TRANSEPIDERMAL WATER LOSS 483 In the steady state, J does not depend on x, so that Jb = - D(c)dc = D(c)dc and (occluded case) (3) fCo c'b fc,8:0 J'b = - D(c)dc = D(c)dc (nonoccluded case) (4) Hence, the difference in flux is given by = i D(c)dc - D(c)dc 1 J'-J • =• D(c)dc (5) Since ca is always greater than c'a, and D(c) is greater than zero, the integral on the right-hand side will always be positive, whatever the form of D(c), which leads to the general conclusion that j, 3 j (6) i.e., the occlusive film always results in a decrease in water loss rate, as expected in- tuitively. As in (1), the assumption was made that the corneum thickness, 8, does not change as a result of increased hydration however, we know that swelling does take place. This ef- fect can only further increase the difference between J' and J. It should be noted also that the use of activities instead of concentrations in the above analysis in no way modifies the result. It is clear now where the analysis in (1) went astray. It assumes, and quite rightfully so (2), that the diffusion coefficient increases with increasing concentration of water in the corneum. However, in their analysis, the authors consider "D to be constant for each membrane, even though this may not be the case for nonuniformly hydrated stratum corneum." This basic internal inconsistency gives rise to the paradox. It is not our intention to dispute the validity of the experimental data quoted in (1). But a different mechanism would have to be found to explain such findings. Other data exist that appear to violate the straightforward diffusion theory. For example, the maximum in the relationship between TWL and ambient relative humidity, as reported by Grice et al. (3) for the in vivo situation, cannot be explained this way. However, other in vivo (4), as well as in vitro (5,6), data do show the expected continuous increase of flux with increasing RH difference across the corneum. In this connection, it may be pointed out that in vivo measurements are notoriously variable and sensitive to ex- traneous influences. CONCLUSION A truly occlusive film, when applied to skin, can only reduce the rate of TWL. When the opposite effect is found in practice, the conclusion must be that the agent applied is not simply occlusive (i.e., inert) but must interact in a more complex way with the stratum corneum.

484 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS ACKNOWLEDGMENT The authors thank Lever Brothers Company for permission to publish this paper. REFERENCES (1) E. R. Cooper and B. F. Van Duzee, Diffusion theory analysis oftransepidermal water loss through occlu- sive films, J. Soc. Cosmet. Chem., 27, 555 (1976). (2) A. F. E1-Shimi and H. M. Princen, Diffusion characteristics of water vapor in some keratins, Colloid Polym. Sci. (in press). (3) K. Grice, H. Sattar, and H. Baker, The effect of ambient humidity on transepidermal water loss, J. In- vest. Dermatol., 58, 343 (1972). (4) A. B. Goodman and A. V. Wolf, Insensible water loss from human skin as a function of ambient vapour concentration, J. Appl. Physiol., 26, 203 (1969). (5) M. M. Rieger and D. E. Deem, Skin moisturizers. I. Methods for measuring water regain, mechanical properties, and transepidermal moisture loss of stratum corneum, J. Soc. Cosmet. Chem., 25, 239 (1974). (6) L.J. Vinson, E. J. Singer, W. R. Koehler, M.D. Lehman, and T. Masurat, The nature of the epidermal barrier and some factors influencing skin permeability, Toxicol. Appl. Pharmacol., 7, 7 (1965).