Is the structure of water different after homeopathic dilution and succussion?

Homeopaths have long been searching for a possible mechanism of action for their remedies, even though they lack that most basic prerequisite for an effective medicine: evidence that it works better than a placebo. In the last twenty years, homeopaths have come to focus on something called “the memory of water”, as being a likely candidate for this mechanism. Briefly, the story is as follows:

Homeopathy is supposed to work on the principle of “like cures like”. Its founder, Samuel Hahnemann, convinced himself of this during his research into the properties of cinchona bark, a potent source of quinine. Quinine was, and is, widely used for the treatment of malaria, and Hahnemann discovered that ingesting large doses of it produced malaria-like symptoms in a healthy person. This led him to believe that what caused symptoms of a disease in a healthy person, could also be used to cure a person suffering from that same disease: “like cures like”.

The trouble was that the doses of the substance that Hahnemann used to produce the said symptoms were so high as to be positively toxic. He therefore decided to work with dilutions of the substances he was testing. He soon discovered another curious thing: the substances, when prepared using a method that involved vigorous shaking (which he called “succussion”) of the substance between dilutions, were just as good at producing symptoms, although just as curiously, no-one else was ever able to replicate the effect. Bizarrely, Hahnemann proposed that, when diluted, the original substance was effective as a treatment against the disease whose symptoms it produced in a healthy person; moreover, the more the substance was diluted, the more potent the effect!

Hahnemann was working in the latter part of the eighteenth century, and knew nothing about the molecular basis of chemistry. He can thus be forgiven for claiming increased potency with increased dilution. Nowadays, we know that the dilutions that he used were so high as to make it improbable that a single molecule of the original substance remained in the water. We therefore have no excuse for claiming anything other than that homeopathic remedies are indistinguishable from plain old water.

Such minor considerations as these have not shaken homeopaths from their convictions: starting in the 1980s with French immunologist Jacques Benveniste and continuing through to the 21st century with the work of Madeleine Ennis in Belfast and Louis Rey in Paris, homeopathy proponents have tried to show that water somehow retains a “memory” of the substances that were once dissolved in it. This memory, they say, accounts for the supposed therapeutic effect.

Benveniste started it all with a paper published in the journal Nature in 1988. His team exposed human basophils (a type of white blood cell that can be stimulated to secrete histamine, a chemical involved in allergic reactions) to solutions of water that had once contained human antibodies, but were now diluted to the extent that not a single molecule of anything once dissolved in it could remain. Amazingly, the basophils responded by secreting histamine, as they would have done if the antibodies had actually been present. More intriguingly, the effect only worked when the solution was shaken violently, a la homeopathic succussion. The basophils had also been exposed to water that had never contained human antibodies, and when this happened, they did not release histamine.

Benveniste published his paper in Nature, who sent a team to investigate his work. The team discovered that, in the original experiments, the experimenters had known which solutions had once contained antibodies and which had only contained pure water. When the experiment was repeated in such a way that the experimenters did not know which solution was which, the effect disappeared. Benveniste was forced to retract his paper. His reputation was destroyed, not so much by this experiment, but by his refusal to accept that there was no such memory of water, and by his later claims, which became increasingly odd, and which culminated in a 1997 paper stating that the effect could be transmitted over telephone lines.

The idea, once ignited, refused to go away. In 2003, the Swiss chemist, Louis Rey published a paper in Physica A, a reputable journal specialising in statistical mechanics. Rey was using the phenomenon of thermoluminescence to study the structure of solids. The technique involves bombarding a cold sample with radiation, warming it up, and analysing the light emitted, which reveals something of the structure of the sample.

Rey diluted solutions of sodium and lithium chloride to homeopathic levels (probably no molecules of the original substance remained), succussed them, flooded them with radiation, and warmed them up. He noticed that the thermoluminescence peaks were characteristic of the original substances that were dissolved in the water, even though the original substances were long diluted out of the solution. His interpretation was that the networks of hydrogen bonds in the samples were different, retaining thus a “memory” of the substance that was originally dissolved in them.

Unfortunately, it seems no-one has yet replicated Rey’s experiment to show that such an effect exists. Other experts in the field of hydrogen bonds in water are unconvinced by his methodology, and given the known problems due to the experimenter effect with many other such attempts to show that water has a memory, one would not be optimistic about an early resolution to the issue. I have no reason to doubt Rey’s integrity, but the example of Jan Hendrik Schon should suffice to warn us that unique results from one person or group should be treated with caution.