196 COAL-TAR COLORS USED IN FOOD PRODUCTS. 



With these data available, and with the general rule stated at the 

 opening of this section in mind, it was regarded as the proper course, 

 when determining upon standards, to require that each color should 

 pass the combined heavy metals and arsenic test of the United States 

 Pharmacopoeia, and in other respects should be as clean and as high- 

 grade as the best of each class given in the preceding analyses. Just 

 how closely it was possible to adhere to this rule will be shown 

 later. 



With respect to the arsenic requirement, there has been a great 

 deal of discussion brought on by those interested in the manufacture 

 of these colors. It was protested that colors could not be made 

 uniformly and continuously under manufacturing conditions, which 

 would contain an amount of arsenic so small that a quantity of the 

 product containing 2 grams of the coloring matter in question would 

 not respond to the Gutzeit test in the United States Pharmacopoeia. 

 The results above given with respect to conformity or nonconformity 

 to the Gutzeit test, were, with a great deal of justification, not regarded 

 as conclusive, because, as has been previously stated, there was no 

 certainty that the amount of coloring matter actually taken did 

 represent 2 grams. In view of the fact that the United States Phar- 

 macopoeia prescribes for the only coal-tar color in it, namely, Methy- 

 lene Blue, that it shall be so free from arsenic that 2 grams fail to 

 respond to the Gutzeit test, it was considered perfectly justifiable 

 to adhere to that requirement until it could be conclusively shown 

 that it was unreasonable and incapable of attainment. Results 

 described in the pages that follow show that it has been possible to 

 make all the 7 permitted colors of Food Inspection Decision No. 76 

 so free from arsenic that they comply with the Pharmacopoeia! 

 test. 



With respect to the heavy metals, no deviation was necessary for 

 the nonferrous metals from the test of the Pharmacopoeia; for iron, 

 however, it was found necessary to increase the limit to substantially 

 0.005 part per 100 of actual coloring matter. The reason for this is 

 that at one time or another all of the seven permitted coloring matters 

 in the course of their manufacture come in contact with, or are con- 

 tained in, vessels of iron, and it seems to be almost impossible to 

 keep iron out to an extent which would bring the color within the 

 pharmacopoeial test. It has been shown that there were 13 colors 

 on the market in 1907 which contained so little iron that they failed 

 to respond to the heavy metals test of the Pharmacopoeia in that 

 respect; but here again the same criticism holds good as in the arsenic 

 test, that there is no certainty that the amounts taken for the heavy 

 metals tests were equivalent in each case to 1 gram of actual coloring 

 matter, and subsequent experience seems to confirm the correctness 



