If you have a representative system model, then indicators, approximations, and estimators can be useful in determining the state of the whole system. With a proven system model it is possible to determine the validity and accuracy of available estimation tools.

My read of the medical/dental research literature relating health consequences to mercury exposure from mercury amalgam fillings leaves me quite puzzled. There is no scientifically established human system model for chronic mercury exposure from amalgams.

Researchers commonly use indicators that have no proven system accuracy to proclaim far reaching conclusions about the amount of mercury exposure. From these unsupportable conclusions, they then extend to claim no negative health consequences result.

Some of the most fundamental scientific concepts -- rates in particular, at times seem to be completely overlooked.

Chronic Mercury and Chronic Rain, Parallel Systems

Let's examine rain as a parallel example to mercury vapor from amalgams, to understand the system model view point.

You have no idea what the previous rainfall has been in the last three years, three months, three days, or three hours. Nonetheless, you are tasked with determining how much rain has soaked into a 100 sq. inches of ground, and the resulting amount of "wetness".

You are not allowed to dig up the earth to measure the wetness within it; that would be the equivalent of killing and dissecting the patient.

The rain appears to be constantly falling. Every time you look out it is wet. Sometimes it looks wetter than at other times. You decide to use an estimator to determine just how much rain has soaked in, the level of "wetness". By measuring the volume of water currently on the surface of your 100 sq. inches, you will predict the amount of wetness within.

You get out your wet and dry vacuum and suck up all the water you can from within the 100 sq. inches. When you go back in and measure the water you've collected, you find very little. In fact, it hardly seems moist at all based on the sampling with your vacuum. You had a very difficult time even measuring the water. Now comes the truly amazing part, your conclusion.

Based on the amount of water you were able to vacuum, you conclude that it is hardly raining at all. Since it is hardly raining at all, you therefore must conclude that the ground has not soaked up much water. With the enormous mass of the earth beneath your 100 sq. inches of surface, there is just not enough water out there to make any serious amount of wetness. Rain in this case does not create significant "wetness" within the ground.

What is wrong with your approach to this problem of measuring rain ?

Before I tell you my answer, let's note the similarity to one of the more popular measurement techniques for determining mercury exposure from amalgams. The following method is sometimes used to predict mercury accumulation within a human body.

The researcher out to determine just how little mercury escapes from amalgam fillings decides to measure the amount of vapor in 100 people's mouths, to determine their body burdens and exposure levels.

The researcher gets out his Jerome mercury vapor analyzer and sticks it into the mouths. Every mouth containing amalgams has mercury vapor, this is confirmed. People who chew gum seem to have even more vapor. The absolute amount of vapor is minute in every measurement.

The researcher compares the measured value of mercury vapor to what has been established as a safe level in the atmosphere of chlora-alkali plants and says: "these mouths all meet the safe standards for occupational mercury exposure". "There's hardly any mercury escaping at all". "What is coming out is so small an amount, that no human with all that mass in their body could ever accumulate very much." "Certainly there is not enough mercury here to cause chronic illness."

Do you see the problem yet ? My rain example and the way people are testing for mercury vapor by sampling mouths are exactly the same. It's the rates! Or rather the ignoring of the need to measure rates. We've measured no rates whatsoever. Yet there are multiple rates to be concerned about.

The issue of mercury toxicity from continuous mercury release from dental amalgams is fundamentally about mismatched rates. Lets go through the rain and the mercury vapor problem together to examine this point further.

The vapor analyzer and the wet and dry vacuum give a sample moment. How much was there the moment before ? How much is there the next moment ? Importantly, is the mercury seen at each moment the same mercury, or different mercury ?

By examining the amount of mercury vapor in the mouth at a moment, just like surface water on the ground at a moment, you simply do not have enough information to reach the conclusions stated.

How fast is the mercury releasing ? How fast is the rain coming down ? To properly measure rain I would need a bucket to hold all the water as it came down over time. For mercury we need some way to evacuate and store all the mercury as it is released over time.

Such methods give the ability to measure the input rate and the absolute amounts of new water and new mercury within a time period, and not just the element present at a moment.

So if we established a rate of mercury release we could correlate the measured moment values. We'd then have a good system model with a reliable estimator for exposure, right ? Wrong! We still know nothing about the rate at which the body is absorbing and transporting the mercury. Lets look at rain again.

