In 1934, 1935, and 1938, he says these teeth are "definitely not normal” but indicates that there may be doubt as to the cause. In 1934 he says:

No attempt is made to diagnose these small white spots or minute white fleckings as the earliest signs of mottling by examination of the person per se. Recourse is always made to group study.

This is what he meant in the Delaney hearings, top of page 1649, when he said:

You don't know what it is.

And as to balancing the light 3 or 4 ways with a mirror, in Public Health Reports, September 10, 1937, page 1252, he says:

At Clovis, the diagnosis of the degree of mottled enamel severity was based on a careful visual examination without the aid of mouth mirrors. *** The signs of chronic endemic dental fluorosis are so objective that little variation in incidence is noted using either one or the other of these methods.

So much for the category of "questionable". "Very mild" is the next category and, as we have seen, sometimes shows brown stain. In Dean 1934, 1935, and 1938, we are told that it is

principally observed on the labial and buccal (i. e. toward the lip or cheek) surfaces; that it involves up to 25 percent of the affected teeth.

And in all three descriptions of "mild" fluorosis we are told:

The white opaque areas * * * involve at least half the tooth surface— and

light brown stains are sometimes apparent, generally on the upper incisors.

In Dean 1942, he describes his new "weighted average" method for calculating a "community index of fluorisis" and gives, on page 26, a greatly abridged description of the categories. It omits all mention of brown stain in the "very mild" and "mild" categories, although in all the seven sets of illustrations brown stain is shown on the upper front teeth of "mild" fluorosis. And, whereas Dean 1934, 1935, and 1938 had said that in mild fluorosis "at least half" of the tooth surface was involved, in 1942 he says that

the white opaque areas * * * do not involve as much as 50 percent of the tooth. Under "very mild" in the 1942 statement he says:

Frequently included in this category are teeth showing no more than about 2 millimeter of white opacity at the tip of the summit of the cusps of the bicuspids or second molars.

And at the Delaney hearings he stated that such was usually the case. However, in the earlier classifications such teeth were called "questionable" unless the areas were also pitted, or larger than 23 millimeter. In a footnote, he explains that

In our earlier studies such teeth were commonly classified as "questionable”; during recent years, however, they have been invariably listed as "very mild." He thus implies, without saying, that the 1942 classification was used in studies II and V.

But as we have seen on page 11, supra, the original reference is to Dean 1938. There is much supporting evidence, both statistical and documentary, of the use of the 1935-38 classification. For example, in Public Health Reports, February 10, 1939 (the same year studies II

and V began) a footnote referring to Dean 1935 (identical with Dean 1938) reads:

This classification of diagnosis has since been abridged by combining "moderately severe" and "severe" into one classification: "severe"-HTD.

I can find no other mention of revision until 1942.

Furthermore, Dean testified (Delaney hearings, pp. 1647-1648) that the recommended figure of 1.0 parts per million was based on "the study of the 21 cities" and was the "result of plotting a curve on the 21 cities." (See fig. 1, appended.) And throughout his testimony kept basing his answers on evidence from "the 21 cities." This is the study by Dean and others (Public Health Report, Apr. 11, 1941, and Aug. 7, 1942) to which I have referred as studies II and V.

THE TRUTH ABOUT THE 21 CITIES

But elsewhere in his testimony (pp. 1652 and 1653) Dean stated that data from Maywood and from Marion were without significance because of changes in the water supply during the lifetime of the children examined. That reduces the 21 to 19.

Repeatedly, and over many years, Dean has pointed out that no conclusions can be drawn from the quantitative studies made where there have been changes in the source or composition of the water during the lifetime of the group examined. He said it in Dean 1938, page 405. He said it in study II. He said it in Public Health Reports, December 6, 1935, on page 1720. And he said it in the American Journal of Public Health, June 1936, page 569.

In the last article cited, on page 573, he lists Galesburg as lacking the "requisites for quantitative evaluation" because of changes in the water up to 1928. Nevertheless, he used Galesburg as one of the key cities in his case for fluoridation. The statement that protection against decay exists even in the absence of visible mottling is based on Galesburg (and rests on a fallacious argument at that). The promised two-thirds reductions in tooth decay is based on the Galesburg-Quincy study. And Galesburg was used as one of the "21 cities." So now we have 18.

