416 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Table IIIb (cont'd) Intensity Ratings of "Body" Shirts Wonhen Men W q- M Men's shirts Mean -+ SD Mean + SD Mean -+ SD Day 1 1 3.29 t.38 3.27 1.39 3.28 1.36 2 3.86 1.29 3.60 1.50 3.72 1.39 3 4.93 1.54 4.87 1.64 4.90 1.57 4 5.07 1.44 5.00 1.46 5.03 1.43 5 3.50 1.70 4.40 1.88 3.97 1.86 6 4.20 1.86 3.27 1.79 3.73 1.86 7 4.93 1.58 3.53 1.88 4.23 1.85 8 5.20 1.78 4.47 1.96 4.83 1.88 9 3.07 1.75 3.13 1.46 3.10 1.58 10 3.33 1.88 2.93 1.58 3.13 1.72 Day 2 11 4.14 1.03 3.80 1.47 12 3.43 t.45 3.13 .92 I 13 2.86 .77 3.13 .92 14 5.64 1.74 5.07 1.16 15 3.50 1.16 3.80 1.01 16 2.13 1.55 2.00 1.00 17 4.46 1.64 3.80 1.61 II 18 2.53 1.41 2.47 .99 19 2.53 1.30 2.67 1.59 20 3.40 1.76 3.47 1.77 Av. 1-20 3.55 3.39 3.97 1.27 3.28 1.19 3.00 .85 5.34 1.47 3.66 1.08 2.07 1.28 4.13 1.63 2.50 1.20 2.60 1.43 3.43 1.74 3.47 Androstenol and III, in their highest dilutions, obtained pleasantness ratings compa- rable to the shirts actually worn. II, at all concentration levels was rated lower in pleasantness than the other test substances at the highest concentration levels its mean ratings were near the low end of the pleasantness scale. Cyclopentadecanolide, at all levels, obtained ratings near the midpoint of the scale and comparable to the female T-shirts. The differences between the pleasantness ratings assigned by male and female subjects were, in general, slight. The male subjects tended to give more favorable ratings to the shirts actually worn, both by women and by men, and somewhat less unfavorable ratings to the higher concentrations of II. Table IV Significance of the Correlation Between Concentration and Perceived Intensity Substance Women Men W q- M z p z p z p Androstenol 6.05 .0001 4.49 .0001 7.40 .0001 III 2.51 .006 3.21 .0007 4.05 .000 l II 7.96 .0001 7.26 .0001 10.76 .0001 CPD 3.39 .0004 3.56 .0002 4.91 .0001

HUMAN BODY ODOR 417 Table V Pleasantness Ratings for "Solution" Shirts Women Men W q- M (n = 29) (n = 30) (n = 59) Substance Sol. 2 Mean • + SD Mean + SD Mean + SD Androstenol 1 2.99 1.37 2.98 1.27 2.98 1.31 2 3.03 1.49 3.10 1.04 3.07 1.27 3 3.63 1.17 3.41 1.01 3.52 1.08 4 4.03 1.06 3.41 0.92 3.72 1.02 5 3.93 1.09 3.60 0.84 3.77 .98 Av. 1-5 3.53 .93 3.30 .82 3.41 .88 III 1 2.85 1.39 2.79 1.21 2.82 1.29 2 2.93 1.32 2.79 1.08 2.86 1.20 3 3.71 1.00 3.60 0.98 3.66 1.00 4 3.78 1.14 3.37 1.03 3.58 1.10 5 3.72 1.11 3.73 1.18 3.73 1.13 Av. 1-5 3.40 .83 3.25 .82 3.32 .82 II 1 1.62 1.53 2.10 1.58 1.86 1.56 2 2.05 1.41 2.30 1.49 2.18 1.44 3 2.60 1.37 2.73 1.58 2.67 1.47 4 3.07 1.18 3.11 1.46 3.09 1.32 5 3.22 1.06 3.01 1.00 3.11 1.02 Av. 1-5 2.51 1.11 2.65 1.24 2.58 1.17 CPD 1 4.01 1.43 3.88 1.34 3.94 1.37 2 4.05 1.20 3.71 l. 10 3.88 1.15 3 3.87 1.11 3.72 0.93 3.79 1.00 4 3.55 0.96 3.43 1.31 3.49 1.13 5 3.87 1.07 3.69 1.04 3.78 1.06 Av. 1-5 3.86 .95 3.68 1.50 3.77 .97 0 = extremely unpleasant, 8 = extremely pleasant. For concentrations, see Table II. PERCEIVED INTENSITY AND PLEASANTNESS RATING The correlations between perceived intensity and pleasantness rating were calculated for each concentration level of each of the four test substances, and using the two-tailed Pearson correlation test on the paired mean pleasantness and intensity ratings of each individual respondent. The findings are presented in Table VII. They show clearly different patterns for the four test substances, and also for male and female respondents. With substance II, correlations are high and negative, especially at high concentration levels, both for male and female respondents. With androstenol and substance III, correlations were also negative at all concentrations, but significant only for female respondents and only at the two highest concentration levels (and, for substance III, at the lowest concentration level). With cyclopentadecanolide there seems to be no covar- iation between perceived intensity and pleasantness. With the shirts actually worn, the correlation, although negative in most cases, is significant only in isolated instances. SEX ATTRIBUTION The findings regarding sex attribution are summarized in Tables VIII and IX. Table