If the ground is very dry and very porous, little rain will be detectable at the surface by vacuuming. Even if there is a heavy rain coming down it will appear to be little by our method. The converse would be true also. If the ground was heavy clay and fully saturated with water, then even a little rain would puddle up noticeably.

If the absorption and transport rate of mercury is faster than the release rate, then the vapor analysis within the mouth will always underestimate the true release rate. Compounding the problem further is the fact that people are biochemically different. Case studies of industrial mercury exposure show wide variances in mercury absorption among individuals facing the same exposure levels.

None of the vapor tests I'm referring to have expressed a human absorption rate for mercury. Studies have been done to show that 80% of elemental mercury vapor breathed into the lungs is absorbed. It is known that mercury vapor can pass directly into the tissues of the mouth. Even when a rate is estimated, there is still going to be differences from person to person.

It is believed that habitual alcohol consumption is one factor that decreases mercury absorption. It has also been shown that the chlorine in the air of chlorine-alkali plants preferentially binds with elemental mercury to produce a less absorbable, non-gas, less toxic mercury compound, namely calomel. There are going to be other factors that enhance and diminish the rate of mercury absorption as well.

Finally, we must depart from the rain analogy when considering mercury exposure from dental amalgam. Mercury vapor is but one of three major forms of release given off by mercury amalgam dental fillings.

In my view, the mercury vapor is a less significant exposure risk. Mercury fillings give off charged atoms directly into saliva. These atoms may be bound to human protein directly, or connect with an electrolytic compound. Such releases to the oral cavity face the same fate as abraded particles -- a trip down the GI..

Abraded particles of amalgam work their way down the GI track where they face further dissolution by the hydrochloric acid within your stomach. From there all the mercury ions pass into the intestines where they are 20% absorbed directly. Once in the intestines the mercury is ingested by Candida yeast and other GI flora. Candida yeast in particular, preferentially contaminate immune system Nuetrophil that defend the cell wall of the GI by consuming the yeast, mercury and all.

Candida yeast absorb all the mercury they can bear. Once loaded, Candida transforms the mercury into the severe poison methyl mercury. The yeast excretes the methyl mercury, and the human GI absorbs it nearly 100%.

Rates Model for Mercury Exposure, Absorption, and Subsequent Toxicity Risk

My central point is that the true model of amalgam mercury exposure, and risk for toxicity is logically based on a system of rates. If actual values for these rates are known for an individual, then a representative model, and predictions of system outcome from a given mercury exposure become possible.

Until such a model is developed and used to validate experimental observations, all conclusions about systemic effects are nothing more than speculations based on incomplete information.

• Mercury releases from dental amalgam in 3 major forms: vapor, charged atoms, and particles. These rates are described mathematically in the form mass/time. Let's collectively call them rate R.
• Methylization by Candida yeasts, other GI flora, and damaged GI cell linings all facilitating uptake and absorption rates. Lets collectively call these rates I.
• Disruptive bonding within the body interferes with enzymes controlling energy and other processes such as nerve function. Again mass/time describes the rates. Let's collectively call these rates B.
• Rate of collection by scavenger molecules such as glutathione; and excretion through harmless bonding, feces, urine, sweat, respiration, hair and nails. Again mass/time describes the rates. Lets collectively call these rates E.

Here is a mathematical description using inequalities for system relationships where the relative rates determine whether dental amalgam mercury exposure is likely to make a person ill or not.

Sick or Suffering People:

• R >= I >= B > E;
• R >= I <= B > E;
• R <= I >= B > E;
• R <= I <= B > E; (the worst case of all, these people are the most susceptible to mercury toxicity)

In all of these cases mercury enters the human system and is chemically bound up disruptively faster than it can be excreted. The net is toxicity at some level. The magnitudes of the different rates will determine just how sick the person may become.

As long as the excretion rates are at a disadvantage to the accumulation and binding rates, further disruptive accumulation and sickness will continue over time.

Healthy or Coping People:

• R >= I >= B < E; (the best case, with the greatest resistance to mercury toxicity)
• R >= I <= B < E;
• R <= I >= B < E;
• R <= I <= B < E.