But that is not all. If we study the water histories of the 21 cities, as given in studies II and V, we find that not only in Galesburg, Maywood, and Marion, but also in Elmhurst, Aurora, Joliet, Elgin, and by specific statement, East Moline, there were changes which make them completely unreliable as evidence. Now we are down to 13, and there is a real question about Lima.

So, actually the 21 cities of which Dean testified boil down to Colorado Springs with 2.6 parts per million, Kewanee with 0.9, Pueblo with 0.6, and 9 cities with 0.2 or less parts per million of fluoride.

These facts were published in Dean's original articles, and it is reasonable to assume that he knew them then. They completely invalidate all conclusions based on the study, which means the entire case for fluoridation. He knew this in 1952, as his testimony proves, and we may assume he knew it all these years. And if the articles were ghost-written and he actually didn't know what was in them at the time, he should have told the world when he found out. Instead, he testified under oath about 21 cities.

THE STRANGE CASE OF AURORA

Aurora is the second key city in the case for fluoridation. It is the one used to prove that front teeth are not affected. It was also one of the cities selected by McClure for his strange studies on fluoride retention and on fluoride effects on bones and growth. (He also used Galesburg, Joliet, and Elgin, as well as Amarillo and Lubbock, Tex., both listed by Dean in 1936 as lacking the requisites for quantitative evaluation.)

It is also the city used as a basis of comparison with Grand Rapids in the Public Health Service experiment. Fluoride was added at Grand Rapids (1945), Aurora was the standard of expected results, and Muskegon was left without fluoride as control.

The water history of Aurora is therefore important. The children studied (studies II and V) were 12 to 14 in the fall of 1939. They were born in 1925, 1926, and 1927.

Up to 1940, Aurora used water from 13 wells (according to study II). The amounts contributed by the different wells was quite variable.

1 Well No. 5 was out of use and no water pumped from it during 1935-39 (and for an unstated prior period). It was repaired and returned to use in 1940. The fluoride content after repair is not given.

2 A mixture of these waters in unknown proportion in 1939 showed 1.2 parts per million. Between 1935 and 1939, these 3 new wells supplied from 40 to 70 percent of the water.

It is clear that the water used by these children while their front teeth were forming (up to about age 3 or 4) was not the same as that used while the bicuspids and molars were being formed later. It is also clear, from the lack of fluorosis in the front teeth that the earlier amount was significantly lower. (This is the only known cause for this finding and has been frequently observed.) In spite of all this, 15 percent of the children had definitely mottled teeth, and 32 percent more had questionable mottling.

Nevertheless, on the same page on which the water history is given, Dean says:

From the standpoint of a population exposed for a long period of time to a water supply containing small amounts of fluorides, Aurora appears to offer many advantages for epidemiological study. Since 1898 the public water supply has been obtained from wells into the Cambrian Potsdam sandstone. (Italics mine.)

He fails to mention that they passed through and received varying amounts of water from the water-bearing Niagaran limestone, and the St. Peter sandstone. From then on he settled for telling people that Aurora has used the same type of water supply for over 50 years.

I believe it can be shown that, without exception, the witnesses for fluoridation testified falsely, but in the other cases it is harder to show that they did it knowingly. In general they were merely parroting in irresponsible fashion, things they had been told. My notes indicate that some of them have perjured themselves elsewhere, but I have not yet had an opportunity to examine the actual record of their testimony.

FAULTY METHODS USED

One of the things they teach you in grade-school arithmetic is that you can't add dissimilar things. This is not because of any narrowminded rules, but simply because the number you get doesn't mean anything. The 5 you get by adding 1 boy and 4 bicycles has nothing to do with the 5 you get by adding 4 boys and 1 bicycle, and neither is 5 of anything.

You can't even add different sized units of the same thing. One foot and four inches isn't five of anything. It is either 16 inches, or 111⁄2 feet.

