Fd Cosmet Toxicol. Vol. 5, pp. 293-308. Pergamon Press 1967. Printed m Great Britain BIBRA Annual Scientific Meeting* Scientific Evidence and Common Sense a s a Basis for F o o d - P a c k a g i n g Regulations J. P. FRAWLEY Hercules Incorporated, Wdmington, Delaware 19899, USA I am honoured by your invitation to meet with you tonight and to discuss with you some of the problems associated with assuring the safety of food-packaging materials. However, I am equally humbled by my own inadequacies to discuss our subject matter as expertly as it deserves. Unfortunately for all of us, there is no individual who can be considered an expert on all aspects of food packaging. Although essentially all of the individual components used in food packaging originate in a chemical plant, the technology for formulating and converting these materials into useful containers varies with every substrate, whether it is plastics, paper or metal. Moreover, the marketing relationship between producer, formulator, con- verter and the food industry is notably different for each segment of this "complex industry. Consequently, it is a formidable task to become an expert for even one aspect. Each of you here this evening possesses expert knowledge in one or more aspects which I wish I had. It is unfortunate that telepathic communication has not reached my level of intellectual development, so I could benefit from your experience. In fact, the only thought waves reaching me suggest that many of you should be delivering this lecture rather than I. Therefore, if you will consider me to be a substitute speaker, you may be a little more tolerant towards my remarks. When was the last time you sat down in the solitude of your study and attempted to write out a geometrical proof that the shortest distance between two points is a straight line ? Most of us would have a difficult time doing it today because, as you recall, it is not sus- ceptible to proof. It must be accepted. Indeed, some of the most difficult things in life to prove are the obvious ones. A number of months ago, I sat down to try to prove something which was obvious to m e - - t h a t there are some uses of food-packaging materials which cannot involve any hazard to health of the consumer of food. I had no preconceived idea of the end point I would reach, but it seemed like it would be fun. Sometimes now I wish I had resisted the temptation and invested my time in some other form of recreation. My main exercise was to try to determine a level of use of any food-packaging component which could be considered safe regardless of its degree of toxicity. Many of you know the conclusion I reached; namely, that any component of a food container or coating which is *Editor's note: This paper was delivered to the Fifth Annual Scientific Meeting of the British Industrial Biological Research Association (BIBRA), held in London on 25 January 1967. The Annual Scientific Meeting, instituted in 1962, provides an opportunity for members and guests of the Association to recetve an address from an eminent tOxicologiston a topic related to BIBRA's field of interest. Previousspeakers have been Professor H. C. Hodge, Dr. A. J. Lehman, Professor L. J. Goldwater and Dr. J. M. Barnes. 293 294 J.P. FRAWLEY present at 0.2 ~ or less is safe beyond any reasonable doubt, and consequently, should not be subject to government regulations. When I first advanced this proposal, it was presented within the framework of existing legislation in my country, as a mechanism for correcting some of our mistakes. Tonight, I shall cast my remarks in a different vein, since your country has not yet committed itself to a regulatory procedure on packaging materials. I do not pretend to be able to propose the best regulatory procedure for Great Britain. There are too many aspects of your business and government operations which I do not understand. However, I can review some of the evolutionary history of our regulatory procedure and suggest some ways to avoid the mistakes we have made. According to the title of my lecture, I am scheduled to talk about scientific evidence and common sense. For no particularly good reason, I am going to reverse this order and address myself to the common-sense aspects of food-packaging regulations first. It has always seemed axiomatic to me that in all matters of environmental health, the degree of hazard should define the degree of control. The amount of attention devoted to each problem should be in relation to the hazard. To distribute our limited efforts on any other basis is a form of gambling with public health. Unfortunately, in this decade of doubt, the scientific community has little control over the area of its explorations. For the most part, the decision is made by national governments as to which area of environmental health should receive concerted attention and at least in the United States, these decisions are not always made on the basis of relative hazard to health. This certainly has been the case for food-packaging materials and it appears that many countries are prepared to follow in our footsteps. In order to give a certain degree of perspective (which I believe is the foundation of com- mon sense) in this matter, let us reflect on a few of the sources of environmental exposure to chemicals. Obviously the overwhelming majority of the chemicals to which we are exposed are the natural ones which constitute our diet. We tend to overlook these and seem / ) Z " - ' , to be content in assuring society that synthetic chemicals will not cause more disease than we already have. Reluctantly, I shall dismiss any further comments on these natural products except to note that it is refreshing to have Dr. L. Golberg and the BIBRA staff occasionally remind us of our prejudice. I f we restrict ourselves to a consideration of synthetic chemicals contributed to our environment, certain obvious sources come to mind" drugs, pesticides, food additives, air pollutants, water pollutants, cosmetics, occupational exposures, household chemicals and food-packaging. As I have already suggested, logically, the amount of time and attention devoted to each of these areas should be in relation to the hazard to health. Sometimes the hazard to health is difficult to evaluate until a significant amount of time and money has been invested. However, after the degree of hazard has been defined, we have the responsi- bility of accepting the facts and of adjusting our effort accordingly. In recent years, in the United States, we have invested more industry, government and university time and money on food-packaging materials than on any environmental health problem other than pesticides and drugs. At the same time we have constructed a complex and restrictive maze of