I made up the waters and supplied them in large bottles to Dr. Bittner in such manner as not to disclose to him the fluoride content of the waters. Dr. Bittner conducted the animal work and examined the mice twice weekly. He recorded the date on which each individual animal developed a diagnosable breast tumor.

Since Dr. Bittner did not know the fluoride content of the waters or even which waters contained fluoride, the effect of possible unconscious bias on the results was eliminated. At the termination of the experiment the data was tabulated by Dr. Bittner who was still uninformed as the amounts of fluoride in the waters.

Dr. Alan E. Treloar, professor of biostatistics in the Medical School of the University of Minnesota, then analyzed the data by statistical methods in order to provide tests of significance between the results obtained with the animals in the three groups. The results of the eperiments are summarized in table 1 so as to show the mean age of the animals at which cancer was diagnosed.

TABLE 1.-Comparison of ages at which mice receiving distilled water or highly fluoridated water as drinking water developed breast cancer

Group C, in which the mean cancer age was 398.3 days, can be considered to be a control group since the mice in this group received distilled water. The animals of group A received water containing 5 parts per million of fluoride and developed cancer at a mean age of 360.4 days. This result is not statistically different from the mean of group C and would be expected to occur 10 times in 100 by chance. The lack of significance between the results obtained with groups A and C is further demonstrated by the results obtained with group B. The animals of group B received water containing 10 parts per million of fluoride, a concentration which was twice that of the water drunk by the animals of group A. The mean cancer age of the animals of group B was 398.6 days, à result which is obviously equal to that obtained with the animals of group C, namely 398.3 days.

Dr. Treloar further writes:

The differences between the means of groups A, B, and C have been subjected to critical statistical examination in the form of an analysis of variance. The ratio of "between means" variance to "within classes" variance, the F value, is 1.88. This falls so far short of the minimum value of 3.09 required by accepted standards for any significance to be assigned, that the differences must properly be regarded merely as chance fluctuations.

In summary, our experiments do not confirm the hypothesis that fluoridated drinking water accelerates the development of breast cancer in mice.

EXPERIMENTS WITH THE HUMAN DEMONSTRATE SAFETY OF FLUORIDATION

Laboratory studies are important in considerations of the safety of fluoridated waters. However, the experiments of prime significance for the human are the experiments of nature. These experiments have been conducted by nature with human subjects over many years in numerous communities in the United States in which high amounts of fluoride occur in the water. In many of these communities the water contains much more fluoride than that which is needed for dental-health reasons.

We have only to observe the results of nature's experiments in order to gain conclusive evidence of the safety of fluoridation. Undoubtedly other persons have or will testify in detail on this point. I will only summarize the observations which have shown (1) the crude death rates from all causes and the specific death rates due to cancer, heart disease, nephritis, complications of pregnancy, and infectious diseases are not different in the communities with high and low fluoride waters, (2) stature and weight and history of bone fractures are not related to the fluoride content of the community water, and (3) no evidence of increased incidence of skeletal osteosclerosis has been found in inhabitants of communities with waters of a high fluoride content.

FLUORIDE NOT NEW IN HUMAN METABOLISM

Intake of fluoride in food and water is not a new metabolic experience for the human and undoubtedly has occurred throughout the whole of the evolutionary development of man. The evidence for this conclusion derives from the wide distribution of fluoride in foods and waters. Practically all soil waters contains some fluoride and many contain large quantities of this ion.

I have never found a single natural article of food to contain no fluoride. Even the highly purified food table sugar (sucrose) according to my analyses contains detectable fluoride (5 x 10-6%). My own studies indicate a daily intake of 0.19 to 0.32 milligrams of fluoride from the food of an average adult human dietary.

Further evidence for the normal and common intake of fluoride by the human is obtained from the fact that in my experience I have never found a human tooth or bone (animal or human) to be devoid of fluoride. These considerations indicate that fluoridation of water which, as you know, is actually a process for adjustment of the fluoride content of water, would not introduce into the body a substance for which the body lacks metabolic experience.