In all of these cases the body is the winner because it can excrete faster than the absorption and binding rates. Some of these inequalities will be more taxing on the body than others. It is always better to not absorb mercury in the first place, than to be the worlds best excretor.

The surest way for everyone to be healthy is to eliminate R all together. Your body would then not need to devote any resources to excreting the mercury toxin.

The extremes of the rate model would predict that for any given rate of exposure R, the person who has a very low rate of input I, and a very low rate of binding B, and an extremely high rate of excretion E, is always going to with stand the poison the best.

The person most susceptible to mercury poisoning will have the highest rate of input I, the highest rate of binding B, and the lowest rate of excretion E.

Below is the data presented by the U.S. dental industry apologists, and my critical review.

Review of the Data From UCSF Dental School

"Gamma-1 to beta-1 phase transformation in retrieved clinical amalgam restorations"; SJ. Marshall, GW. Marshall, Jr., H. Letzel; Dental Materials 8:162-166, May 1992.

"Mercury content of amalgam restorations"; SJ. Marshall, GW. Marshall, Jr., H. Letzel; General Dentistry; Nov-Dec, 1989; 473-7.

The tables published in their research reports did not include the "Initial Hg" and "Missing Hg" columns, only the "Residual Hg" column is theirs.

Initial Hg is the amount the manufacturer specified or included with the amalgam supplies. Missing Hg is calculated by subtracting their residual numbers from the manufacturers specified initial mix. I don't have initial values for all their samples just yet. I will fill in the table as I get that information.

Important: The convention with amalgam researchers is to normalize all content percentages against a 100% value for the whole alloy, including all the constituents, and not just one ingredient.

Unaware of this convention, I originally interpreted the data below to be normalized to a 100% value for mercury only. As you can see by the numbers, that mistake would lead one to think more than half the mercury was missing on a consistent basis.

Low Copper Amalgam - SJ. Marshall et. al.

Please note the high variance in residual mercury between samples of the same brand:

Ageston shows a 5.2% variance normalized to the amalgam. ?% normalized to initial Hg.
Cavex-SF shows a 5.1% variance normalized to the amalgam, ?% normalized to initial Hg.
New True Dentalloy shows a 8% variance normalized to the amalgam, ?% normalized to initial Hg.
Shofu Spherical shows a 4.5% variance normalized to the amalgam, ?% normalized to initial Hg.
Standalloy F shows a 9.9% variance normalized to the amalgam, ?% normalized to initial Hg.

The researchers explain away the maximum amounts they can of the variation based on the handling of the dentist installing the amalgam, and the batch of metal powder alloyed in the amalgam.

From the data and explanations they gave, there is little doubt that some of the missing Hg and variances in residual Hg can be explained away. I'm not willing to accept their partisan approach to minimizing the amount that escaped into the humans. There is no reason to believe amalgam dissolution and mercury release rates are uniform from filling to filling, mouth to mouth. Some people are getting it much worse.

I find it heartening for the truth that they would even admit some mercury was in fact missing, most likely due to the experience "in vivo".

Since Candida yeast collect mercury, selectively feed it to immune cells, and magnify the toxicity by converting it to methyl mercury, releasing it into the gut, any released mercury is too much.

High Copper Blends of Amalgam - SJ. Marshall et. al.

Please note the high residual mercury variance between samples of the same brand:

A76 Degussa shows a 3.7% variance normalized to the amalgam. ?% normalized to initial Hg.
Dispersalloy shows a 4.2% variance normalized to the amalgam, 8.4% normalized to initial Hg.
Luxalloy shows a 8.4% variance normalized to the amalgam, 15.4% normalized to initial Hg.

Abandoning the conventions of the amalgam researchers, lets discuss the amount of mercury missing as a percentage of all the mercury installed with the filling, 100%. For two of these brands I have specifications for the initial amounts. Below is the maximum amounts of mercury missing from the presented samples.

Dispersalloy shows a worst case sample with 15.8% missing from the initial 100% of Hg installed in the filling.

Luxalloy shows a worst case sample with 17.8% missing from the initial 100% of Hg installed in the filling.

I'm willing to allow some of this variation to be accounted for by batch variation and differences in operator technique in condensing the raw mix. My talk with a metallurgist experienced in amalgam research leads me to allow 5% of the mercury as a reasonable maximum for these factors.