government controls which is too involved to be understood by the regulated industries. After 8 yr of effort we have now clearly demonstrated that the return on our investment has been negligible and that the health hazard is slight in comparison with other sources of chemical exposure. Consequently, I believe we have a responsibility to the public of restoring a more equitable balance to our programme. By an equitable balance, I do not mean that we should ignore all aspects of food packaging, but find some FOOD-PACKAGING CONTROL 295 common-sense approach to allow us to concern ourselves with potential hazards and not with predictably safe practices. It is to this assignment that I applied myself last s u m m e r - - t o try to develop a scientific basis for a start, and only a start, towards a common-sense approach to food-packaging regulation. However, before proceeding with the scientific evidenc6 1 have collected and the conclusions I have drawn, I have made two statements which require documentation. First, that the return on our investment has been negligible and second, that our regulatory scheme has been too complex to serve its intended purpose. Following enactment of our law, the major task facing the industry was evaluation of current industry practices. Many of you are familiar with some of the larger research programmes undertaken by different segments of the industry, for example, the petroleum wax studies by the American Petroleum Institute, the can enamel studies by the can pro- ducers and the rosin product studies by Hercules. Many other programmes which received less publicity were conducted on regenerated cellulose, polyethylene and other polymers, paper coatings and wet strength resins, to mention only a few. The net result of this invest- ment of millions of pounds has been that more than 90~o of the prior industry practices have been confirmed as safe, through a combination of low toxicity and low migration, and have been endorsed by our Food and Drug Administration (FDA), by inclusion on our permissive list. This general endorsement of the vast majority of industry practices testifies to the fact that most uses of packaging materials are inherently safe. However, before sufficient facts were accumulated to confirm the low degree of hazard associated with packaging materials, we had committed ourselves to an "omnibus" per- missive list, containing every conceivable chemical which might even" remotely come in contact with food. We now have 94 separate food-packaging regulations or lists dealing with different uses of packaging chemicals, from 967 ingredients* for adhesives to 8 ingre- dients for zinc-silicon dioxide matrix coatings. These regulations contain over 43,000 words, about 10,000 words more than the regulations for all intentional food additives. In 1966 alone, the 8th yr of the law's existence, over 200 new uses of packaging chemicals were approved and published in our Federal Register. Even a superficial examination of these regulations reveals that the vast majority of these words are devoted to the enumeration of thousands of chemicals for one or more specific uses under which most cannot migrate to food at an unsafe level, regardless of their degree of toxicity. Unless you work with these regulations on a daily basis, are familiar with the multiple cross-references and have sufficient technical training to understand what chemicals are covered by some of the vague generic terms, it is almost impossible to determine the approved uses of a given chemical. We have created a complex maze of regulations, too lengthy and involved to be understood by most of the regulated industry, with the unanticipated result of a growing apathy towards correct interpretation. This type of "omnibus" permissive list came about in the United States at the insistence of some segments of industry, coupled with a change in interpretation of our laws by the FDA---a change which revoked the long-established principle of de minimis non curat lex (the law does not concern itself with trifles) by claiming that the law does not recognize any level of a chemical as insignificant. This denial of the existence of a toxicologically-insigni- ficant level or biological zero is analagous to a denial of the existence of night, on the grounds that you cannot prove the absence of light. In all countries outside the United *967 does not count the numerous reaction products cleared for one o r m o r e of these chemicals. If these a r e counted, the number exceeds 3800. 296 j.P. FRAWLEY States, the term toxicological insignificance has real meaning and significance. I believe it is a sound scientific principle which must be preserved and eventually I hope will be restored in my country. It is this principle which I believe is the very crux of intelligent common- sense regulation of food-packaging materials. Let us take stock. Experience has taught us that most uses of food-packaging materials are safe beyond any reasonable doubt. Experience has also taught, at least in the United States, that an " o m m b u s " permissive list not only diverts the energies of our corporations, government and universities into predictably unprofitable research, but can lead to some, if not general disregard for the law. This leaves us with the conclusion that some mechanism should be found for subjecting to a permissive list only those uses of packaging materials which pose a potential hazard to health; that is, constructing a " n o n de minimis list" or "relevant" list. Here is where I invite each and everyone of you to sit down and define those uses which can be assumed to be safe. As I stated earlier this exercise sounds like fun, but it is just plain hard work. I am not certain that my solution is the best, but I assure you it is the result of many hours of reflection and analysis. In its briefest form, I believe that any chemical suitable for use in food-packaging is safe for man at a level of 0.1 p p m in the total diet. Extrapolating this dietary concentration to a practical and meaningful guideline for regulatory purposes, use of a chemical in a container at a level of 0-2 ~o or less will contribute less than 0-1 p p m to man's diet. On the surface, this may not appear to propose a major improvement, but application of this guideline would permit deletion of 75 ~ of the citations in the United States regula- tions and I am certain would permit more efficient use of manpower, in establishing permissive lists Jn other countries. I do not ask that you accept this conclusion on faith. So, as briefly as possible, I should like to review the scientific basis for this conclusion. First of all, what level of a compound, which is suitable for use in food packaging, but of unknown toxicity, can