CONTROLLED FLUORIDATION THE EQUIVALENT OF NATURAL FLUORIDATION

There are no reasonable grounds to doubt that fluoridation with a properly chosen fluoride salt (or compound) gives the equivalent of natural fluoride in water. In either case all evidence points to the existence of fluoride ions from whatever source is to deny one of the most fundamental premises of chemistry which requires the uniformity of identity and properties of an element.

Evidence of the equivalence of fluoride in water from natural and so-called artificial sources has been obtained from their equal behaviors

in analytical chemical reactions and from laboratory experiments with animals. Furthermore, the reduction in dental caries attack rates produced by artificial fluoridation programs in Grand Rapids, Evanston, Newburgh, and so forth, demonstrates the physiological identity of fluoride added at the waterworks to that dissolved into waters from soils and rocks.

TOXICITY IS NOT QUALITATIVE

Misunderstandings have arisen in discussions of fluoridation of water owing to the fact that very large amounts of fluoride are toxic. When sodium fluoride is employed as a rat or cockroach poison, the dry and nearly pure compound is used. The aim is to produce acute lethal fluoride intoxciation in the vermin, a circumstance which would require an impossibly large intake of fluoridated water. In water fluoridation the fluoride compound is highly diluted to one part of fluoride in a million parts of water.

It is not possible to state categorically that any substance is a poison without defining the amounts of the substance or the conditions of its administration. And this applies to fluoride also. Numerous substances, including many required by the body, are toxic when taken in large amounts. Dr. Doty has already referred to vitamin A. For example, the essential vitamin D produces toxic effects in human when given in careless overdosages. One further example, air contains 21 percent oxygen and we breathe air in order to obtain oxygen for essential body needs. If an animal breathes pure oxygen, particularly under pressure, it will suffer ill effects and if the experiment is continued long enough the animal will surely die. Under these conditions oxygen is a poison.

Mr. HALE. Thank you, Dr. Armstrong. Are there any questions? The committee is indebted to you, sir.

Dr. ARMSTRONG. Thank you.

STATEMENT OF DR. FRANCIS F. HEYROTH, M. D., PH. D., ASSOCIATE PROFESSOR OF INDUSTRIAL TOXICOLOGY AND ASSISTANT DIRECTOR OF THE KETTERING LABORATORY, DEPARTMENT OF PREVENTIVE MEDICINE AND INDUSTRIAL HEALTH, COLLEGE OF MEDICINE, UNIVERSITY OF CINCINNATI, CINCINNATI, OHIO Mr. HALE. Dr. Heyroth, the committee will hear you.

Dr. HEYROTH. Mr. Chairman and gentlemen of the committee. My name is Francis F. Heyroth. I hold the degrees of M. D. and Ph. D., both from the University of Cincinnati. I am associate professor of industrial toxicology and assistant director of the Kettering laboratory in the department of preventive medicine and industrial health, college of medicine, University of Cincinnati. I also serve as associate professor of biological chemistry in the college of medicine and as a member of the Cincinnati Board of Health.

The Kettering laboratory studies the dangers to workmen and consumers that attend the manufacture and use of potentially toxic substances. Its work on fluorides began more than 20 years ago shortly after fluorosis became recognized in Denmark as an occupational abnormality, occurring among men engaged in grinding a mineral known as cryolite or sodium aluminum fluoride. In the course of

routine X-ray examinations of these Danish workmen, an increase in the density of the shadows of the bones of many of them was noted. This was of a type not hitherto encountered. It was subsequently established that it was an effect associated with the accumulation of fluoride in the bones.

Our laboratory has made many observations on experimental animals and on human volunteers in the effort to learn the fact of known amounts of fluorides when ingested daily over prolonged periods. We sought to find what part of the dose is excreted in the urine and how much is retained in the body, where it is stored in the bones. Our goal was to determine the maximum amount that a workman might absorb daily into his tissues without storing too much in his bones.

I happen to be a member of the Cincinnati Board of Health. When the Cincinnati Board of Health was asked for an opinion on the advisability of fluoridation, I reviewed the data of our laboratory and the extensive published literature on the toxicity of fluorides at the request of my fellow members. My report to the board was recently reprinted by the British Ministry of Health as an appendix to the report of its mission to investigate fluoridation in North America.

