THE ANALYST. JUNE, 1902. FLUORIDES AS BUTTER PRESERVATIVES, WITH OBSERVATIONS ON THEIR INFLUENCE ON ARTIFICIAL DIGESTION. BY OTTO AND CHARLES W. HEHNER. (Read at the Meeting, Nay 7, 1902.) THE evidence taken by the Departmental Committee on the use of preservatives and colouring matters in food contains only one very brief reference to fluorides, and that of a negative character, one witness being asked whether he had met with any article of food containing fluorides, the reply being that he had not met with them, but had heard of their use. During the last few months there appeared on the London markets a good deal of Brittany butter which either did not contain any boric preservative at all (while174 TEE ANALYST. the Brittany practice had hitherto been to add from 0.4 to 0.5 per cent.of boric acid), or only unusually small quantities ; yet these butters possessed the keeping qualities of butter containing preservatives. Our attention was directed to these butters by a firm of the highest standing in the butter trade. After searching for salicylic, benzoic, and sulphurous acids, with negative results, we succeeded in obtaining in a few, but by no means all such instances, distinct evidence of the presence of a soluble fluoride. Fifty grammes of the butter were melted in a platinum basin, the aqueous layer separated from the fat, rendered slightly alkaline with sodium hydroxide, evaporated, the residue ignited, transferred to a platinum crucible, strong suIphuric acid added, and the crucible covered with a waxed slip of glass upon which a distinguishing mark had been scratched with a pin.The crucible and contents were gently heated upon a sand-bath for two hours, and the glass siip examined after the removal of the wax. We found that 1 milligramme of calcium or sodium fluoride showed distinct etching of the glass. Instead of melting the butter in a platinum basin, the melting out can be done equally well in glass, but a little alkali should be added. It soon became apparent that in the presence of boric preservative fluoride could not thus be discovered, for in presence of both strong sulphuric acid evolves boron fluoride. By the use of dilute sulphuric acid the formation of boron fluoride is avoided,and as long as the amount of boric acid is not greater than five times that of the fluoride, a reaction (etching of the glass) can be obtained.But we found it far preferable to get rid of the boron compounds before testing for fluorides. The aqueous liquor is separated from 50 grammes of the butter, and without clarifying it calcium chloride is added, the liquid heated to boiling, and a small excess of sodium carbonate used to precipitate the calcium compounds. The precipitate consists of calcium borate, fluoride, carbonate, with some phosphate and perhaps sulphate. It is filtered off, burnt and treated with hot dilute acetic acid ; this dissolves out the carbonate, borate and phasphate. The residue is again collected on a filter, ignited, and treated with strong sulphuric acid. Distinct etching of the glass results in presence of fluoride.We have thus often been able to get fluorine reactions when, without separation of borate, the tests were quite negative. We also endeavoured to make this test a quantitative one, though with no conspicuous success. The final precipitate containing the calcium fluoride was weighed after the etching process, the sulphuric acid was driven off, the residue strongly ignited, and the weight of the resulting calcium sulphate again ascertained. Since calcium fluoride, with its molecular weight of 78 yields 136 parts of calcium sulphate, the increase of weight should furnish the measure of the amount of fluorine. But even in cases where no etching could be observed, and where we must presume fluorides to have been absent, there was a distinct residue left after the acetic acid treatment, and this, in most cases, slightly increased in weight after being converted into sulphate. It is pro- bable that the ignited calcium phosphate was rather more insoluble in the dilute acetic acid than was expected, and tricalcium phosphate, on ignition with atrong sulphuric acid, we have satisfied ourselves, gains notably in weight, the ignited residue corresponding in composition with a mixture of (calcium sulphate) and calcium pyrophosphate with a little metaphosphate. But when a fluorine reactionTRE ANALYST.175 was obtained, the increase in weight was nearly always materially higher than in the negative cases. The increase in weight corresponded from about 0.04 to 0.08 per cent. of sodium fiuoride. This amount is probably overestimated by about 0.02 per cent., so that the actual quantity of sodium fluoride present is from 1 to 4 grains per pound. These figures make no pretence to