be assumed to be safe in the human diet? I know of no better approach to answering this question than to examine our toxicological experience and tabulate the experimentally determined safe level for all the compounds which have been studied. Because 90-day toxicity studies are generally considered inadequate for calculation of safe levels and because only a small number of these are published, I decided to review as many 2-yr chronic toxicity studies as I could find and to tabulate the "no-effect" level confirmed for each. I can make no claim that I have found every 2-yr chronic toxicity study which has been conducted. I can only claim that I have tabulated the "no-effect" levels from every chronic study which I could find, without any selection or rejection except irradiated foods. In total, I was able to locate 2-yr chromc toxicity studies on 220 different substances (see Appendix), and although this may seem like a modest number, it represents between 4 and 7 million pounds in toxicological research. Last September I had been able to locate only 143 such studies, but with the co-operation of many of my colleagues in the field of toxicology, I estimate that I now have collected about 90 ~o of all such studies which have been conducted. Table 1 presents the distribution of "no-effect" levels for all of the 220 compounds. It is apparent from Table 1 that a small percentage of compounds will be extremely toxic--having a "no-effect" level in experimental animals below 1 ppm, but that the majority will exhibit no toxic effect even at 100 ppm. Only 19 of the 220 compounds demonstrated FOOD-PACKAGING CONTROL 297 Table I. Distribution of "no-effect" levels m 2-yr chromc studtes "No-effect" All level compounds (ppm) (220) <1 5 <10 19 <100 40 <1000 101 < 10,000 151 any toxic effect below 10 ppm. We might conclude that the odds of detecting a toxic effect at 10 ppm from any "unknown" compound are approximately 1 in 10. Let us now look at Table 1 a little more closely and examine the nature of these 19 compounds which had a "no-effect" level at 10 ppm or less. Table 2 shows the same in- formation as Table 1, except two additional columns have been added which subdivide these 220 compounds into two categories: (1) a "heavy metals and pesticides" category and (2) an "all other compounds" category. I believe this breakdown is worthy of careful examination. The most apparent conclusion is that all 19 of the compounds, which were toxic below 10 ppm, were pesticides and heavy metal compounds. Equally significant is the fact that 39 of the 40 compounds, which had "no-effect" levels below "100 ppm in experi- mental animals, were also pesticides or heavy metal compounds. The only compound m the "all other compounds" category which was toxac below 100 ppm was acrylamide. Table 2. Distribution of "no-effect" levels in 2-yr chronic studtes Heavy "No-effect" All metals and level compounds pesticides Others (ppm) (220) (88) (132) <1 5 5 0 <10 19 19 0 < 100 40 39 1 < 1000 101 72 29 <10,000 151 86 65 It is obvious that the degree of toxicity of pesticides and heavy metals (which were used as pesticides at one time) is quite different from that of other commercial chemicals. This should represent no surprise because pesticides are synthesized, screened and selected for their toxicity to one or more forms of life before becoming commercial products. This contrast is more clearly shown by Fig. 1 which depicts the distribution of "no-effect" levels for the two categories of chemicals. It is obvious that the average toxicity of a pesticide is about 100 times as great as the average for other chemicals. 298 J.P. FRAWLEY Therefore, if we exclude heavy metals and pesticides from our consideration, experience has indicated that only a very occasional (fewer than 1 out of 100) commercial compound will have a "no-effect" level below 100 ppm and that an infinitely small number will exhibit any toxicity at 10 ppm or less. 50-- 9,°- 30- a 2o- I0 -- D //7///4"//////~ o o t I to Io0 Io00 IO,0OO Ioo,000 O~etory no-effecl level, ppm FZG. 1. Histogram showing distribution of "no-effect" levels in 2-yr chronic studies on 220 compounds. Shaded areas denote pesticides and heavy metals, and blank spaces other chemicals. Now if we apply the conventional 100-fold margin of safety to these experimentally determined "no-effect" levels, all 132 of the non-pesticidal chemicals are safe for man's diet at a dietary concentration of 0.1 ppm or higher. Many of these materials are used at much higher levels in food, but had we assumed that they were all safe at 0.1 ppm and permitted their use up to that level without toxicity studies, we would have been correct in 100 ~o of the cases and would not have exposed the public to any health hazard. For accuracy, I should point out that in the above calculation I have dealt only in orders of magnitude. If these "no-effect" levels are subdivided into small groups as 10, 30, 100 or 300 ppm, it can be concluded that all were safe to animals at 30 ppm. Moreover, I did not take into consideration the larger food consumption per kg of body weight of rats and dogs versus man. Combined, these additional calculations show that the 0-I ppm level in the human diet provides less than 1/1000th the mg/kg/day intake of the most toxic of the 132 compounds. Some may argue that I have been unnecessarily conservative, and perhaps I have been. Now let me consider briefly the other aspect of our problem migration to food--and attempt to develop some guidelines which will tell us which uses may contribute more than 0.1 ppm to the diet of man, and which uses cannot. First of all, it is obvious to everyone that any major component of a food container must be assumed to possess the capability of migrating to food at a level in excess of 0.1 ppm, unless proven otherwise. However, it is equally obvious, that some uses will not contribute 0.1 ppm to the diet. Our problem is to find this dividing line. Undoubtedly, this dividing line is different for each type of substrate. It will be different for films than for bottles. It will be different for one polymer than for another. We could establish a whole series of levels for each food-packaging substrate and its intended use, but most of these differences are not sufficient to justify complication of regulations. Therefore, in order to determine the level of addition which would contribute less than 0.1 ppm to the diet with all types of substrates, I selected the substrate which is the most permeable and susceptible to extraction by food, namely, papel. In addition 1 have selected FOOD-PACKAGINGCONTROL 299 an additive which is very readily extracted from its substrate, namely, rosin size. This combination of substrate and additive, I believe, represents