388 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS WDi 15 I0 I--I Normal { n= 7) K'q Medium-dry (n= •3 ) Dry (n= 7) 1.0 1.5-3.5 4-5 Average Dryness Scores Figure 3. WDi determined electrometrically in skin with different degrees of dryness evaluated subjectively ! = absent, 1.5-3.5 = slight to moderate, 4-5 = intense. The bars represent standard error. scores for the dry skin subjects were less consistent. This was to be expected in view of the great complexity involved in recognizing and quantifying parameters associated with so-called dry skin (5, 10). Despite this, the regression coefficient for the dry group is high enough (0.7) to suggest that different degrees of dryness were detectable within this population. Figure 3 shows the relationship between the WDi values and the subjective skin dryness. To simplify analysis, the dry skin group is presented as two categories, medium dry and dry, based on the scores ranging from 1.5 to 3.5 and 4 to 5, respectively. The numerical differences in the figure indicate a direct relationship between the WDi values and the degree of superficial skin dryness. However, statistical analyses of data (Table I) show that the differences were significant (P = 0.05) only when comparing the normal group to either the medium dry or dry group. A typical pattern of C readings obtained from two subjects, one of whom had dry skin, is illustrated in Figure 4. This shows that occlusion causes a gradual increase in capacitance associated with a drop in impedance. Immediately following the occlusion with petrolatum (Figure 4 insert) a temporary increase in Z and decrease in C can be Table I Statistical Comparison Between Dry, Medium Dry, and Normal Skin Evaluated Electrometrically Geometric Mean Ratio Comparison of WDi Values P Value 95% Confidence Interval Dry/Normal 2.47 .016 1.20, 5.08 Medium/Normal 1.97 .036 1.05, 3.71 Dry/Medium 1.25 .528 0.61, 2.57

ELECTROMETRIC ASSESSMENT OF SKIN DRYNESS 389 SECTION A 0.80[ 0.32[ .'_ --•ø• ø'•ø'•'"'"'ø'--'--o o o •0 -0.64 I .10 I- ß I ß Capacitance I I• '•" •:'• ' .•:e I I I i i i io 2o 3o Minutes After Occlusion SECT/ON 0.80 - 0.52 y 0.00 -0.64 -o • ß Capacitance o•. o Impedance (100 Hz) -i .60 I i i i0 20 30 Minutes After Occlusion Figure 4. Changes in skin capacitance and impedance measured in normal (A) and dry skin (B) after occlusion with a thin film of white petrolatum. T = petrolatum treated, U = pre-test control measurement. measured. The latter effects are consistent with the low dielectric constant of this occlusive material. Both the increase in C and the decrease in Z which are observed after a brief time following occlusion have been interpreted (1,2,3) as being related to an increase in skin surface hydration. The increase in C with hydration (which may appear paradoxical) can be explained based on experiments of Clar [cited in (7)] who found that the dielectric constant (K) of