accuracy, but cannot be far off the truth. This, on analysis, was found to contain about 98 per cent.of sodium fluoride, the remainder being moisture. A strong trace of lead, doubtless due to the use of lead vessels in the preparation of the hydrofluoric acid, was also present. With the precipitate obtained from some of the samples of butter several glass plates in succession could be strongly etched. We have so far found fluoride in twenty samples of butter; in ten saniples there was no boric preservative; while in ten others this was also present in amounts varying from 0.18 per cent. to 0.36, or from 12.6 to 25.2 grains per pound.We have also searched for fluoride in a few samples from other countries in which circumstances appeared to render the presence of some preservative other than boric likely, but with negative results. So far the use of fluoride seems to be confined to certain French butters. As traces of fluorine must be present, and are stated to have been found, in milk-, we tested the ash of 50 grammes of milk, but with quite negative results. The reactions obtained with 50 grammes of butter must therefore be due to the added material. The Preservatives Committee of the Local Government Board recommended that borate be the only permitted preservative in butter and margarine. As, however, it might be argued that no evidence as to the use of fluorides in butter was before the Committee, the recommendation in question might not apply to that preservative, it becomes important to inquire into the physiological action of fluorides when taken in the quantities that might be consumed in butter.Taking the ma.ximum amount we have met with-namely, about 4 grains to the pound-it would follow that in 2 ounces of butter, which a person might very easily consume, 8 grain of sodium fluoride would be contained. We take from the American Dispensatory (17th edition, p. 3634) the following statements : ‘‘ Waddell states that the alkaline fluorides are not pronounced irritants, and when taken internally in doses of a grain to a grain and tl half continuously they reduce the force and the frequency of the pulse, at the sanie time depressing the temperature and increasing somewhat the flow of urine, but not distinctly affecting either the respiratory or the cutaneous functions.Th.is accords with the physiological studies of Tappeiner, who found in animals the soda-salt to powerfully depress blood-pressure by acting on the vaso-motor centres. - Death, after profound collapse, was produced by centric failure of respiration. Dr. Waddell also affirms that there is an enormous decrease in the number of red blood corpuscles, which he believes, but does not prove, to be the result of a direct action upon the spleen .” These statements raise, to say the least, a presumption that the quantities of fluoride that are used as butter preservatives are not without injurious action upon the living body. A number of experiments we have made with digestive ferments We obtained some of the fluoric butter preservative from Brittany.176 THE ANALYST. show clearly that fluorides have, in some cases at least, a powerful iuhibitory action.As regards some fermentative actions this is already well known. Thus, in Oppen- heimer’s book on I ‘ Ferments,” translated by Mr. C. A. Mitchell, we find the statement that sodium fluoride prevents the action of rennet (p. 144), and that fluorides have a marked injurious effect upon the zymase obtained by Buchner from yeast (p. 242). In EfTront’s method for the suppression of bacterial organisms in the manufacture of alcohol, it has been found (Maerker, ‘‘ Das Fluss-saeureverfahren in der Spiritiis- fabrikation,” Berlin, 1891) that as little as 10 grammes of ammonium fluoride in 100 litres of water may injure the germinating power of barley, and that more than 3 grammes in 100 litres may interfere with the diastatic power of malt.Now, since butter with about 13 per cent. of water was found to contain from 0.02 to 0.06 per cent. of fluoride, calculated as the sodium salt, the samples contained an aqueous solution of from 0-16 to 0.46 per cent. strength, which solution inhibits the growth of micro-organisms in the butter. In the experiments which we are about to give a more dilute solution was employed in every case. Ptyalin Experiments.