the most extreme case of migration and values determined from rosin sized paper should represent a maximum for any component of any packaging media. Indeed, such data would be excessive for most uses of packaging components. In our initial efforts to study the migration of rosin size from paper, we used typical simulated solvents: various aqueous solutions, hexane, maize (or corn) oil, etc. This was wasted effort because, in water and oil, the extraction was a direct function of time and temperature and did not plateau until essentially 100 Yo of the rosin size was extracted and the integrity of the paper sheet was destroyed. Nevertheless a century and a half of experience has shown paper to be a satisfactory packaging material. Although these studies clearly demonstrated that rosin sized paper would be an appropriate choice for developing maximum migration data, they contributed nothing to the evaluation of safety of rosin size which was our principal motive at that time. As a consequence of this failure of the simulated solvents test to help define the amount actually migrating to food, we prepared radioactive samples of rosin size, incorporated them into typical commercial paper and paperboard at known levels, packaged a wide variety of food in contact with these paper samples at typical package ratios, stored them at typical storage temperatures for typical storage times and determined the rosin size content of each food by counting the radioactivity. The study was far more extensive than I shall describe, because we used several types of paper (greaseproof, waxed, unwaxed, etc.), containing three different levels of rosin size, 24 different types of food (water, ice-cream, oysters, apricots, green beans, dry breakfast Table 3. Maximum migration of rosin stze* from uncoated paper under typical storage conditions Temperature Time Migration Food (°F) (days) (ppm) Mdk products Water 34 14 5'9 Ice-cream 10 28 0"3 Vegetables Green beans 34 7 1'3 Green beans 72 14 4"1 Lettuce 34 7 2.4 Potatoes 72 28 0.2 Meats Ground beef 34 5 8'7 Ctucken 34 3 7.2 Beefsteak 34 7 4'9 Sausage 34 5 124 0 Fruits Apricots 72 28 0-1 Apples 72 28 1"2 Grain products Puffed rice 72 14 58 Wheaties 34 28 7.0 Flour 72 28 0"2 Doughnuts 72 3 0.9 Others Sugar 72 28 0"2 Butter 34 14 32'8 *4 Yoin paper. 300 J.P. FRAWLEY food, sugar, doughnuts, ground beef, butter, bacon, sausage, to name just a few) and anal- ysed each sample at several different storage intervals and temperatures. For our purposes, I have selected only the uncoated and unwaxed paper and only the maximum migration levels obtained for the 18 commodities packaged in these uncoated papers under typical commercial storage conditions. Adrmttedly this gives unrealistically high values for rosin size which are not typical of industry praetice, but for our present purposes, the worst case must be presented. Table 4. Calculatton o f maximum migration o f rosin size* to total diet Contribution Percentage Average to total of migration diet Commodity group diet (ppm) (ppm) Milk products 31 3"1 1'0 Vegetables 20 20 04 Meats 18 38 2 69 Fruits 13 0'5 01 Grain products 10 3-5 0'4 Sugar 5 02 0-0 Butter, oils 3 32"8 09 Total ... 9"7 *4 % in paper. Table 3 shows the maximum migration value for 18 food commodities at various typical storage times and temperatures when exposed to paper containing an average of 4~o rosin size. It is obvious from some of these values that high levels of migration can occur with some foods, whereas other foods contain much less rosin size. The data in Table 3 can be quickly considered since the individual values are of no great significance, but the com- posite of these values can be helpful. Table 4 shows the average consumption of these various commodity groups in the USf, the average migration to that commodity group and a calculation of the maximum level of rosin size in the average total diet, if 100 Yo of the diet were packaged in uncoated paper containing 4 Yo rosin size. Undoubtedly there are some differences in dietary habits between our countries, but I doubt that they would significantly alter the calculation. As I mentioned previously, three different sizing levels were used in these studies. Table 5 shows the final dietary calculations for the same foods, under the same conditions for paper containing 2 or 1 ~o rosm size. The extrapolation is remarkably good. The last column shows the migration in ppm expressed on the basis of a unit of 1 ~o rosin size in the paper or container. For each per cent addition to the container, man's diet would contain a maximum of 2 ppm of the additive, if the entire diet was in contact with that container. One further calculation is necessary in order to arrive at a realistic determination of the Table 5. Maximum migration o f rosin size to diet Level of size in paper Migration (%) (ppm) ppm/~o 4 9-7 2"4 2 4'4 2"2 1 1 "9 1 "9 "fU.S. Department of Agriculture Dietary Evaluation of Food Used in Households in U.S., 1961. FOOD-PACKAGING CONTROL 301 level of addition which will contribute no more than 0.I p p m to the diet. Obviously, 100 7o of man's diet is not in contact with paper, or any other single type of food container. There are five major types of food container substrates, glass, metal, paper, plastics and regenerated cellulose. In addition, there is a significant portion of food which is not packaged or is packaged in bulk so that most of the individual units are never in contact with the container. It is impossible to get reliable figures revealing the percentage of the food-pack- aging market shared by each type. However, it is conservative to assume that no more than 25 Yo of man's diet is in contact with any given type of food package or packaging additive. Table 6 shows this final calculation of the maximum migration to the diet which would result from the use of a component at a level of 1 ~o in the container (as directly measured from the rosin size experiments) and the maximum migration from a level in the container of 0.2 Yo. Undoubtedly, this calculation is an exaggeration for most uses of packaging com- ponents which possess greater insolubihty or which are used in substrates more resistant to penetration than paper. Nevertheless, it permits the conservative conclusion that any compon- ent of an article contacting food which xs present m the article itself or its coating at a level of 0.2 % or less by weight will contribute to the diet a level which can be of no possible public health significance. Consequently, such trivial uses should not be included on lists of components permitted in food packaging. Table 6. Calculation o f maximum contribution to the diet Maximum total diet migration ... 2 ppm/each Yo Maximum diet in contact ... 