
A recurring position in integrative health holds that minerals are either organic and absorbable or inorganic and useless. The instinct behind it is sound and the axis it uses is the wrong one. Organic versus inorganic is a statement about carbon chemistry, and carbon chemistry does not predict what crosses an intestinal wall.
The critique is correct about supplement forms. Calcium carbonate — ground limestone, oyster shell, dolomite — requires substantial gastric acid to liberate the calcium ion, absorbs at roughly a quarter to a third under favourable conditions, drops further in the older and the hypochlorhydric, and contributes to the arterial and articular mis-deposition that follows calcium supplied without the vitamin K2 and magnesium cofactors that direct it to bone. Magnesium oxide is worse: cheap, common in mass-market supplements, absorbed at single-digit percentages, and mostly functioning as a laxative. Against these, the citrate, malate, glycinate, and taurate forms absorb several times better. The practitioner who reads labels on this axis is doing something real.
The axis that does predict it is whether the mineral arrives already dissociated as a free ion or arrives as a salt still requiring gastric acid to liberate it. That single distinction sorts the field correctly, and it sorts it differently from the organic-inorganic split. Calcium carbonate is inorganic and fails. Magnesium oxide is inorganic and fails. Magnesium glycinate is organic and succeeds. And dissolved sea minerals, marine plasma, and the mineral content of a spring are all inorganic by the chemist’s definition and succeed — because they arrive as Mg²⁺ and Ca²⁺ already in solution, already hydrated by the water molecules surrounding them, with the dissolution accomplished before the glass was filled.
This is why the marine sources earn their standing rather than borrowing it. René Quinton built an entire therapeutics on the isomorphism between seawater and the internal environment, and the ocean is not a mineral slurry but a living medium — plankton and algae cycling its content through biological form, its whole mass in perpetual motion, its trace spectrum complete rather than curated. Unrefined sea salt carries that spectrum; the refined product carries two elements of it. The claim to make is not that these are organic. It is that they are ionic, complete, and biologically cycled, and that the axis on which they win is the one that governs.
So the sentence to retire is minerals in water are inorganic and poorly absorbed. It is wrong twice — the chemistry of the term and the physiology of the claim. The defensible version is narrower and unassailable: water is an insignificant mineral source at any concentration a person actually drinks, and an unreliable one besides, since its content is set by geology rather than by need and cannot be titrated. Deliberate repletion beats incidental dosing, not because the incidental dose fails to absorb but because a practitioner should not outsource mineral status to whatever stone the water last touched.
That argument survives every study anyone can bring. The absorption argument does not survive one.
Hydration Status and Function
The physiology worth carrying is not about how much to drink. It is about what the body does when the answer is wrong in either direction.
Cell volume is a regulated variable and not merely a consequence of intake. Cells swell and shrink against osmotic gradients, and the volume state itself operates as a signal — hepatocyte swelling promotes anabolic activity, protein and glycogen synthesis, while shrinkage shifts the cell toward catabolism and breakdown. Chronic mild dehydration is therefore not a subjectively unpleasant state with no downstream consequence. It is a persistent catabolic bias.
At the other extreme sits dilutional hyponatraemia — sodium falling below about 135 millimoles per litre because water intake outpaces the kidney’s maximum free-water clearance of roughly 800 to 1,000 millilitres per hour. Water moves into cells down the osmotic gradient. In the brain, confined by the skull, that swelling produces headache, confusion, seizure, and at the extreme death. Endurance athletes have died of it. So have participants in water-drinking contests.
The condition is not caused by low-TDS water — sodium loss and excess volume cause it — but the practitioner running ultra-pure water is systematically closer to it than the practitioner drinking mineralized water, and four situations compound the exposure: extended fasting, heavy sauna use, ketogenic eating with its characteristic sodium and magnesium wasting, and prolonged endurance output. Under any of them, daily electrolyte repletion is not optional. It is the condition on which the rest of the protocol is safe.
This is the one place where the low-TDS position carries genuine risk, and it deserves to be stated at full strength rather than buried in a caution paragraph. Purity has a cost, and the cost is that nothing arrives incidentally.
The Harmonist Position
Four claims, marked by status.
Doctrine. Water is the medium of life, not a nutrient among nutrients. It participates in Logos, receives and releases form, and is not inert to what it meets. The purification architecture is sound because contaminant removal has no plausible downside and a documented upside, and because the practitioner should govern what enters the body by intention rather than by municipal accident.
Evidence. Contaminant reduction through reverse osmosis followed by distillation is established. Molecular hydrogen at 0.5 to 1.6 mg/L is measurable and mechanistically specified, with 81 clinical trials logged as of 2023 and the bulk of the literature still preclinical. Exclusion-zone formation at hydrophilic interfaces is real. Water-borne minerals absorb well and supply an insignificant fraction of requirement — 12.5 litres of hard water for a day’s calcium, 14 litres of good water for a day’s magnesium. Low-TDS water does not leach minerals from tissue.
Tradition. No cartography theorizes dissolved solids, and the absence should be named rather than papered over. The Indian, Chinese, Shamanic, Greek and Abrahamic streams all carry water discipline, and all of it concerns source, movement, vessel, temperature and timing — living water over standing, the spring over the cistern, the copper vessel overnight, the fast broken with warm water rather than cold. That convergence is real and it is about something else. Manufacturing a traditional endorsement of a TDS target would be a flattening.
Open. The mechanism by which imposed ordering persists through storage and gastric transit. Whether the exclusion zone requires a fourth phase or is conventional interfacial chemistry. Whether the hardness-cardiovascular association is causal or confounded by everything else that varies between water districts.
The protocol that follows is the one the Water article describes. Reverse osmosis first, then distillation of the permeate — the membrane takes dissolved solids, heavy metals and fluoride, the still takes the volatile organics that pass through a membrane, and the two together approach as close to zero as a domestic architecture reaches. Restructure, because water holds form and the practitioner works with a medium rather than on a commodity. Enrich with hydrogen. Then replete deliberately, from marine and ionic sources chosen for their spectrum, at a dose measured rather than inherited.
No remineralization cartridge. A cartridge redelivers, at the tap and without discrimination, the arbitrary geological content the architecture just spent two stages removing — and it does so downstream of every decision the practitioner might otherwise make. The point of purifying to zero is not that minerals are unwanted. It is that from zero, every mineral entering the body is chosen.
The water is the solvent. The minerals are a decision. Keeping those separate is the whole of the position, and it is why the purification and the supplementation belong to different spokes of the same wheel rather than competing for one.
The Practice
Test the source before designing anything. A dissolved-solids meter costs almost nothing and tells you almost nothing on its own; a full panel — heavy metals, volatile organics, disinfection byproducts, nitrate — tells you what the architecture actually needs to remove. The practitioner who buys a system before testing has bought a solution to an unexamined problem.
Put the membrane at the point of use, not the point of entry, unless the whole-house case has been made on other grounds and the corrosivity has been handled downstream of it. Store in glass, and understand that the reason is the purity itself.
Then measure what the water no longer carries. Serum magnesium is a poor marker — most of the body’s magnesium is intracellular and serum is defended late — so red blood cell magnesium is the test worth running, alongside serum sodium and potassium if fasting, sauna, or endurance work is part of the practice. The Monitor discipline is what makes ultra-pure water safe. Without it, the protocol has removed a floor and installed nothing in its place.
Water is not made good by having nothing in it. It is made good by having nothing in it that you did not put there.
See also: Hydration, Water, Supplementation, Monitor, Purification, Wheel of Health.