By Dana Ullman MPH
Below is dialogue between Lionel Milgrom and Wayne Turnbull. Lionel is an experienced chemist and a homeopath. He has also written about homeopathic research for The New Scientist as well as other publications. Lionel, John Morgan (the head pharmacist at Helios, a leading homeopathic pharmacy in England ), and Francis Treuherz (a respected homeopath) observed the making of the homeopathic doses of histamine. Although these 3 were seemingly complimentary to the pharmacological process during the making of the medicine (and in front of the TV cameras), afterwards (and BEFORE the unveiling of the result of the experiment), Lionel Milgrom realized that there were some problems in the making of the potencies that may have influenced the outcome. Whether these problems were serious or not isn’t essential. I am including it here because Turnbull’s email provides verification that he was not following Ennis’ protocol and because the sloppiness in the making of the medicine is a part of the problem that occurred.
FROM LIONEL MILGROM) Sent on December 5, 2003, before the results of the study were revealed…
Dear Wayne (Turnbull of Guy’s Hospital),
Thanks for allowing us into your lab last week.
I’ve been thinking about this experiment all week, and I have to admit that there were a couple of things that I found troubling. I didn’t mention them at the time (or rather I did but not so that you would notice, and certainly not on screen), mainly because I thought that in the context of what was trying to be achieved, it probably didn’t matter. Now, with hindsight, I am not so sure.
If I didn’t explain it too well, as well as being a homeopath, I am also a practising chemist with over 30 years experience in research and teaching. As a chemist, I was appalled by the way the histamine was weighed out and the original stock solution of histamine made up by Wendy. At first, I thought that it was my professional pride that was injured. Actually, it could be more serious than that.
First, she tried to accurately weigh out 50 mgs of histamine on a balance that is a) not very accurate for weighing such small quantities: it is more suited to roughly weighing tens and hundreds of grams, and b) it was not protected from drafts. I saw the scale pan vere from between 40 and 60 mgs due to gale that was blowing out of the near-by fume cupboard. This means an error in the weighing of something like 20-25%. There was a four-figure balance opposite. Why was this not used, as by all standards it should have been?
Then, horror or horrors, she added distilled water to the weighing boat while it was still on the scale-pan! I doubt if you routinely make up standardised solutions in your lab via this technique – if you do then the firing squad surely beckons! How do you know that all the histamine was dissolved in the boat? Apart from being sheer sloppy technique, this is a surefire method of ensuring the short life of what is not an inexpensive piece of laboratory equipment. Hence, there is now another potentially large source of error in the concentration of the histamine solution.
Let us assume, that having acquired this histamine solution of for all intents and purposes, highly inaccurate concentration, the further dilutions and succussions were carried out correctly (certainly, from what John and I saw, this would seem to be a reasonable assumption). This now means that the same inaccuracies have been transferred to the dilutions. Thus, a 15c dilution actually isn’t. John runs a highly respected homeopathic pharmacy and I am absolutely certain that such inaccuracies in potency would never pass muster in his establishment. So why were they allowed in yours? More to the point, the Ennis data (by the way, I believe, that she now has some reservations about whether her protocol was actually followed to the letter) seems to indicate a periodicity in the activity of the potencies. Because of the now large inaccuracies carried over from the weighing and making up of the original histamine solution, even if such periodicities are observed, we will now not know which potencies they are and whether they duplicate or not Ennis’s original findings.
So why didn’t I mention this at the time? Simple. As the histamine was to be effectively diluted out of existence, I thought that it didn’t matter all that much. Having thought long and hard, I now believe that it might, and I am kicking myself that I never said anything. Consequently, and without even seeing the flow cytometry results, I must now sadly conclude that whatever the result of this experiment, it was actually a complete waste of everybody’s time.