25 Yo Concn in Maximum concn package in diet (%) ~pm) 1 '0 0-5 0"2 0"1 As most of you know, I submitted this conclusion to the profession and to the industry in the United States last September. It was worded differently, by describing a level of 0.2 70 as GRAS*, because of the particular structure of our law. It has received overwhelming and gratifying support and only a few questions have been raised. Our own Food and Drug Administration has authorized me to tell you that they are giving it serious consideration, but could not reach a decision prior to this meeting. At this point, I can only say that after 6 months, I am even more convinced that the conclusion is sound and conservative; and, that it offers a common-sense approach to a reduction in the extensive waste of time and money of both industry and government on problems which can be of no possible public health siLmificance. F r o m a regulatory point o f view, I believe it offers one mechanism of avoiding preparation and constant modification o f an omnibus permissive list which experience has shown becomes so unwieldy and com- plicated as to invite disregard. I do not believe that type of situation is in the best interest of the consumer, industry or government. Now I have intentionally reserved for my closing remarks a discussion of the few questions and mental reservations about this proposal which have been raised by groups throughout the world. Perhaps, some of these points will answer some of the questions you would like to raise. • Goncrally recognized as safe. 302 J . P . FRAWLEY The most frequent comment is a concern that despite our toxicological experience to date, we cannot assume that the next compound will not be toxic at 0.1 ppm. The same basic concern has been expressed in another way, by expressing doubt that toxicity data from 2-yr chronic studies represent a valid cross-section of chemicals, since some of the more toxic ones are rejected by short-term toxicity tests. It is, of course, possible that some chemical may be synthesized at some time in the future which would be toxic at 0.1 p p m in the diet. However, it is almost impossible for such a compound to become an intentional component of a food container. For a compound to be toxic for man at 0.1 p p m presumes that it will be as toxic or more toxic than any commercial pesticide. For it to be used as a component of a food container presumes that it must be manufactured, packaged, distributed, and in other ways handled several times before contacting the food. It is inconceivable that a compound as toxic as this could pass through so many hands, in an industry not accustomed to handling highly toxic substances, without revealing its toxicity through injury to personnel. Once recognized, safe handling of such a compound would require such extreme industrial hygiene precautions as to be incompatible with converting operations and food-packaging practices. It seems to me that in order to produce an unsafe food package, due to incorporation of a toxic ingredient at a level of 0-2 ~o or less, it would require a deliberate or intentional act on the part of a manu- facturer to poison the public, without at the same time poisoning his own workers. N o amount of legislation or regulation can protect against such insanity. It has been suggested by a few of my colleagues that the extreme toxicity of such materials as aflatoxin rules out an assumption that any chemical is safe at even one part in a thousand million unless it has been tested. I believe that such an assumption is valid, if we limit our discussion to certain uses or industries. Again, I believe common sense tells us that it is inconceivable that anyone could manufacture millions of pounds of aflatoxin, or any substance of extreme toxicity and distribute it for use as a stabilizer in plastics or wet- strength resins for paper without finding out that it was too toxic for that industry. N o company can afford to lose customers that way. Accidental contamination with aflatoxin or other extremely toxic substances is another matter, but this is outside the considerations of a permissive list. Whether a permissive list contains 200 or 20,000 substances, accidental contamination is no more or less likely. Quality control, inspection, and personal attention to details in manufacture are all necessary ingredients to the prevention of contamination of any product. A few individuals have questioned the validity of my estimate that no more than 25 of the diet will be in contact with the same packaging substrate or chemical. In rare circum- stances, of course, some individuals may eat canned foods almost exclusively and some may eat fresh or unpackaged food almost exclusively. These variations in dietary habits, along with other intraspecies differences have been taken into consideration as part of the basic concept of our 100-fold margin of safety. Moreover, as mentioned above, the conservative calculations used above provide a 1000-fold margin of safety. The principal objection to this proposal in the United States has been administrative. Adoption of this proposal would obviate the need for m a n y of our packaging regulations and would suggest a complete rewriting of Subpart F. For example, the "general adhesives" and "defoamer in paper manufacture" regulations would be replaced by a statement o f good manufacturing practice that adhesives should not contact the food (as is already provided despite the fact that thousands of chemicals are enumerated) and that defoamers may be used only prior to and during sheet-forming process (as is also already provided). FOOD=PACKAGINGCONTROL 303 I t h i n k the time necessary to accomplish this task would be time well spent a n d would be rapidly recovered in reduced administrative costs. Let me close with this concept. Y o u r c o u n t r y and mine have a limited n u m b e r of com- petent specialists in the field of e n v i r o n m e n t a l health. There are m a n y problems facing our society which are competing for their time a n d attention, as well as for financial support. Included are c o m m u n i t y air a n d water pollution, industrial exposures, household chemicals, pesticides, drugs, cosmetics, food additives, confined e n v i r o n m e n t s of space cabins a n d s u b m a r i n e s a n d food-packaging. H a v i n g some professional responsibility in all of these segments, I a m convinced that food packaging constitutes the least hazard to health o f all o f these. Yet in recent years it has c o m m a n d e d more time and attention t h a n a n y other area other t h a n drugs a n d pesticides. It is o u r moral and professional responsibility to invest o u r time a n d m o n e y in research which is likely to provide the greatest protection to health. I suggest