These things don't work in arithmetic or anywhere else in mathematics. Neither do they work in physics, in chemistry, in engineering, or anywhere else. Any time you do them, you are in trouble.

Yet, when the men who should know better add up decayed, filled, and missing teeth, getting something they call a DMF rate; and when they tell us things based on comparison of such rates; we bow in reverence and do whatever they tell us to do even if it kills us.

UNRELIABLE EXAMINATIONS

But that is not all. You can't even tell how many teeth are decayed. In the Journal of the American Dental Association, December 1941, pages 1959-1961, Radusch (chairman of the Minneapolis caries study group), reports a comparison of work done by 8 dentists examining 33 patients. Each patient was examined by 3 (in 1 case only 2) examiners who could take as much time as they liked and knew their findings would be checked by the others. The reported findings show a standard deviation of from 40 percent above to 40 percent below the average number of cavities found.

The same sort of thing occurs when the same man repeats his own examinations. Dirks, Amerongen, and Winkler (Journal of Dental Research, June 1951, pp. 346-359), found that by cleaning and drying the teeth, using rubber dam, separators, and spotlights, spending 4 to 5 hours on each examination, with 2 examiners each making 2 examinations and comparing results, they would get fairly reproducible findings. They discarded the method as too cumbersome and still too inaccurate to be reliable.

X-RAY ALSO UNRELIABLE

When you use X-ray, you are no better off. It is more reliable for cavities between the teeth, but less reliable for cavities on the exposed surfaces. In the Journal of the American Dental Association, August 1934, Ennis tells us, on page 1373, that on examining 883 students he found 1,662 decayed teeth clinically as against 1,372 found by X-ray but that the examinations agreed in only 237 cases. Decay

was found clinically but not by X-ray in 1,425 teeth, while it was found by X-ray but not clinically in 1,135.

And to cap it all, Durkett (Journal of Dental Research, vol. 20, pp. 70-76, 1941), found that 54 percent of cavities found microscopically were found by neither X-ray nor clinical examination.

EXPERIMENTS MEANINGLESS

These things are not too serious from a personal standpoint. If a cavity is of any consequence, it will get big enough to find. From an experimental standpoint, however, they are devastating. Neither individual nor statistical comparisons can have any meaning. You can compare two mouths and be very sure that one is better than another, but you can't attach any number with any possible meaning to the differences.

For these reasons, the experiments at Newburgh, Grand Rapids, and elsewhere can never prove anything if they go on a hundred years (except probably that they have produced mottled teeth, and possibly that they have caused more or less serious other damage, and when these things become known, it will be far too late to undo the damage).

And there are other reasons why most of these experiments will never prove anything. They were not designed to ascertain facts, but to prove a point. Blayney, at Evanton, has used faulty methods, but has made a serious and sincere attempt to run a respectable scientific experiment. He, however, is frowned upon in public health circles for being honest.

All this boils down to the plain fact that there is no possible basis for the promise of 65 percent reduction in tooth decay. Neither the experimental nor the "epidemiological" evidence proves anything. And you couldn't even know if it occurred. It is our fantasy, and fraudulent fantasy at that.

It is true that there are places where there is more fluoride and less tooth decay than in others. There are also places where the reverse is true. We aren't told about the latter. Figure 2, appended, shows examples.

Figure 3 shows the lack of any reliable relationship between the amount of fluoride in the water and the damage done to teeth. The figures are accepted at face value from the literature. If anyone wants to prove that they are unreliable, it's all right with me. Almost all of them are quoted from Dean. Ockerse found (Dental Caries, Department of Health, Pretoria, South Africa, 1949, p. 51) that

it was not possible to correlate the degree of mottling with the amount of fluroine in the drinking water.

He attributed this to variations in the amounts present in the same supply. As we have seen, there are other important reasons.

OTHER FLUORIDE DAMAGE

We have seen that McClure claims to have proved that fluoride cannot accumulate in the body because it is almost all put out in the urine. We know his proofs are false. But in any case, "the proof of the pudding is in the eating."