The study of the putative effects of ultra-high dilutions is not even a fledgling science. It is difficult and exacting work, something I know from my own hard experience. It is absolutely essential that effective and agreed protocols be found for whatever physical and/or biological effect one wishes to study and to be rigidly adhered to time and time again, until they are proved faulty for whatever reason. This means that the protocols themselves have to constantly monitored and checked for faults. And faulty protocols in the bio-medical sciences are by no means uncommon. In my own work of late, this is something I have had to come to terms with on several occasions. Again, as I am finding by hard experience, the exactitude of the perhaps more naive physical sciences does not easily or readily transfer to the bio-medical sciences.
Those attempting this work have to be prepared to accurately and laboriously repeat over and over again the necessary steps in preparation and dilution/succussion that go with the territory. And there are no short cuts. In this game, there is no such thing as a killer ‘one-off’ experiment, however much one would wish it. It grieves me to have to say that it is about time that such work was taken seriously and not used as televised gimmicry. My advice would be to scrap this experiment and start all over again, and this time not under the glare of television lights.
With Best Wishes
Dr Lionel R Milgrom
FROM WAYNE TURNBULL:
At 17:26 12/12/03 +0000, you wrote:
> Thank you for your mail. I hope I can address the issues that you’ve raised. First of all, as far as I’m concerned this whole project was never discussed, presented or conducted as an exercise into the fundamental validity of either homeopathy or the memory of water. The result is NOT that homeopathy has no credibility, or that the memory of water is nonsense. The result only allows us to say that with 16C, 17C and 18C dilutions and a sample size of 420, no significant difference could be found between histamine and water in their effect upon the IgE mediated degranulation of normal human basophils. Not much of a sound bite I agree, but that’s the finding. It doesn’t discredit homeopathy or water memory. There is a significant but subtle difference between disproving a theory and not proving a theory.
> After taking the decision to get involved in this project, and after a lot of discussion with Mark Golden at ABC News, I have tried to be both accomodating and inclusive regarding the involvement at any stage, with the homeopathic community. My personal opinion of homeopathy and water memory is an irrelevance, and I am not going to put that opinion on record here. I have no axe to grind and no agenda to serve. I got involved because it was an opportunity to perform “real science”. Here’s an experiment, you don’t know the answers, let’s see what happens. Given an assay that I was confident to stand behind, the eventual results would be valid, regardless of whether they turned out to be positive or negative. The pursuit of scientific enquiry requires an abstract objectivity. Not only do I believe that I am perfectly capable of such impartiality, I had hoped that I was successful in communicating that impartiality to Dana, John, Francis and yourself. I have tried to be as receptive as practicable when comments have been made, because a consensus between all parties is essential when performing this experiment, otherwise we arrive at the situation we now find ourselves in, with phrases such as television gimmicry being tossed around.
>In answer to your specific criticisms regarding the initial dilution procedure, several points need to be made.
>I would agree that the balance used is not optimal for controlling the measurement of small amounts of material, but I would consider that it is acceptable. If it wasn’t acceptable, we would not have used it.
>My recollection of the “scale pan vere” is different to yours. Although there was a variation in measurement from 40 to 60 mgs, the balance settled at 50. I don’t know of a chemical balance that will produce a weight measurement that remains rock-solid, with no wavering whatsoever.
>I find your statement regarding an apparent draught in the room as extremely odd. There could not possibly be a gale issuing from the fume cupboard, as, by their very nature, fume cupboards DRAW air from the room they’re placed in, and draw that air very weakly.
>The 4 point balance was not available. As a dept we are in the process of re-siting a lot of personnel and equipment into new lab space.
>The addition of distilled water to the boat was performed to ensure that all of the weighed histamine was transferred to solution. The distilled water used was taken from the 5 mls already aliquotted for this 1% w/v solution. Washing the boat in this way ensures as much of the weighed histamine ends up into a defined volume (in this case 5 mls).