that m y proposal is a start toward restoration of a proper balance. APPENDIX No-effect levels established by 2-yr feeding studies No-effect No-effect level level Compound (ppm) Compound (ppm) Acrylamide x 40 Carbarsone (p-Ureidobenzenearsonic Aldrin2 <0.5* acid)t8 1000 Alkyl ketene dimera I000 Carboxymethylcellulose (CMC) x2 10,000 Allethrm' 4000* fl-Carotene" 1000 Amiben (2,5-Dichloro-3-aminobenzoic CatechoD~ 1250T acid)s 10,000" Chlorbenside (p-Chlorobenzyl Ammonium sulphamate 6 500 p-chlorophenyl sulphide)' 20~ Antimony chloride' <5001" Chlorbenside, sulphone derivative4 20* Arsomc acld8 50* Chlordane' 2-5* 1-Ascorbyl palmitateQ 2500 bis(p-Chlorphenoxy)methane ~ 300 Barium chloride' 20001" p-Chlorophenyl p-chlorobenzene- Benzene hexachloride, technical (BHC)' 10" sulphonatC 25* a-Benzene hexachloride (a-BHC)' 10" Chlorpropamide 62 1250 //-Benzene hexachloride (fl-BHC)7 < 10" Chlortetracycline" 10,000 7-Benzene hexaehloride (7-BHC)' 50* Citrus Red No. 2 (C.I. (1956) No. 12,156)" 500 dLBenzene hexachloride (8-BHC)7 <800* Copper chromate 4 500* Benzoic acid10 5000 Cub6" 50* Biureaxx 7500 Cupric chloride 4 500'1" Butoxypolypropylene glycol (tool wt Dalapon4 300* 800)1 640 D & C Orange No. 5" 10,000 Butylated hydroxyanisole (BHA)1° 5000 D & C Orange No. 10" 10,000 Butylated hydroxytoluene (BHT)x2,13 1000 D & C Red No. 927 500 Butyl 3,4-dihydro-2,2-dimethyl-4-oxo-1, D & C Red No. 1028 500 2H-pyran-6-carboxylate (Indalone)" 40,000 D & C Red No. 21" 10,000 1,3-Butylene glycoFs 30,000 D & C Red No. 2726 I0,000 2-(p-tert-Butylphenoxy) isopropyl DDD fiDE)' 10" 2'-ehloroethyl sulphite (Aramite) le 100" DDT' 1" tert-Butylphenyl salicylatex 2000 Dehydroacetic acid ~ 1000 Cadmium chloride' < 10t Diazinon4 0.75* Calcium disodium ethylenediamine- Dichlone 4 <500* tetmacetate 1,t 7 5000 1,1-Dichloro-2,2-bis(p-ethylphenyl)ethane Captan' 1000~ (Perthane) 4 100" *Pesticide "l'Heavy metal T--Turnouts at higher levels. 304 J. P. FRAWLEY Appendix (contd) No-effect No-effect level level Compound (ppm) Compound (ppm) 2,4-Dichloro-6-o-chloroanilino-s- 2-Heptadecyl glyoxahdine acetate triazine (Dyrene)4 5000* (Glyodinp 210" 2,4-Dichlorophenoxyethyl sulphate, n-Heptyl-p-hydroxybenzoate4e 1500 sodmm salt~ 200* 1-[5-(3a,4,5,6,7,7a-Hexahydro-4,7- 4,4'-Dtchloro-a-trichloromethyl- methanoindanyl)]-3,3-dimethylurea benzhydrol (Kelthane)~ 20* (Herban) 3 500* Dicyandiamide~-" 2500 Hydroxyethylcellulose47 10,000 Dieldrin ~ 0-5* Hydroxypropylmethylcellulose~ 50,000 O,O-Diethyl O-3-chloro-4-methyl-1- Hydroq uinone~1 10,000T oxo-2H-l-benzopyran-7-yl d-Isoascorblc acid ~ 10,000 phosphorothioate (Co-Ral) 4 2* d-Isoascorbyl palmitate~x 2500 Di(2-ethylhexyl) phthalate ~ 1300 Isopropyl N-(3 chlorophenyl) carbamate Dl-n-hexyl azelates° 5000 (CIPC) 48 2000* D~-isobutyl adipatO 5000 4,4'-Isopropylidene bis(2-Isopropyl- Ddauryl thiodapropmnic acid ~ 30,000 phenol)1 1000 Dimethyl carbate ~4 10,000 Light Green SF Yellowish4° 10,000 2,4-Dimethyl-2-methylene-1,2,4- Malathion 4 100" thmdmzolidme-5-thtonO 100* Malelc hydrazide4 20,000* Dimethyl phthalatO 4 20,000 Maneb a 25* 3,5-Dlmethyltetrahydro-1,3,5,2H- Melamine-formaldehyde resin thmdtazine-2-ttuone (Mylone)~1 < I0" (Parez 607) 28 50,000 O,O-Dimethyl-O-(2,4,5-trichlorophenyl) Mercaptobenzothtazole4 120* phosphorothioate (Ronnel)4 10* Mercury acetate 7 2-5* 3,5-Dimtrobenzamide~ 600* Methoxychlor~,49 200* 3,5-Dimtro-o-toluamidO 62* O-Methyl-O-(4-tert-butyl-2-chlorophenyl) DJphenyP, 83 500* methylphosphoroamidothloate Diphenylamine~ 100" (Ruelene)l,50 30* 3-(2-Dtphenylyloxy)-1,2-epoxypropane~ 2000 O-Methyl-O-(2,4-dichlorophenyl) Distearyl thzod~propiomcaczd~ 30,000 isopropylphosphoramidothioatex 10" Dmron 4 125" Methyl p-hydroxybenzoate5x 20,000 Dodecyl benzene sodium sulphonate Methyl methacrylate~6 100 (Santomerse no. 3)s~ 2000 Methyl naphthaleneacetlc acid ~ 2500* Dodecyl gallate x° 350 Methylpolysdoxane1 3000 Dodine a,~n 50* Methyl salicylate5~ 10,000 Endosulphan~ 30* Monuron 4 250* EPN ~ 5* 1-NaphthyI-N-methylcarbamate 1 200* Epoxidized soybean oil (Paraplex a-Naphthylthiourea58 50* G-60) ~ 25,000 N~cotme 4 62 Epoxldized soybean oil (Paraplex Nordlhydroguaiaretic acid 21 2500 G-62) ~ 5000 Nylon (Zytel)54 100,000 4-Ethoxyphenylurea (Sucrol, dulcin) z7 < 1000 Octadecylamines5 500 Ethoxyqum4 120* Octyl gallate1° 350 Ethyl acrylate~ 100 p-tert-Octylphenoxy-polyethoxy Ethyl 4,4'-dichlorobenzilate4 50* ethanols (Triton X-405p 6 14,000 2-Ethyl hexanediol-l,3xs 40,000 Parathion 4 1" 2-Ethylhexyl diphenyl phosphate Petrolatum n~ 50,000 (Santtc~zer 141)a9 1250 Petroleum wax no. 2s8 I00,000 Ethyl phthalyl ethyl glycolate~ 5000 Petroleum wax no. 858 100,000 Fast Green FCF 4° 10,000 Petroleum wax no. 1258 100,000 FD & C Blue No. 1~x 5000 Petroleum wax no. 15~8 100,000 FD & C Blue No. 24x 1000 Petroleum wax no. 2058 I00,000 Ferbam ~ 200* Phenacetins9 630 Glycerol4~ 100,000 PhenoP 1 10,000 Glycerol monostearate's 250,000 Phenyl mercuric acetate v 0" 1" Gum gumaO ~ 5000 o-PhenylphenoP° 2000* Gum rosin, pale 4n 500 Pimaricinel 500 Hel~tachlor epoxide4 0.5* Piperonyl butoxide4 700* FOOD-PACKAGING CONTROL 305 Appendix (contd) No-effect No-effect level level Compound (ppm) Compound (ppm) Polyacrylamide (Separan AP30) s~ 10,000 Sodmm lauryl sulphatC t I0,000 Polyacrylamlde (Separan NP10) ~2 10,000 Sodmm lauryl trioxyethylene Polyethylene glycol (tool wt 200) t 40,000 sulphonate 72 5000 Polyethylene glycol (mol wt 400) ~ 20,000 Sodmm monofluoracetatC <5* Polyethylene glycol (tool wt 1500)x 2000 Sodium nitratC 3 10,000 Polyethylene glycol (mol wt 1540)t 40,000 Sodium #-sulphopropionamidex 10,000 Polyethylene glycol (mol wt 4000) ~ 40,000 Sodmm tnpolyphosphate63 5000 Polymenzed turpentine resin3 2000 Sorbic acld x° 50,000 Polyoxyethylene(20)sorbitan Sorbitan monopalmitate (Span 40) 63 50,000 monolaurate (Tween 20) e3 50,000 Sorbitan monostearate (Span 60) 84 50,000 Polyoxyethylene(20)sorbitan Sorbitan tristearate (Span 65) 6s 50,000 monooleate (Tween 80) 6s 50,000 Sulphenone (p-chlorophenyl phenyl Polyoxyethylene(20)sorbitan sulphone)4 I00" monopalrnitate (Tween 40)~s 50,000 Tall oil rosin, pale45 2000 Polyoxyethylene(20) sorbitan Tartar emetm (Potassium antimonyl monostearate (Tween 60) 64 50,000 tartrate) 4 <500q Polyoxyethylene(20)sorbitan Tartaric acid 74 12,000 tristearate (Tween 65) ~n 50,000 Tartrazine ~s 10,000 Polyoxyethylene(8)stearate (Myrj 45) ~ 20,000 Terpene polychlorinates (Strobane) 4 50* Polyoxyethylene(40)stearate (Myrj 52) a~ 50,000 Tetradifon4 300* Ponceau 3R (C.I. (1956) No. 16,155) ~ 5000 Thiodiproplonic acid 2~ 30,000 Ponceau SX (C.I. (1956) No. 14,700) ~6 50,000 Thiourea TM 500* Potassium bromate ~ 627 Thiram 4 200* 1-n-Propoxy-2-amino-4-nitrobenzene Toxaphene 7 25* (P.4000)s 7 < 1000 2,4,5-Tnchlorophenoxyethyl sulphate, Propyl gallatC ~ 10,000 sodmm salt ~ 200 Propyl p-hydroxybenzoatest 20,000 Tri(polynonylphenyl) phosphite Pyrethrum 7 1000" (Polygard) 77 3300 Rosin, disproportionated4~ 500 Tylosin 7a 10,000 Rosin, fully dimerized~s 500 Vinyl chloride--vinyl acetate Rosin, partially dimenzed45 500 copolymer78 120,000 RotenonC 2* Vmyhdene chloride-vinyl chloride Saccharin~7 10,000 copolymerx 50,000 Selenium 7 <3* Wood rosin, dark 45 500 Sodium alginate~7 50,000 Wood rosin, fully hydrogenated45 500 Sodmm alkylbenzenesulphonatC 8 5000 Wood rosin, hydrocarbon insoluble Sodium bisulphite69 500 residue45 500 Sodium chromate TM 300* Wood rosin, pale45 2000 Sodium cyclamate37 10,000 Wood rosin, partmlly hydrogenated~6 2000 Sodmm 2,2-dichloropropionatC 300 Yellow AB 8° 500 Sodmm dioctyl sulphosuccinatC~ 5000 Yellow OB 8° 500 Sodium hexametaphosphate63 5000 Zineb 4 500* Sodium lauryl glycerylsulphonatC~ 5000 Zlram 4 250* REFERENCF~ 1. Well, C. S. & McCotlister, D. D. (1963). 