The 10.8% and 12.8% still missing, need another explanation. Evaporation and runoff during setting; solid state conversion, corrosion, and electro-chemical dissolution -- all release mercury into your mouth. These are the other explanations for the missing mercury.

None of this information accounts for abraded particles which reduced the initial mass of the filling, leaving no clues behind, corrosion byproduct or any other estimator.

High Copper Single Particle Amalgam - SJ. Marshall et. al.

Please note the high variance in residual mercury between samples of the same brand:

Indiloy shows a 2.5% variance normalized to the amalgam. 7.1% normalized to initial Hg.

Sybraloy shows a 4.0% variance normalized to the amalgam, 9.0% normalized to initial Hg.

Tytin shows a 1.7% variance normalized to the amalgam, 13.5% normalized to initial Hg.

Abandoning the conventions of the amalgam researchers lets discuss the amount of mercury missing as a percentage of all the mercury installed with the filling, 100%. For each of these brands I have specifications for the initial amounts. Below is the maximum amounts of mercury missing from the presented samples.

Indiloy shows a worst case sample with 9.0% missing from the initial 100% of Hg installed in the filling.

Sybraloy shows a worst case sample with 14.4% missing from the initial 100% of Hg installed in the filling.

Tytin shows a worst case sample with 16.9% missing from the initial 100% of Hg installed in the filling.

Again, I'm willing to allow some of this variation to be accounted for by batch variation and differences in operator technique. A 5% placement variation of residual mercury is a reasonable maximum for these factors.

The 4%, 9.4%, and 11.9% of initial mercury still missing, need another explanation. Evaporation and runoff during setting; solid state conversion, corrosion, and electro-chemical dissolution -- all release mercury into your mouth. These are the other explanations for the missing mercury.

Still, none of this information accounts for abraded particles which reduce the initial mass of the filling, leaving no clues behind, corrosion byproducts or other estimators.

Absolute Values of Missing Mercury

Now lets talk about the maximum amounts of mercury your body is exposed to from a dental filling. There are so many assumptions, presumptions, opinions, points of view, and vested interests, I will simply present a table of possible values.

You choose what you believe is the right amount for you. Dr. SJ Marshall assumed the average filling weighed 0.4 grams each. I'm willing to do that and assume the average filling started with 50% mercury.

Below is the table of possible exposure rates given the number of teeth with average amalgams, and your favorite average loss of mercury.

Oral galvanism, when it includes mixed metals, will push the percentages up. Abrasion and wearing of the actual filling are not accounted for in any of the missing mercury numbers presented on this page from Dr. Marshall's paper.

According to a Danish report concerned with mercury pollution in the environment, the average amount of mercury released at cremation is 1.8 to 3.8 grams per cadaver. We humans carry quite a bit of mercury in our teeth, all the way to the grave it seems.

For some, the mercury they carry to the grave may in fact be what carried them to the grave.

Conclusions

1. By using engineering principles to model the complex human system of dental mercury that is release into the mouth: methylization, human absorption, harmful binding, and excretion; it is easy to see major fallacies in the way estimates of potential mercury poisoning are made in some of the published research literature.

2. Scientific principles obvious to most technical people are not showing up in these studies of mercury exposure from dental amalgam. We entrust our good health and desire for long life to the decisions based on the results of these flawed studies.

3. There is compelling evidence from multiple independent researchers that mercury amalgam fillings release mercury into the body on a continuous basis.

4. The proper unit of measurement for the mercury released from an average amalgam over time is milligrams (mg).

5. Case histories of known mercury exposure, and the rates model offered in this essay, both indicate that there will be a wide variance in health consequences between individuals, when confronted with the same rate of mercury exposure.

6. Candida albicans is scientifically proven to transform elemental mercury into the far more poisonous methyl mercury. Many people with CFS and other chronic illnesses have documented Candida yeast overgrowth in the intestines.

7. Given the clear hazard of even small amounts of mercury, it is unreasonable to continue using mercury alloys for dental restorations.

8. People with CFS, bearing mercury amalgam fillings, and experiencing Candida overgrowth in the GI track need to take heed of the knowledge conveyed here. The facts and reasoning presented are compelling that mercury amalgam fillings are the suspected root of your puzzling health problems.