>”John runs a highly respected homeopathic pharmacy and I am absolutely certain that such inaccuracies in potency would never pass muster in his establishment. So why were they allowed in yours?” Unfortunately the answer to this question is because you allowed it to happen. Francis, John and yourself were actively encouraged to express any comments, reservations or objections that you had at the time. We were more than willing to listen. I believe we went out of our way to express this position. To labour the point once more, without the agreement of all participants, the eventual outcome has little value. If you remember, when the histamine was being weighed, I made the suggestion that we might take 500 mgs into 50 mls of water, instead of 50 mgs into 5 mls. As I recall, you agreed, at the time, that this was unnecessary.
>The eventual dilutions to 18C take the diminishing concentration of histamine well past Avogadro’s constant. Even if there is an error of 20% in the initial 1C dilution, the critical part of the serial dilution process is the accuracy and precision of that repeated dilution. In this way the error at 18C will still be 20%. Any error is maintained, not amplified.
>A couple of final thoughts. The restriction of my daily workload does not afford me the luxury of checking my e-mail several times a day. I log on when I can, and I respond accordingly. I am more than willing to discuss any and every issue that might be raised, and I am certainly not avoiding any contact when criticism is made and then portrayed as inconvenient to my apparent point of view. I am however extremely unhappy at the portrayal of myself, my dept and my colleagues within the division as sloppy and cavalier, and naturally I strenuously disagree. The most disappointing element is the implication that such alleged sloppiness must, of course, be endemic throughout the rest of the lab work. I find that attitude neither helpful nor acceptable.
The degranulation protocol that we use was never portrayed as a replication of Dr Ennis’s methodology. The differences between the 2 that have been referred to elsewhere in correspondence, I would consider to be minor. I reach this conclusion because the results that we obtained were validated by several positive and negative controls. It is very basic lab practice when attempting to evaluate samples of unknown activity to include known controls. Only if these controls perform as they are expected to, can you then reliably assign a value to an unknown sample. Whatever the differences in methodology, they are not significant enough to have affected the performance of these controls. I actually had to abandon 2 runs of samples because the positive control sample was unacceptably low.
I did not get involved in this project because I thought it would be a laugh, or because it was an opportunity to get on camera. We got involved because we felt that we had the expertise and the attention to detail to contribute to the body of work already in existence. I’m disappointed that we’ve been unable to communicate this point of view in an adequate fashion.
Please feel free to pass on this mail to whomsoever you feel may be interested.
From: Lionel Milgrom
Sent: Saturday, December 13, 2003 4:45 AM
To: Wayne Turnbull
Cc: Francis Treuherz; Dana Ullman, MPH; Harald Walach; [other e-mail addresses removed for publishing]
Subject: Re: ABC Experiment
Many thanks for your e-mail and exegesis. I would just like to raise a few points.
If this project was never discussed as an exercise into the fundamental validity of homeopathy or the memory of water hypothesis (not theory, yet!), what could it have been for, other than a bit of TV entertainment? To call it ‘real science’ is actually a bit of a non-sequatur, especially as it now turns out that there were significant differences in the protocol you used compared to that of Ennis’s. Consequently, whatever the result, one has to question its validity. After 30+ years experience as a lab-based scientist, I think I am capable of understanding the difference between disproving and not proving a hypothesis by now.
You are utterly correct in stating that any scientific enquiry requires the utmost objectivity and impartiality. Does that include inaccurately weighing and standardising the original histamine solution (by the way, there WAS a draft next to the balance – whether it was blowing out or in is immaterial: it led to severe scale-pan vere)? I now understand why this is important and significant. And I did remark at the time, but obviously not strenuously enough, which I accept.
