ANALYSES OF CERTIFIED PERMITTED COLORS. 205 



of this discrepancy been proven at the time the first analysis was 

 made certification would have been denied to the eight lots above 

 mentioned. 



These results further show the position taken early in the work 

 by several of the manufacturers that a requirement of not more 

 than 1 part of metallic arsenic in not less than 264,000 parts of 

 coal-tar color could not be complied with on a commercial scale 

 to be untenable; on this basis only 4 out of the 86 lots examined 

 would have been excluded. Further, the position of some manu- 

 facturers and dealers that the arsenic requirement ought, for prac- 

 tical manufacturing reasons, not to be more rigorous than one part 

 of metallic arsenic in 26,400 parts of color, or 1 part of arsenious 

 oxid (As 2 O 3 ) per 20,000 parts of color, is not borne out by the data. 



SUGGESTED REQUIREMENTS FOR CERTIFIED COLORS. 



Although the material embodied in this report gives a very good 

 idea of the composition and quality of substantially 30 different 

 lots of permitted colors prior to the issuance of Food Inspection 

 Decisions Nos. 76 and 77, and of 74 lots of certified colors, yet these 

 data are hardly sufficient to furnish a basis for standards with which 

 each color specimen must comply in detail. The fitness or unsuit- 

 ability of any lot has been determined by the examination of the 

 analytical data obtained thereon in the Food and Drug Inspection 

 Laboratory at New York; such examination has been applied to 

 the particular color under investigation with respect to its general 

 relationship to the results theretofore achieved. If in some minor 

 quality, as, for example, freedom from salt, the sample was not up to 

 what had been previously accomplished, but in a major quality, as 

 for example, ether extractive, it was equal to or better than what 

 had been previously accomplished, and the pharmacopoeial tests for 

 arsenic and heavy metals were satisfied with the exception of iron, 

 and the amount of iron was within the limit previously stated, 

 0.005 per cent, and the other factors showed a fairly close conformity, 

 such a defect as its salt content would not act as a bar to the pass- 

 ing of the lot; however, no matter how good a color might be in 

 respect to such determinations as ether extractive, if it failed to 

 comply with the United States Pharmacopeia requirement for 

 arsenic or for heavy metals it was not accepted. 



These arsenic results have been tabulated to show the distribution 

 of arsenic content (As 2 O 3 ); the numbers at the top are the Green 

 Table numbers; the numbers in the body of the table indicate the 

 number of specimens of the arsenic content stated in the first column ; 

 the last column shows the totals of all colors of the arsenic content 

 (As 2 O 3 ) corresponding thereto. 