5. Hazleton, L. W., Paynter, O. E. & Weir, R. J. Relatlonslup between short- and long-term (1964). Safety evaluation studies on 2,5- feeding studies in designing an effective dichloro-3-aminotmnzoicacid (Amiben). Toxic. toxmity test. J. agric. Fd Chem. 11, 486. appl. Pharmac. 6, 349. 2. Davis, K. J. & Fitz.hugh, O. G. (1962). Tumori- 6. Sherman, H. & Stula, E. F. (1965). Toxicity genic potential of aldrin and dieldrin for mice. studies on ammonium sulfamate. Toxic. appl. Toxic. appL Pharmac. 4, 187. Pharmac. 7, 497. 3. Hercules Incorporated (Unpublished data). 7. Lehman, A. J. (1952). In Chemicals in Foods: A 4. Lehman, A. J. (1965). Summaries o f Pesticide Report to the Associatwn o f Food and Drug Toxictty. The Association of Food and Drug Officials on Current Developments, Part II. Officials of the United States. Kansas. Pesticides. Vol. XVI, No. 2. p. 47. The 306 J.P. FRAWLEY Association of Food and Drug Officials of the hydrochloride tn rats. Toxic. appl. Pharmac. United States. 3, 654. 8. Prier, R. F., Nees, P. O. & Derse, P. H. (1963). 24. Dacre, J. C. (1965). Chronic toxicity and The toxicity of an organic arsenical, 3-nitro-4- carcmogenicity studies on Citrus Red No. 2. hydroxyphenylarsonic acid. II. Chronic Proc. Univ. Otago reed. Sch. 43, 31. Toxicity. Toxic. appl. Pharmac. 5, 526. 25. Hansen, W. H., Davis, K. J. & Fitzhugh, O. G. 9. Fitzhugh, O. G. & Nelson, A. A. (1946). (1965). Chrome toxicity of cub6. Toxic. appl. Subacute and chronic toxielties of ascorbyl Pharmac. 7, 535. palmitates. Proc. Soc. exp. Biol. Med. 61, 195. 26. Industrial Bin-Test Laboratories (Unpublished 10. Joint FAO/WHO Expert Committee on Food data). Addltwes---Sixth Report (1962). Evaluation 27. Davis, K. J. & Fitzhugh, O. G. (1962). of the Toxicity of a Number of Antimlcrobials Pathologzc changes noted in rats fed D & C and Antioxidants. Tech. Rep. Ser. WId Hlth Red No. 9 for two years. Toxic. appl. Pharmac. Org. 228. 4, 200. 11. Oser, B. L., Oser, Mona & Morgareidge, K. 28. Davis, K. J. & Fitzhugh, O. G. (1963). (1965). Studies of the safety of azodicarbon- Pathologic changes noted m rats fed D & C amide as a flour-maturing agent. Toxic appl. Red No. 10 [monosodium salt of 2-(2-hydroxy- Pharmac. 7, 445. l-naphthylazo)-l-naphthalenesulfonicacid] for 12. Deichmann, W. B., Clemmer, J. J., Rakoczy, two years. Toxic. appl. Pharmac. 5, 728. R. & Bmnchine, J. (1955). Toxicity of di- 29. Amertcan Cyanamid Company (Unpublished tertiarybutylmethylphenol. A.M.A. Archs rod. data). Hlth 11, 93. 30. Hodge, H. C., Maynard, E. A., Downs, W. L., 13. Joint FAO/WHO Expert Commtttee on Food Salerno, Lillian & Packer, J. T. (1962). Chrome Additives--Ninth Report (1966). Spectfica- oral toxicity studies of di-n-hexyl azelate m tions for the Identity and Purity of Food rats and dogs. Toxic. appl. Pharmac. 4, 247. Additives and their Toxicological Evaluation: 31. Smyth, H. F., Jr., Carpenter, C. P. & Wed, Some Antimicrobials, Antioxidants, Emulst- C. S. (1966). Toxicologic studies on 3,5- tiers, Stabihzers, Flour-Treatment Agents, dimethyltetrahydro-l, 3, 5, 2H-thiadiazine-2- Acids and Bases. Tech. Rep. Set. Wld Hlth thione, a sod fungicide and shmicide. Toxic. Org. 339. appl. Pharmac. 9, 521. 14. Lehman, A. J. (1955). Insect Repellents. Vol. 19. 32. Kerr, K. B., Abrahamson, P. R. & Ipson, J. R. p. 87. The Association of Food and Drug (1965). Evaluation of the safety of 3,5-dinitro- OffÉcials of the United States. benzamide. Toxic. appl. Pharmac. 7, 488. 15. Scala, R. A. & Paynter, O. E. (1964). Safety 33. Ambrose, A. M., Booth, A. N., DeEds, F. & evaluation studies on 1,3-butylene glycol. Cob, A. J., Jr. (1960). A toxicological study Toxic. appl. Pharmac. 6, 358 of biphenyl, a citrus fungistat. Fd Res. 25, 328. 16. Oser, B. L. & Oser, Mona (1962). 2-(p-tert- 34. Paynter, O. E. & Weir, R. J., Jr. (1960). Butylphenoxy)isopropyl 2-chloroethlyl sulfite Chromc toxicity of Santomerse No. 3 from (Aramite). II. Carcinogenicity. Toxic. appL olefin (dodecyl benzene sodium sulfonate). Pharmac. 4, 70. Toxic. appl. Pharmac. 2, 641. 17. Oser, B. L., Oser, Mona & Spencer, H. C. 35. Levinskas, G. J., Vidone, Lena B., O'Grady, (1963). Safety evaluation studies of calcium J. J. & Shaffer, C. B. (1961). Acute and chronic EDTA. Toxic. appl. Pharmac. 5, 142. toxicity of dodine (n-dodecylguanidineacetate). 18. Oser, B. L., Morgareidge, K., Wemberg, M. S. Toxic. appL Pharrnac. 3, 127. & Oser, Mona (1966). Carcinogenicity study 36. Larson, P. S., Firmegan, J. K., Haag, H. B., of carbarsone. Toxic. appL Pharmac. 9, 528. Smith, R. B., Jr., & Hennigar, G. R. (1960). 19. Shelanska, H. A. & Clark, A. M. (1948). Chronic toxicity studies on two epoxidized Physiological action of sodmm carboxy- soybean oils in the rat and dog. Toxic. appL methylcellulose on laboratory animals and Pharmac. 2, 649. humans. Fd Res. 13, 29. 37. Fitzhugh, O. G., Nelson, A. A. & Frawley, 20. Bagdon, R. E., Zbinden, G. & Studer, A. J. P. (1951). A comparison of the chrome (1960). Chronic toxicity studies of fl-carotene. toxicities of synthetic sweetemng agents. J. Am. Toxic. appL Pharmac. 2, 225. pharm. Ass. Sci Ed. 40, 583. 21. Lehman, A. J., Fltzhugh, O. G., Nelson, A. A. 38. Borzelleca, J. F., Laxson, P. S., Hennigar, & Woodward, G. (1951). The pharmacological G. R., Jr., Huff, E. G., Crawford, E. M. & evaluation of antioxidants. Adv. Fd Res. 3, 197. Smith, R. B., Jr. (1964). Studies on the chronic 22. Delahunt, C. S., P'an, S. Y., Dardin, V. J. & oral toxicity of monomeric ethyl acrylate and Schneider, J. A. (1960). Prolonged administra- methyl methacrylate. Toxic. appl. Pharmac. tion studies with chlorpropamide. Toxic. appl. 6, 29. Pharmac. 2, 195. 