THE NATURE OF H. R. 2341

This bill represents an attempt to initiate Federal legislation on the fluoridation of public water supplies now in progress in hundreds of cities and towns as a public-health measure for lessening the prevalence of dental decay. The House Select Committee to Investigate the Use of Chemicals in Foods, the Delaney committee, devoted 7 days in 1952 to hearings on fluoridation and issued a report on this subject dated July 10, 1952. The opening two sentences of the final paragraph of its conclusions and recommendations read as follows:

The advisability of fluoridating the public water supply of the Nation is essentially a local problem, to be determined for itself by each community. Your committee is not recommending that Federal legislation be enacted in this field.

About 6 months later, H. R. 2341 was introduced and referred to your committee. It reflects the unwillingness of those opposed to fluoridation to abide by the locally arrived at decisions of the many communities that have adopted this public health measure and seeks to negate their decisions by Federal edict. Those promoting this bill may be doing so in the belief that the United States Public Health Sesrvice is seeking to promote the universal fluoridation of the water supplies of the Nation. If so, they are incorrect. Fluoridation is recommended only for those communities whose water supplies contain from natural sources less than the amount that has been shown to be optimal for dental health. In most areas this is 1 part per million; in a few, it is somewhat less. Where grossly excessive amounts occur naturally the Public Health Service has advised and insisted on the installation of equipment for the removal of the excess.

The clause in H. R. 2341 that states that "no agency of any State or of any municipality or other political subdivision of the State shall treat any public water supply with any fluoride compound" seems an attempt to exercise powers that are reserved to the States. If the bill becomes law some of its provisions might create difficulties where

fluoridation is in progress. An interstate common carrier might be enjoined from obtaining its supply of drinking water from a city where the water is fluoridated.

Any justification that the bill may have depends upon the validity of the premise in its opening clause, "To protect the public health from the dangers of the fluorination of water." The evidence that this is false and that fluoridation is safe has been brought together in three publications by me which constitute the essence of my statement and which I have submitted to the clerk for distribution to the committee. And, I might interpolate, in preparing this report I felt deeply the sense of responsibility which had been placed upon me by my official position.

1. My report to the Cincinnati Board of Health, dated January 26, 1951.

2. Reprints of an article "Toxicological Evidence for the Safety of the Fluoridation of Public Water Supplies," published in the American Journal of Public Health, 42, 1568-1952.

3. Reprints of a letter to the editor of Industrial and Engineering Chemistry, 45, 2369-1953-entitled "Effectiveness and Safety of Fluoridation of Public Water Supplies."

These may be summarized briefly as follows: The evidence derives from four sources, (1) animal and (2) human experimentation, (3) epidemiological studies of the health of residents of communities whose water bears excessive amounts of fluorides, and (4) the results of the medical examination of workmen exposed to fluorides in industry. Values for the maximum daily intake of fluoride that may be tolerated by animals of various species have been found from the results of the very great number of experiments performed since 1890. Expressed in milligrams of fluoride per kilogram of body weight, these values are: dairy cattle, 1 to 3; swine, 5 to 12; rats, 10 to 20; guinea pigs, 12 to 20; chickens, 35 to 70. I shall mention only one experiment done in our own laboratory. Each of two 11-months-old littermate dogs was given daily 65 milligrams of fluoride; the third littermate was given none. One dog was given its fluoride as the sodium salt, while the other was given it as cryolite. The administration was continued for 5 years and 5 months. It was stopped because of the death of the dog that had been given none. During life, no osseous changes could be detected roentgenographically in any of the dogs. Storage occurred in the bones, for the ash of those of the dog given sodium fluoride contained 10 times the amount of fluoride found in those of the dog used as control, and given none. No noteworthy histopathologic changes were found in the organs of the fluoride-fed dogs.

The painstaking human balance studies of our laboratory referred to in my introduction offer direct evidence as to the manner in which men and women disposed of the fluoride they took. One of our volunteers took daily 12 milligrams of fluoride as tablets of sodium fluoride over a period of 130 weeks. This daily dose represents the amount present in 12 quarts of fluoridated water with 1 part per million. Careful medical examination made at the end of this period revealed no evidence of any harm. Other persons took lesser doses. In none of our volunteers did any radiographic change occur in the density of their bones, even though one of them had taken 21 grams of sodium fluoride over a little more than 2 years.