-Ten grammes of air dry arrowroot were gelatinized with hot water and the bulk made up to 250 C.C. ; 5 C.C. of sublingual saliva were added when the solution had a, temperature of 38” C.After one hour at that temperature iodine gave only a brown coloration. The cupric reducing power corresponded to 42.63 per cent. of maltose calculated on the original starch taken. A precisely similar experiment was made at the same time, 0.1 gramme of ammonium fluoride (containing 55.8 per cent. of fluorine) being added after the arrowroot had been gelatinized. After an hour’s digestion iodine gave a strong blue colour ; the solution had no cupric reducing power. A similar experiment was made with an equivalent quantity of sodium fluoride. After one hour’s digestion the solution gave a blue colour with iodine and did not reduce Fehling solution. I t follows that salivary action is prevented by a solution containing 0.04 per cent. of sodium fluoride, or its equivalent in ammonium fluoride.Peptic Digestion.-Three lots of 50 grammee each of hard-boiled white of egg, which had been rubbed through a fine sieve, were heated to 40” C. with 250 C.C. of a 0.2 per cent. HC1 solution containing enough pepsin to dissolve the whole of the white of egg. One lot did not receive any fluoride the second 0.1 gramme, and the third 0.05 of ammonium fluoride. After five hours the whole of the white of egg iu solution 1 had dissolved, the amount of egg had not visibly diminished in solution 2, while a very large proportion was left undissolved in sample 3. It is clew that as little as 0.02 per cent. solutions of fluoride greatly interfere with peptic digestion. The difference in the appearance of the contents of the test- bottles was so striking that quantitative experiments seemed unnecessary.Rennet Expeyiments.--Contrary to the statement in Oppenheimer’s work, quoted above, we could not perceive any interfering action of a small amount of fluoride upon the milk-curdling action of a solution of rennet. Nor was such interfering action to be expeoted with quantities of fluoride smaller than amounts entirely precipitable by the calcium salts contained in the milk. As soon, however, as the calcium equivalent was overstepped, the fluoride stopped the rennet action.T€IE ANALYST. 177 Diastatic Action.-The proportions of starch (arrowroot) and water were the same as in the saliva experiments, but instead of the saliva 25 C.C. of a cold water infusion of malt were. used, the temperature being maintained at 65" C.for one hour. The iodine reaction after that time was markedly different in the fluids with and without fluoride ; in the former case only a darkening in the colour was the result, while in the latter a strong erythrodextrine reaction was always observed. The reducing power, calculated in percentages of maltose upon the weight of the arrowroot taken, was found to be as follows : Without With ammonium fluoride. ammonium fluoride. 67.4 ... ... ... 57.4 62.1 ... ... ... 45.0 At the same time, it appears that the injurious action of fluorides upon the salivary starch- solvent action is far more pronounced than in the case of the corresponding malt- diastatic effect. From some qualitative tests made by us it also appears that the inhibitive influence of fluorides upon the ferment, which forms mustard oil from myrosin, is but slight.Against the use of fluorides in the manufacture of alcohol, as in Effront's process, no objection can be raised from a sanitary point of view, because the chemical does not enter into the final product. The case is far otherwise when the fluoride is directly admixed with food materials and is consumed with them. At present the use of fluoride, as far as our experience goes, appears to be confined to a single country-France. We hope that our remarks may be the means to arrest the further spread of its employment. We may further take this opportunity to express the hope that the President of the Local Government Board will no longer delay the issue of regulations concerning the use of preservatives, the Committee's report having been presented nearly six months ago.These figures bear out Maerker's warning previously referred to. DISCUSSION. Mr. LLOYD said that about fifteen years ago he had been told by the agent of one of the colonies that a means of preserving butter had been discovered which could not be detected by analysis, and, after much trouble, he had found that fluorides were being used. He had reported that fluorides were far more injurious thsn some other preservatives, and, as far as he knew, they were given up in favour of boracic acid. Last year, however, he had been informed that in Brittany fluorides, in the form of sodium fluoride, were again being used, and he regretted to say that, as soon as the possibility of the prohibition of the use of preservatives in England became known, the material was introduced on this side of the Channel.Moreover, he was afraid that its use might not be confined t o butter. The Preservatives Com- mittee had recommended that the use of 4 per cent. of boracic acid should be permitted in the case of butter-a substance which had been washed free from the constituents causing decomposition-and that 4 per cent. should be allowed