Dilutions or delusions?

Dilutions or delusions?

"An experiment whose conclusions have no physical basis is described in this week's [june 30] issue," say the editors of NATURE. With the comment, they feature a paper by 13 researchers testing the hallmark of homeopathic medicin: the use of minute doses of drugs that bring on symptoms similar to the disease to stimulate a cure.

The international team appears to have demonstrated that while blood cells called basophils react to an antibody solution after the equivalent of 120 tenfold diluations. The antibodies normally bind to and disrupt basophil membranes until the membranes can no longer hold a laboratroy stain, a process known as degranulation. Such an infinitesimal concentration, however, should not contain enough antibody molecules to damage the cells. In theory, all the antibody would have been diluted out. But the scientists found 40 to 60 percent of the basophil membranes disrupted by such a weak solution, catching attention from the French press and homeopathy fans worldwide prior to the study's publication.

Six laboratories in France, Italy, Israel and Canada duplicated the elaborate series of blinded experiments, which involved diluting the antibody solution, labeling cells and detecting degranulation after exposure to the solution. At the request of NATURE's editors, the study's primary author, Jacques Benveniste of the University of Paris-Sud, has agreed to repeat the procedure under the watchful eyes of independent investigators.

"It's just important that the work is published," observes Patricia Fortner of the University of Toronto and one of the study's authors, "so we can get feedback from around the world."

The researchers build their hypothesis to explain the startling findings around the molecular behavior of water. The solution had to be vigorously shaken for at least 10 seconds before it would cause degranulation. That means, they say, that "specific information must have been transmitted during the dilution/shaking process. Water could act as a "template' for the [antibody] molecule, for example, by an infinite hydrogen-bonded network, or electric and magnetic fields."

NATURE Editor John Maddox offers a simpler explanation: "It must be some systematic error." Although the original paper was reviewed for two years before publication, Maddox says the results of Benveniste's second run of the experiment will appear in NATURE later this summer.