39. Treon, J. F., Dutra, F. R. & Cleveland, F. P. 23. Dessau, F. I. & Sullivan, W. J. (1961). A two- (1953). Toxicity of 2-ethylhexyl diphenyl year study of the toxicity of chlortetracyclme phosphate. A.M.A. Archs ind. Hyg. 8, 170. FOOD-PACKAGINGCONTROL 307 40. Hansen, W. H., Long, Eleanor L., Davis, K. J., 55. MacDonald, W. E., Deichmann, W. B., Nelson, A. A. & Fitzhugh, O. G. (1966). Radomskl, J. L. & Austin, B. S. (1962). The Chronic toxicity of three food colourmgs: chronic toxicity of octadecylamine in the r a t - Guinea Green B, Light Green SF Yellowish a supplemental report. Toxzc. appL Pharmac. and Fast Green FCF in rats, dogs and mice. 4, 610. Fd Cosmet. Toxicol. 4, 389. 56. Larson, P. S., Borzelleca, J. F., Bowman, E. R., 41. Hansen, W. H., Fitzhugh, O. G., Nelson, Crawford, E. M., Smith, R. B., Jl:., & Hermigar A. A. & Davis, K. J. (1966). Chromc toxicity G. R. (1963). Toxicologic studies on a prepara- of two food colours, Brilliant Blue FCF and tion of p-tertiary octylphenoxy-polyethoxy Indigotine. Toxic. appl. Pharmac. 8, 29. ethanols (Triton X-405). Toxic. appL Pharmac. 42. Hine, C. H., Anderson, H. H., Moon, H. D., 5, 782. Dunlap, M. K. & Morse, M. S. (1953). 57. Oser, B. L., Oser, Mona & Carson, S. (1965). Comparative toxicity of synthetic and natural Toxicologic studies of petrolatum m mice and glycerin. A.M.A. Archs ind. Hyg. 7, 282. rats. Toxic. appL Pharmac. 7, 382. 43. Ames, S. R., O'Grady, M. P., Embree, N. D. 58. Shubik, P., Saffiotti, U., Lijinsky, W., Pietra, & Harris, P. L. (1951). Molecularly distilled G., Rappaport, H., Toth, B., Raha, C. R., monoglycerides, III. Nutritional studies on Tomatls, L., Feldman, R. & Ramahi, H. monoglycerides derived from cottonseed od. (1962). Studies on the toxicity of petroleum J. Am. Oil Chem. Soc. 28, 31. waxes. Toxic. appl. Pharmac. (Supp.) 4, 49. 44. Johnson, V., Carlson, A. J., Kleitman, N. & 59. Woodard, G., Post, K. F., Cockrell, K. O. & Bergstrom, P. (1938). Action of gum guaiacum Cronin, M.T.I. (1965). Phenacetin: Long-term upon the animal organism. Fd Res. 3, 555. studies in rats and dogs. Toxic. appL Pharmac. 45. Frawley, J. P. (1965). Investigations estab- 7, 503. hshmg the safety of rosin products for food, 60. Hodge, H. C., Maynard, E., Blancher, H. J., food packaging applications. Nay. Stores Rev. Jr., Spencer, H. C. & Rowe, V. K. (1952). Ter. Chem. 74 (12), 4. Toxicological studies of ortho-phenylphenol 46. Paynter, O. E., Hazleton, L. W. & Garlock, m (Dowicide I). J. Pharmac. exp. Ther. 104, E. A. (1966). Toxicologic evaluation of the 202. n-heptyl ester ofp-hydroxybenzoic acid. Toxic. appl. Pharmac. 8, 350. 61. Levinskas, G. J., Ribelin; W. E. & Shaffer, 47. Smyth, H. F., Jr., Carpenter, C. P. & Weil, C. B. (1966). Acute and chronic toxicity of C. S. (1947). Chronic toxicity of hydroxyethyl pimaricin. Toxic. appl. Pharmac. 8, 97. cellulose for rats. J. Am. pharm. Ass. 36, 335. 62. McCollister, D. D., Hake, C. L., Sadek, S. E. 48. Larson, P. S, Crawford, E. M, Smith, R. B., & Rowe, V. K. (1965). Toxicologic investiga- Jr., Henmgar, G. R., Haag, H. B. & Finnegan, tions of polyacrylamides. Toxic. appl. Pharmac. J. K. (1960). Chronic toxicologic studies on 7, 639. isopropyl N-(3-chlorophenyl) carbamate 63. Joint FAO/WHO Expert Committee on Food (CIPC). Toxic. appl. Pharmac. 2, 659. Additives---Seventh Report (1964). Specifica- 49. Hodge H.C., Maynard, E. A. & Blanchet, H. tions for the Identity and Purity of Food J., Jr. (1952). Chronic oral toxicity tests of Additives and Their Toxicological Evaluation: methoxychlor [2, 2-di-(p-methoxyphenyl)-l,1, Emulsifiers, Stabdizers, Bleaching and Matur- 1-trichloroethane] in rats and dogs. J. Pharmac. ing Agents. Tech. Rep. Ser. Wld Hlth Org. exp. Ther. 104, 60. 281. 50. McCollmter, D. D., Oyen, F., Olson, K. J. & 64. Fitzhugh, O. G., Bourke, Anne R., Nelson, Rowe, V. K. (1965). Toxicology of 4-tert- A. A. & Frawley, J. P. (1959). Chronic oral butyl-2-chlorophenyl methyl methylphosphor- toxicities of four stearic acid emulsifiers. Toxic. amidate (Ruelene). Toxzc. appl. Pharmac. 7, appl. Pharmac. 1, 315. 490. 65. Hansen, W. H., Davis, K. J., Fitzhugh, O. G. 51. Matthews, C., Davidson, J., Barren, E., & Nelson, A. A. (1963). Chronic oral toxicity Morrison, J. L. & Richardson, A. P. (1956). of Ponceau 3R. Toxic. appl. Pharmac. 5, 105. p-I-Iydroxybenzoic acid esters as preservatives, 66. Davis, K. J., Nelson, A. A., Zwickey, R. E., II: Acute and chronic toxicity in dogs, rats Hansen, W. H. & Fitzhugh, O. G. (1966). and mice. J. Am. pharm. Ass. 45, 260. Chronic toxicity of Ponceau SX to rats, mice 52. Webb, W. K. & Hansen, W. H. (1963). and dogs. Toxic. appL Pharmac. 8, 306. Chronic and subacute toxicology and path- 67. Nilson, H. & Wagner, J. E. (1951). Feeding ology of methyl salicylate in dogs, rats, and tests with some algin products. Proc. Soc. exp. rabbits. Toxic. appl. Pharmac. 5, 576. BioL Med. 76, 630. 53. Fitzhugh, O. G. & Nelson, A. A. (1947). 68. Tusing, T. W., Paynter, O. E. & Opdyke, D. L. Chronic oral toxicity of alpha-naphthyl (1960). The chronic toxicity of sodium alkyl- thiourea. Proc. Soc. exp. Biol. Med. 64, 305. benzenesulfonate by food and water admim- 54. Haskell Laboratories, DuPont Company (Un- stratlon to rats. Toxic. appl. Pharmac. 2, pubhshed data). 464. B 308 J . P . FRAWLEY 69. Fltzhugh, O. G., Knudsen, L. S. & Nelson, (1964). Chronic rat and dog toxicity studies A. A. (1946). Chronic toxicity of sulfites. J. on tartrazine. Toxic. appl. Pharmac. 6, 621. Pharmac. exp. Ther. 86, 37. 76. Fitzhugh, O. G. & Nelson, A. A. (1948). Lwer 70. Food and Drug Admimstratlon (Unpublished turnouts m rats fed thlourea or thioacetarnide. data). Science, N.Y. 108, 626. 71. Fitzhugh, O. G. & Nelson, A. A. (1948). 77. Carson, S., Oser, B. L. & Oser, Mona (1964) Chronic oral tox~c~ties of surface-active agents. Toxicologic studies w~th the ant~oxidant J. Am. pharm. Ass. Sci. Ed. 37, 29. Polygard. Toxic. appl. Pharmac. 6, 342. 72. Tusing, T. W., Paynter, O. E., Opdyke, D. L. 78. Anderson, R. C., Worth, H. M., Small, R. M. & Synder, F. H. (1962). Toxicologic studies & Harris, P. N. (1965). Toxicological studies on sodium lauryl glyceryl ether sulfonate and on Tylosm: Its safety as a food additive. sodium lauryl trioxyethylene sulfate. Toxic. Toxic. appl. Pharmac. 7, 478. appl. Pharmac. 4, 402. 79. Smyth, H. F., Jr. & Weil, C. S. (1966). Chxomc 73. Lehman, A. J. (1958). Nitrates and nitrites in oral toxicity to rats of a vinyl chloride-vinyl meat products. Q. Bull. Ass. Fd Drug Off. acetate copolymer. Toxic. appL Pharmac. 9, U.S. 22, 136. 501. 74. Fitzhugh, O. G. & Nelson, A. A. (1947). The 80. Hansen, W. H., Nelson, A. A. & Fitz_b.ugh comparative chronic toxicities of fumarlc, O. G. (1963). Chromc toxicity of Yellow AB tartaric, oxalic and malelc acids. J. Am. (1-phenylazo-2-naphthylamlne) and Yellow pharm. Ass. Sci. Ed. 36, 217. OB (1-o-tolylazo-2-naphthylamme). Toxic. 75. Daws, K. J., Fitzhugh, O. G. & Nelson, A. A. appl. Pharmac. 5, 16.