in the case of cream, which was much more difficult to preserve than butter, for the178 THE ANALYST. substances liable to decomposition, instead of being washed out, were all contained in it. So far as he knew, sodium fluoride was being used-or, at any rate, was ready to be used-not merely for butter, but for milk and cream.He thought that a debt of gratitude was due to the authors of this paper for having described so clearly a method by which it could be detected. It had taken him a very long time, in years gone by, to detect the preservative in the colonial butter he had referred to, and if he had not been perfectly certain that gome preservative was present, he might have considered the search hopeless, and have said that no preservative was present. I n regard to the statement that the action of fluorides upon rennet was only marked when sufficient was present to neutralize or combine with the lime present in the sample of milk experimented with, he would point out that the action of rennet was a time action, and the question was not whether the action might be actually stopped, but whether it might not be delayed; and he would like to hear whether the authors had made any experiments which would show whether the action of rennet was delayed to any extent by quantities of fluorides too small to stop it altogether, for he fancied that the digestion of milk in the infant must be to some extent similar to the action of rennet, and therefore that a substance added in small quantities, if it delayed the action of rennet, would probably also delay diges- tion in the infant, and would, on that account, be injurious, although it might not actually prohibit entirely the action of rennet.Mr. ALLEN said that probably it was within the knowledge of the authors and of others present that tihe use of silico-fluoride of sodium as a preservative had been patented, and that the compound was manufactured to a considerable extent at Warrington.It was interesting, seeing that the association of boric acid with fluorides had been men- tioned, to note that boron compounds also were manufactured largely at Warrington. He would like to hear if the authors had any suspicion that the two were used in con- junction. Possibly the boron was in the form of boro-fluoride of sodium (NaBFl,); or borax and hydrofluoric acid might be used, and would probably form B compound (HBF1,) eminently suitable for the purpose in question. Mr. RICHMOND mentioned that it had been proposed to use fluo-boracic acid as a preservative, and some years ago he had examined a sample which contained it. He did not think, however, that it had ever been extensively used.Mr. BRIANT said that sodium fluoride had been patented some five or six yews ago as a preservative for beer, but in that case, again, no practical application had followed. Mr. FISHER said that he remembered that some years ago potassium boro- fluoride had been recommended as a strong and powerful antiseptic and an excellent preservative, and he had made a few experiments with it at the time, but he had never met with it professionally, and did not know that it had come into use. Mr. 0. HEHNER, in reply, said that their experiments had been confined to ascertaining the presence of fluorides, and he could not say whether boro-fluoride had been also used or not. In many cases boro-fluoride wm certainly absent, for they had found fluoride in the absence of boron. He doubted whether at the present The use of sodium fluoride had probably grown out of it.THE ANALYST. 179 time any Australian butter contained fluorides. Their experiments in regard to the action of rennet were quite qualitative. The results obtained had been that without fluorides coagulation took place in a very short time at blood heat, while after the addition of fluorides the milk remained fluid for a long time. The inhibitive action of the fluoride was only apparent after a certain quantity had been added. The first few drops of the solution, which made no difference, were, he surmised-though he had not sctually determined whether such was the case-necessary for the precipitation of the calcium. Their remarks had been intended mainly for the nformation of analysts, but partly also for the information of the Board of Agriculture. He presumed that the limitations recommended by the Departmental Committee on Food Preservatives could only be regarded as extending to those preservatives which had been referred to in the course of the Committee’s inquiry. Only such substances could be prohibited or permitted as were known to be used at the time of the inquiry. As, however, formal regulations had not yet been issued, there might be still time to extend the prohibition to fluorides, which, in his opinion, were n j urious .