In April 2025, a team at the University of Massachusetts Amherst, led by graduate student Anthony Raykh, published a finding in Nature Physics that surprised even them. While experimenting with oil, water, and magnetised nickel nanoparticles, Raykh found that no matter how vigorously the mixture was shaken, it consistently returned to the same elegant urn-like shape. Even altering the size of the magnetic particles did not change the effect. Simulations by collaborators at Tufts and Syracuse universities showed that the strongly magnetised particles increased interfacial tension rather than decreasing it - the opposite of what particles normally do in emulsions. The result was what the team called a "shape-recovering liquid."
The structure was not fragile. It was robust. The harder they shook it, the more reliably it reformed.
It is worth noting that the nickel particles, simply placed in the oil-water mixture without agitation, produced no such structure. They would settle at the interface or sink. The urn shape appeared only after vigorous shaking - the mechanical agitation was essential. Neither the substance alone nor the agitation alone was sufficient. It was the combination that produced the structure.
When I came across this information, being a homeopath, I was intrigued. I did some research using Claude AI. It led to a very interesting trail of evidence, and I came across several interesting, established scientific facts that span multiple domains - water structure, colloidal chemistry, nanoparticle self-assembly, biomineralization and - magnetotactic bacteria! All pointing to one conclusion - as described in the title of this article. Read on.
It is commonly assumed that liquid water is a formless medium - molecules jostling randomly with no persistent arrangement. Research shows otherwise. Liquid water contains the local structural arrangements found in all seventeen known phases of ice. It flickers constantly between a denser, disordered state and a more ordered, ice-like state with stronger hydrogen bonds. At room temperature, it forms three-dimensional cage-like clusters extending beyond 3 nanometres, encompassing thousands of molecules, displaying fractal behaviour and forming percolating networks. Five-membered rings are a dominant topology, and specific cluster shapes - pentamers, hexamers forming cage structures - show up repeatedly as preferred arrangements.
When water is vigorously shaken, measurable physicochemical changes occur - shifts in UV scattering spectra that persist for several hours, changes in dissolved gas concentration, formation of reactive oxygen species, and luminescence in the blue spectrum. These are attributed to nanobubble formation and gas chemistry.
In colloidal chemistry, different nanoparticles self-assemble into different patterns. Gold nanoparticles, silica nanoparticles, and polymer micelles each produce distinct arrangements because their physical properties - size, charge, surface chemistry - impose different constraints on how they can arrange themselves.
Calcium carbonate can crystallise as vaterite (snowflake-shaped), aragonite (needle-shaped), or rhombohedral calcite. The presence of trace amounts of iron or nickel ions, in concentrations as low as 1.5 millimolar, triggers an entirely different crystal habit - star-shaped crystals emerge instead. A trace presence of a different substance changes the structure entirely.
Even ocean floor manganese nodules form through colloidal chemical processes: dissolved manganese precipitates around nucleation points - a shark's tooth, a shell fragment - with negatively charged hydrated manganese dioxide colloids adsorbing metal ions and growing concentrically over millions of years, forming layered structures that reflect the specific chemical environment of their formation.
There is no known case in chemistry or biology where two different substances produce identical structures in a medium. Substance identity determines structural outcome. This is universal.
Strangely, even organisms do this. Magnetotactic bacteria produce chains of magnetite crystals inside their cells. The crystals have species-specific morphology - cuboctahedral, elongated prismatic, or bullet-shaped - depending on which species of bacterium is producing them. The same mineral, the same raw material, but different organisms impose different shapes. Different information applied to the same substrate produces different forms.
If water
then we need to consider the following hypothesis:
Different substances, mechanically agitated with water under identical conditions, should produce structurally distinguishable results.
This hypothesis, fortunately, can be tested:
Take five substances with different physical properties - for example, silica, iron, nickel, sulphur, and sodium chloride. Prepare identical samples of purified water. Subject each sample to the same process of mechanical agitation with the substance. Use a control sample of water subjected to identical agitation without any substance.
