THE ANALYST. 121 LOCAL GOVERNMENT BOARD REPORT. On the Presence of Tin in Certain Canned Goods. G. S. Buchanan and S. B. Schryver. (Reports of Inspector of Foods, No. 7, 1908, 1-30.)-1n July, 1906, the attention of the Local Government Board was directed to the condition of certain consignments of old tins of meat essence, which had been returned from the South African campaign. These were subjected to a detailed chemical examination, and, whilst no appreciable changes had taken place in the organic contents of tins, the sarriples were found to be highly contaminated with tin. Since that date a large number of samples of other tinned foods have been examined, and the results of the investigation are embodied in this Report. In order to ascertain whether there is a, liability of absorption of tin when com- paratively non-irritant, organic compounds are given by the mouth, Dr.Schryver himself for three weeks took quantities of a tin salt (sodium stannic tartrate) equiva- lent, during the first week, to 1 grain of tin per day; in the second week the quantity was increased to 2 grains of tin, and during the third week to 3 grains of tin per day. The quantity of tin excreted in the urine and fmes wag estimated, and from the results given it is seen that during the first fortnight there was no evidence of accumulation of tin in the system ; in the second week a small quantity of tin was found in the urine, showing that some of the metal had been absorbed from the alimentary tract. When 3 grains of tin were ingested daily, there was evidence of accumulation in the body, the amount being about 2 grains of tin during the week.No ill-effects were felt as a consequence of the doses taken. Whilst the urine con- tained but relatively small quantities of tin, experiments on dogs showed that when tin enters the body by a channel other than the alimentary tract, the metal is excreted in larger quantities in the urine than in the fmes. The canned goods examined consisted of flesh foods, fish, meat essences, extracts and soups, fruits, vegetables, puddings, and jams, the majority of the samples having been in the tins for from five to eight years. The meat samples appeared to have taken up into their substance a comparatively small proportion of tin, the quantity being generally less than 0.5 grain per pound ; only in one instance (canned tripe) did it reach 2 grains per pound.Meat extracts of similar age had absorbed larger quantities of the metal, 3.3 and 5.3 grains per pound being found in two samples respectipely ; the same was the case with the meat essences, but to a less degree. One sample of beef essence, however, contained 21.8 grains of tin per pound; this sample contained beads of solder, and it was found that a very conspicuous con- tamination by tin may take place in a short time if pieces of solder find their way into the contents of the can. Contamination by tin in this way is the more serious, as it is accompanied by risk of simultaneous contamination by lead. Only one of the three samples of soup examined contained tin in large quantities; in this instance the soup contained fruit acid derived from tomatoes, and was contaminated with 3.5 grains of tin to the pound.Canned lobster had absorbed as much as 2-4 grains of tin per pound, whilst two samples of canned vegetables (parsnips and122 THE ANALYST. carrots) showed 1.5 and 2.2 grains of tin per pound respectively. Apple and plum puddings and a sample of apricot jam contained about 3 grains of tin per pound. No samples of canned fruits most commonly consumed in this country were included in the South African samples, but samples of canned pears, apricots, and peaches, from a London importer, who specially selected them as old stock and stated them to be one or two years old, contained not more than 1 grain of tin per pound. The investigation brought out clearly that the tin, after solution in the liquid contents of the can, becomes in course of time absorbed in, or chemically combined with, the solid contents (meat, fruit, etc.) in such a way that the latter contain comparatively larger quantities than the liquid itself.Commenting on the results o€ the investigation, Dr. Buchanan says that they suggest that in the case of food contaminated with tin the larger portion of the tin does not become absorbed from the gastro-intestinal tract, and that accumulation of tin in the body in these circumstances takes place at a slow rate. Whilst it may be concluded that there is not a serious risk of chronic poisoning by the absorption of relatively small quantities of tin as a result of a diet which consists largely of canned foods continued over a lengthy period of time, it still remains necessary to consider the irritant action on the gastro-intestinal tract, which may result from the presence of considerable quantities of tin salts in the food.It appears to be well established (a comprehensive account of alleged cases of tin-poisoning and researches on the toxicity of tin by other workers is given in the Report) that such irritant action may take place, and the condition of the stomach, the susceptibility of the individual, and other circumstances must be taken into account. On the whole, it seems clear that in any kind of canned foods quantities of tin approximating to 2 grains to the pound are not only unusual and unnecessary, but also must be regarded with grave suspicion in consequence of the risk of irritant action of the tin they contain.I t is suggested that special attention should be directed by sanitary officers, with