242 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS 2O •,s IO IO 20 30 MINS Fig. 2. Charge leakage with time on 1 g of hair washed in a standard shampoo. Tested at 70 per cent Relative Humidity, using ten hair strokes. 25 20 15 IO - 5 - I I 0.25 0.5 0.75 I.O HAIR WT. IN GMS. Fig. 3. Variation in charge with weight of hair sample. All samples washed in a standard shampoo. Tested at 40 per cent Relative Humidity using ten hair strokes. As can be seen, the curve is a typical exponential decay curve. Although the charge leaks away fairly rapidly, its magnitude is sufficiently constant during the ten seconds required to take a reading. The general reproducibility of the results indicates that this leakage is not important in these experiments. Table 1 The mean charges generated on each of five 1 g locks of hair after identical treatment Relative Humidity 40 per cent ' 10 hair strokes per deter- mination *Mean Relative Charge Units Standard Deviation Lock A B 1 19.4 0.98 2 22.5 1.34 3 20.4 1.03 4 22.0 0.98 5 22.1 0.65 *Mean of 5 determinations Standard deviations of the means in column A 1.35 Mean of standard deviations in column B = 1.0

STUDYING STATIC ELECTRICITY PRODUCED ON HAIR BY COMBING •,43 The reproducibility of the results To obtain some indication of the reliability of the method, the charge was determined on five similar locks of hair after being given the identical shampoo treatment. The relative humidity was set at 40 per cent so that any variability would be as large as possible. The results are given in Table 1. As will be seen, the accuracy of the charge determinations is slightly greater than the sample to sample reproducibility, but both are adequate. The effect of the weight of the hair sample on the charge A number of samples of hair of different weights but identical length were given the standard shampoo treatment and the charge determined at a relative humidity of 40 per cent. The results are shown in Fig. 3. The charge was found to increase with weight, up to an optimum value of 0'75 g and then it diminished, presumably due to the decrease in the number of hair/comb points of contact, as the hair was forced into strands between the teeth of the comb. The effect of the number of strokes of the hair through the comb Half gram and gram samples of hair were given the standard shampoo treatment, and the charges produced by varying the number of strokes of the comb were determined at a relative humidity of 40 per cent. The resultq are shown in Fig. 4. 0'5 GM. HAIR n- 20 - U I0 i I I 5 IO 15 20 No, OF STROKES '!?ig. 4. Variation in charge with number of hair strokes. Two hair samples of different weight, washed in a standard shampoo and tested at 40 per cent Relative Humidity. It is not surprising that very little difference was obtained for 0.5 g