Then examine the water using the tools that already exist for studying water structure: Raman spectroscopy, NMR spectroscopy (both 1H and 17O), infrared spectroscopy, X-ray diffraction, neutron diffraction, UV-Vis spectroscopy, and electron microscopy. These are the same instruments used in the water cluster research cited above. No new technology is needed.
There are three possible outcomes:
1. All five samples look identical to each other and to the control. Then mechanical agitation with a substance leaves no structural trace in the water. The hypothesis is wrong.
2. All five samples look identical to each other but different from the control. Then the process of agitation alters water's structure, but the substance's identity is irrelevant. The process leaves a trace in the water, but not the substance.
3. All five samples are structurally distinguishable from each other, and point clearly to the substance that was introduced. It has left an imprint on the water.
If the last, i.e. (3), happens, then the substance's physical properties shaped the water's structural arrangement during agitation, and different substances produce different results. Moreover, these structural changes persist.
A fourth possibility was suggested to me: that the five substances produce different structures, but randomly - different each time the experiment is repeated. However, the UMass finding already addresses this: nickel produced the same structure every time, no matter how vigorously or how often it was shaken. The structure was reproducible. If the same holds for other substances, each would have its own consistent signature.
In 1988, the French immunologist Jacques Benveniste published a paper in Nature claiming that highly diluted antibody solutions retained biological activity - as if the water had "remembered" the antibody. The paper was investigated, the results could not be reliably replicated, and Benveniste's career was effectively destroyed.
Later, Luc Montagnier, the Nobel laureate who co-discovered HIV, published work claiming that DNA emits electromagnetic signals that produce structural changes in water persisting even at extreme dilutions. He was widely ridiculed.
Both researchers were attempting to demonstrate a biological effect of structured water - that the water could do something therapeutically. This required the scientific community to accept two extraordinary claims simultaneously: that water retains substance-specific structure, and that this structure has biological activity.
It should be noted that the UMass team attributed the shape-recovering effect specifically to magnetic interactions between the ferromagnetic nickel particles, and that when non-magnetic silica particles were tested, the effect did not occur. Whether non-magnetic substances can produce analogous stable structures through other physical mechanisms - charge, particle geometry, surface chemistry - remains untested. This is precisely why the proposed experiment matters.
A simpler first step would be to stay within the magnetic mechanism that already works: use iron, cobalt, and magnetite nanoparticles alongside nickel. Same type of force, different substance. If the shapes differ, the substance matters, not just the magnetism.
For non-magnetic substances, a different approach may be needed. Rather than using bulk particles, one could try nanoparticles - created by what Hahnemann called "trituration," a mechanical grinding process that reduces substances to the micro-nano scale (a well-established process in mechanochemistry). At this scale, surface properties change fundamentally: insoluble substances become soluble, reactivity increases, and the particle surface is structurally different from the bulk material. Whether nanoparticles of non-magnetic substances, introduced into water through mechanical agitation, produce substance-specific structural changes is an open question - and a testable one.
If this finding holds true, and different ferromagnetic substances do produce different structures, the study could then be extended to non-magnetic substances to see whether they too have some structural impact. It may not be a visible shape - it may be something we have not yet seen or measured.
In a conversation with a patient who is a scientist, I raised the question of why the simpler, purely structural question had not been tested independently. His response was candid: the homoeopathic association makes the question professionally dangerous. A materials scientist investigating substance-specific structural traces in mechanically agitated liquids would be doing clean science. But if the work can be connected to homoeopathy, the career risk is significant. The question is avoided not because it has been answered, but because asking it is costly.
I asked him: how can science hold a position on a question without looking at it?
He said: that's just how it is.
It seems that mechanical agitation of a liquid in the presence of a substance can produce stable, reproducible structures. A physician and scientist over two centuries ago called this process succussion, and it was the mainstay of his medicinal preparations. In 2025, science showed it makes a difference.
Now we know it happens with nickel. But that is like saying apples fall. If we find that not just apples, but many other things fall, the gravity of that finding would be phenomenal.
The question remains open. The tools exist. The experiment is simple - just the UMass experiment, with two more substances.