The basic thesis in the memory of water hypothesis is that a material diluted out of existence (and agitated, or succussed, at each dilution step) can still exhibit effects. If this is true, then it could be that the past dilution history (and therefore the original concentration of the solution) matters: if it is not true, then it would not matter. At the moment, there is no agreed or clear-cut hypothesis to explain how such ultra-highly diluted solutions could exhibit effects, certainly not in any of the conventional molecular or bio-molecular sciences. However, I would refer you to the work of Samal and Geckler (Chemical Communications, 2001, page 2224; there is no volume number) which shows quite categorically that past dilution history of solutions admittedly, not past Avogadro’s limit) affects the outcomes of measurements made upon them. I would also refer you to a much earlier paper by Del Guidice, Preparata, and Vitello (Physical Review Letters, 1988, vol 61, pp1085-1088), where a physical mechanism for a water memory effect is postulated. The point is that both of these journals are front-rank peer-reviewed journals in their fields (chemistry and physics, respectively) with high impact factors. It is therefore incumbent upon us to make sure that we really do ‘real science’ on this topic, for its own sake and not because some TV company thinks it might be a good wheeze.
Chemistry and biochemistry are based on the notion of palpable molecules interacting with each other. Ergo, remove the molecules and the interactions must cease. The effects, if any, of intervening solvent molecules, especially ones as complex as water en masse, are exceptionally difficult to account for and therefore include in our hypotheses. In biochemistry, in particular, concentration on active bio-molecules (e.g., DNA, proteins, etc) tends to preclude or ignore the effect of the intervening water molecules in which all biochemistry is played out. In any one cell, the actual physical nature of the water inside it (e.g., viscosity, local structure, etc) varies from compartment to compartment. Consequently, any attempt to even simply model the intimate molecular effects of intra- and extra-cellular water is a nightmare. This is only to highlight the difficulty of seriously constructing and testing a working hypothesis for a water memory effect. But let there be no doubt that not only is water far stranger than we think, it is also much more difficult to explain than we think.
If dilution history does not matter, then it cannot matter whether 40, 50, or 60 mgs of histamine were originally weighed out, for eventually, there will be no molecules of histamine left in solution. And the reason why I mention all of this is that it appears to me that by NOT weighing and standardising accurately, whether you know it or not, you have already pre-judged the issue. This means objectivity and impartiality are compromised and therefore one has to conclude that experiments will suffer accordingly.
I find it hard to believe that even with all the resiting, etc going on in your establishment, a four-figure balance could not have been found. I too work in a highly respected academic environment where such inconveniences occur on a daily basis, and this is not a problem for the promulagation of serious science. I would also reiterate that accurate weighing cannot by its very nature be done on an open rough balance: this is something which is taught at first-year undergraduate level. Also, the correct technique for making a standardised solution involves carefully transferring the weighed material (dry) to a volumetric flask, ensuring its dissolution and even distribution in water, and then making up to the mark. This does NOT involve at any stage, or for any reason, addition of water to the weighing vehicle on the scale pan. Only then can you be absolutely certain that ALL the material weighed out has been properly transferred and used in making up the solution. Consequently, I would submit that on top of the original 20-25% inaccuracy in the weighing, there is a further unknown inaccuracy in making up the original solution. I am sorry to labour this point….actually, I am NOT sorry to labour this point. It just IS sloppy, bad, technique and should not be allowed in ANY laboratory regardless of its academic status. However, I would agree that the further dilutions were done properly so that further error after weighing and solution manufacture probably did not occur.
My understanding was that the ABC ‘experiment’ was going to be a reproduction of the BBC experiment, which was supposed to be an attempt to reproduce Ennis’s work. So, I’m afraid I do not understand what you mean by saying that you were not trying to follow Prof Ennis’s protocol. As this was the one that claimed success for the effects of ultra-highly diluted solutions of histamine on basophil degranulation, don’t you think it would have been a good idea to follow it to the letter? And if there are changes made to the protocol which you consider to be only minor, does this not mean that once again, you have already pre-judged the issue and therefore there has to be a question mark over your objectivity and impartiality? I would be fascinated to hear what Prof Ennis makes of this.
Having said this, I am sure I speak for all of us involved in this exercise that we appreciated the time and effort you put into this and your accommodation and inclusivity. On your suggestion, I have circulated your reply to my first e-mail, and I shall be doing the same with this one.
With Best Wishes