368 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS METHODS MATERIALS Glycerol (Analar Grade) was obtained from BDH Ltd., Poole, England, and stored in tightly closed containers at room temperature until required. Simple o/w creams and lotions were prepared, based on mineral oil emulsions, in which the glycerol replaced part of the water phase. TRIAL DESIGN For the instrumental measurements, single applications of glycerol in aqueous solutions or cosmetic emulsions were applied evenly by an experienced operator using a dose of 0.05 ml per 20 cm 2 of inner forearm skin, unless otherwise stated. Appropriate con- trol-treated sites were included and readings were taken at intervals over the subsequent eight hours. Young adults of both sexes, aged between 18 and 35 years, were used. Six to eight volunteers, each offering two or three test sites per forearm, were normally found to be sufficient. All treatments and readings were carried out under conventional, double-blind conditions. Wherever possible an equivalent contralateral site was used as a within-person control, and all data were obtained from normal healthy skin. In order to minimise sweating, the ambient conditions did not exceed 20øC and 60% RH. For the photographic assessment of skin condition, 15 female volunteers, aged between 19 and 57 years and with dry skin on the backs of their hands, took part in a double- blind crossover trial. The glycerol lotion was compared with the same lotion without glycerol. Treatment was at least twice daily, ad lib, home-use. Each phase of the trial was of two weeks duration, with one week without treatment between the two halves of the crossover. MEASUREMENT OF THE RATE OF TRANSEPIDERMAL WATER LOSS (TEWL) A Servomed EP1 Evaporimeter (Servomed AB, Stockholm-Viillingby, Sweden) was used to measure water flux through the skin of volunteers before and during treatment. This method is based on the estimation of the vapour pressure gradient immediately adjacent to the surface of the skin (7) and permits the surface investigated to be exposed to normal ambient air even during the short period of measurement. The instrument's detector was placed against the skin area to be tested and allowed to equilibrate for 30 seconds. Then the internally computed TEWL was recorded. The probe consists of thermistors and surface relative humidity detectors set 0.5 and 1.0 cm from the skin surface. A 5-cm tube of paper, 1 cm in diameter, was attached to the back face of the detector in order to reduce fluctuations in the reading caused by turbu- lence of the air in the vicinity of the probe. The modification allowed the operator to take frequent readings without unduly occluding the test area and generated more re- producible data than when local air movements caused readings to fluctuate. MEASUREMENT OF SKIN SURFACE TOPOGRAPHY There are a number of parameters used by engineers to define the roughness or smooth- ness of surfaces, and these can be used to measure the roughness of skin surface replicas

GLYCEROL TREATMENT OF SKIN 369 (8). Possible parameters include mean peak size, number of peaks, and total trace length (9). In this study the parameter Rtm has been chosen. Rtm is the average of the five largest "peak-to-valley" heights encountered when five consecutive 1-mm lengths of the surface plane are traversed by a stylus. On the skin surface, R•m increases with age as the major visible lines increase in depth (10) thus it is a useful guide to an individual's self-assessment of how aged his or her skin looks. A decrease in R•m indi- cates a reduction in the observed skin roughness, i.e., skin looks smoother and younger. In this work on glycerol, R•m values were measured using the Talysurf 10 Profilometer (Rank Taylor Hobson Ltd., Leicester, England) on skin surface replicas. Silicone rubber replicas of predetermined areas of the stratum corneum were made be- fore and at intervals up to 24 hours after treatment using "Silflo" (J and S Davies Ltd., London N12), a dental replicating material, at a silicone:catalyst ratio of 20:1. This produced a stable negative replica after two to five minutes. The negative replicas of the skin surface were used as moulds to cast positive epoxy resin replicas (Ciba Geigy (UK) Ltd., Duxford, Cambridge, England) for subsequent microscopic examination and sur- face profile (R•m) measurements. After mounting the positive cast in the profilometer a diamond stylus (10 Ixm tip diameter, nominal stylus force 2 mN) traversed the surface of the cast at a rate of 30 mm per minute and in a direction at right angles to that of the major lines. The profile was amplified and displayed on a chart recorder (vertical magnification X 200, hori- zontal magnification X 10). Six traces, each 6-mm long, were taken at 1-mm intervals across the replica, and the Rtm was recorded in microns (ixm). MEASUREMENT OF THE COEFFICIENT OF FRICTION (•) Immediately after wetting the skin with water there is a marked increase in drag or friction (Ix) which is easily detected by rubbing the area with the finger (! 1). In dry weather this increased drag returns to normal in four to eight minutes. The high fric- tion of moist skin in good condition is not unpleasant and is not mistaken for rough skin by touch. Figure 1. A simple electrical analogue of the skin showing resistors (R• and R2) and a capacitor (C) connected in parallel.