340 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS as possible, then makes small changes in the concentration dials, "p," until the three meters labeled "a" indicate zero. The production control problem is equivalent to adjustment of a trial formula. A sample of the batch being processed is measured, and adjustments to the batch formula are obtained from the computer. In most production problems a colorimetric adjustment is satisfactory if appropriate procedures have been established. When the system is properly set up, adjustments can be obtained in less than ten minutes, including measuring time, and one correction is usually sufficient. SYSTEM OPERATION Several steps are required to set up and operate the color control system which has been described. Before any matching or control work can be undertaken, the pigments must be calibrated. A mixture of each pigment with white at a known ratio is made and the reflectance meas- ured these measurements are converted to KfS values and set into the pigment plug-in boxes. When a standard is to be matched, it is first measured on a spectro- photometer, and the appropriate KfS values are set into COMIC. A tentative choice of pigments is then made, and the corresponding pigment plug-in boxes are inserted into the computer. If the operator is unable to obtain a straight line on the computer oscilloscope by adjust- ing the concentration dials, he makes a different pigment selection and tries again. When a straight line or the best line possible with avail- able pigments is obtained, the operator reads the pigment formula from the concentration dials. The proposed formula is then made up and compared either visually or instrumentally with the standard. If it is not a sufficiently close color match, the reflectance values of the sample are converted to K/S values and are placed in the COMIC along with the KfS values of the standard to be matched. Alternatively, the difference in tristimulus values along with the appropriate dR/d(K/S) values may be used. A correction of the formula is then obtained. A new formula is made up, and the entire process is repeated until a satisfactory match has been obtained. Adjustments in a production batch are handled much as are adjust- ments to a trial formula. Usually tristimulus values of the batch sample are measured and placed in the computer along with the values for the standard. After nullins the three meters, the operator reads from COMIC the adjustment required to bring the batch to the standard.

INSTRUMENTATION IN COSMETIC COLOR CONTROL 341 APPLICATION TO COSMETIC PROBLEMS The example cited to show the operation of the color control com- puter is a real one chosen from some of the work which has been done on compressed cakes. There appear to be no difficulties in applying these methods to the color control of powder in this form. The same holds true for liquid make-up, although the best method of preparing a sample must be established. Multilayer draw-downs* have been found to be very effective on the few problems studied the results were excellent, even with measurements made on the liquid make-up in a glass cell. In the case of lipsticks and nail polishes, there are problems of compu- tation on shades containing very little white, just as in paints or plas- tics. There is also the problem of establishing a satisfactory method of sample preparation for lipsticks. These problems may make it im- practical to apply these methods to the very brilliant shades, but the more pastel shades should not prove difficult. The success of these methods in the control of hair dye has already been established by one manufacturer who has been using them for this purpose for about two years. CONCLUSIONS On the basis of work done on cosmetic problems, it can be predicted that the use of instruments for color control will be as successful in cosmetics as it has been in paints, plastics, and textiles. Certainly not all color problems will be solved by instrumentation, although a suffi- ciently large portion will be solved to more than justify the use of these methods. In the future further improvements in both techniques and instruments can be expected, but the general methods described here have been so successful that we may be sure they are here to stay. (Received November 15, 1965) REFERENCES (1) Judd, D. B., and Wysecki, G., Color in Business, Science, and Industry, 2nd ed. page 387 if, John Wiley & Sons, Inc., New York, 1963. (2) MacAdam, D. L., J. Opt. Soc. Am., 32,247 (1942). (3) Davidson, H. R., and Hanlon, J. J., Ibid., 45,617 (1955). (4) Davidson, H. R., Hemmendinger, H., and Landry, J. L. R., J. Soc. Dyers Colourists, 79, 577-589 and 601-603 (1963). * A Bird Applicator normally used for paint draw-downs was used, but we found it neces- sary to allow the first coating to dry, then place a second on top of it.