the co-operation of the Public Analyst, to canned foods which there is reason to suspect to be of unusual age--e.g., more than one or two years-and in particular to such canned goods as those in which the liability to act on tin is specially great. The presence of more than 2 grains of tin to the pound may be taken to signify that the food has become potentially dangerous to health, and calls for further examination of other samples with a view to dealing with the whole consignment in accordance with the results obtained. I t would be advantageous to be able to ascertain, from marks or labels on the cans, the date and place of preparation of canned goods.For the estimation of tin two processes were used, one colorimetric and the other gravimetric. I n both methods the organic matter of the sample was destroyed by hea,ting with sulphuric acid ; in most cases 50 grams of the material were employed for the estimation, this quantity being heated in a round-bottomed Jena flask with 50 grams of potassium hydrogen sulphate and 50 C.C. of concentrated sulphuric acid, further quantities of the acid being added, if necessary, to keep the contents of the flask in a fluid condition. When the solution had become perfectly colourless it was cooled, diluted with water to about 600 c.c., treated with hydrogen sulphide, and allowed to stand overnight. In some caseg, when large quantities of calcium salt are present in the foodstuff, calcium sulphate separates out when the acid solutionTHE ANALYST. 123 is diluted ; this must not be removed by filtration before precipitation with hydrogen sulphide, as it contains tin, possibly in the form of a double sulphate. The pre- cipitate of sulphide and sulphur was collected on a filter and washed a few times with water, and the tin then estimated either colorimetrically or gravimetrically as follows : C'olorimetric 1Method.-The filter and precipitate were transferred to a wide test- tube, boiled with 5 C.C.of concentrated hydrochloric acid, and the solution filtered through a small conical Buchner funnel, the residue being washed with 2.5 C.C. of hydrochloric acid. The test-tube was next closed with a cork carrying a tube, and carbon dioxide was passed into it.A piece of zinc foil was then added to the strongly acid solution while still hot; and as soon as the last trace of metal had dissolved, the reagent was added. The latter was prepared by dissolving 0.2 gram of dinitrodiphenylaminesulphoxide in 100 C.C. of & sodium hydroxide solution, and filtering. Two C.C. of the reagent were used in each test, and were introduced into the test-tube while the current of carbon dioxide was continued. After boiling for a few minutes, the solution was diluted to 100 C.C. with cold water, filtered, and the filtrate treated with a drop of dilute ferric chloride solution. The violet coloration produced was then compared with that yielded by known amounts of tin.The standard solutions were prepared from a solution containing such a quantity of stannous or stannic chloride that each 5 C.C. contained 14.28 milligrams of tin. This is a convenient amount, as 14-28 milligrams in 100 grams is equivalent to 1 grain per pound. One-quarter of a grain of tin per pound gives a marked coloration if 50 grams of the sample be taken for the estimation, and 0.1 grain per pound gives an appreciable colour. The colorimetric process was found to be useful for sorting purposes and for estimating less quantities of tin than 1 grain per pound, but for larger quantities the gravimetric method was preferred. The reagent used in the colorimetric process is prepared as follows : Ten parts of nitric acid of specific gravity 1-48 are mixed with an equal weight of nitric acid of specific gravity 1-40; the mixture is cooled with ice, and small quantities of thio- diphenylamine are added with continual stirring until 1 part by weight has been added altogether.The temperature must not rise above 5" C. during the addition, Aftier standing a few hours, the separated nitro-body is collected on an asbestos filter and washed, first with concentrated nitric acid, then with nitric acid of gradually diminished strength, and finally with water. I t is then extracted with hot alcohol, in which it is not appreciably soluble. Gravimetric Method.-In this method the washed precipitate of sulphide and sulphur obtained as described in the colorimetric process was dissolved in a small quantity of hot 10 per cent. sodium hydroxide solution, and the sulphide re- precipitated by the addition of glacial acetic acid. This treatment was necessary in order to remove sjlica derived from the flask. The sulphide was then collected on a filter, washed with hot water, dried, oxidised, and weighed in the form of oxide. w. P. s. Ferrosilicon and its Properties. Wilson. (Report of the Inspector of Factories, England, 1907.)-The poisonous action of certain samples of ferrosilicon is ascribed to the generation of arseniuretted and phosphorettsd hydrogen, owing124 THE ANALYST. to the action of moisture. The explosive properties of ferrosilicon are probably due to the formation of ferrous carbonate and spontaneously inflammable hydrogen silicide (SiH,), under the combined action of carbon dioxide and water. Other less probable theories attribute the explosions to the existence of the ferrosilicon in a strained condition (similar to that of Rupert’s drops), owing to rapid cooling; and to the freezing of contained water in winter. The quantity of ferrosilicon imported into England is about 4,000 tons per annum. A. G. L.