336 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS armature, which is suspended upon a highly compliant, virtually frictionless bearing of pressurized gas. These properties enable the measurement of viscoelastic response of very soft tissues under conditions of very low driving forces and displacements. Although the technique does not measure a fundamental elastic property of stratum corneum, it does provide a convenient and rapid way to determine the effect of various agents on its mechanical properties (19,20). For example, a factor of two decrease in the elastic modulus can be observed immediately after the application of water to the skin. Thus, one has evidence that the stratum corneum contributes significantly to the elastic modulus, since a rapid response would not result from the action of water on the dermis. We have applied the GBE technique to measure the mechanical properties of skin having varying degrees of dryness, and to determine the effect of humectants such as glycerol (1,2,3-propanetriol) on the mechanical properties of skin. The results of these studies are reported in the following sections. MATERIALS AND METHODS CLINICAL GRADING AND GLYCEROL TREATMENT We report data from two separate clinical experiments. In the first experiment, the outer portion of the lower part of the left leg (28 female panelists) was graded by two trained judges according to the scale in Table I prior to measurement with the GBE. Grading was an average over the entire area rather than the small site measured in- strumentally. The purpose of this first experiment was to determine how well the GBE elastic modulus correlates with visual skin grade. In a second investigation, five female panelists applied 0.6 ml of 40% glycerol solution to a 100 cm 2 area on one calf (outer aspect) and the same amount of water to the other calf. Treatments were applied once in the morning and once four hours later in the afternoon for three days, omitting the last treatment. Visual grade assessments and GBE measurements were made immediately prior to each treatment. The treatment sites were graded by a single trained judge prior to measurement with the GBE. Panelists followed their usual bathing/showering habits throughout the study. The panelists were assumed to have achieved equilibrium with the room temperature and humidity, which were 72øF and 54 --- 3%, respectively, thrbughout the study. EQUIPMENT AND MEASUREMENT PROCEDURE The electrodynamometer is basically that described by Christensen et al. (17) and Hargens (18). For in vivo measurements, the GBE head is supported in a three-dimen- Table I Leg Grading Scale ½ 0 Skin smooth and lustrous, no detectable dryness or follicle irritation 1 Skin lustrous, slight ashing visible in cracks 2 Skin dry, smooth, but without lustre, moderate ashing covering general surface area 3 Skin slightly rough overall with slight scaling high amount of ashing covering total leg area 4 Skin has moderate to high roughness, moderate scales formed with some small, fine cracks high ashing overall 5 Skin is rough with much scaling large cracks with high ashing overall l/2 grade units are allowable.

MECHANICAL PROPERTIES OF SKIN 337 sional translation stage firmly mounted to a Ralmike's portable toolmaker's bench (South Plainfield, N.J., Catalogue #085-3). This arrangement allows for maximum mechanical stability, required for accurate stress-strain measurements, and simulta- neously provides a portable yet secure work surface for mounting the electronics (Figure 1). A Tektronix FG-501A function generator, slightly modified to enable accurate adjustment of the D.C. offset using an auxillary 10-turn potentiometer, was used to drive the force coil of the GBE head at a sinusoidal frequency of 2 Hz. The other electronic components are analogous with the description by Christensen and Hargens (17,18). One major improvement is the use of a digitizing oscilloscope (Tektronix 5223 with 5B25N time base generator) for capturing force and displacement traces. These traces are transferred to a Tektronix 4051 computer equipped with a signal processor (4051R07) to enhance data analysis. Every eighth point of each trace was recorded on magnetic cartridge tape for later analysis. For instrumental measurements, panelists would lie on their side on a portable cot, and were requested to remain as still as possible. Attempts to restrain movement of the leg with straps were abandoned when it became apparent that this method applied Figure 1. In vivo GBE Apparatus. a) Ralmike's portable toolmaker's bench #085-3. b) Nitrogen tank (regulated to 5 psi). c) Tektronix 4051 computer. d) Shaevitz CAS-025 signal conditioner for LVDT. e) Tektronix 5223 digitizing oscilloscope. f) Tektronix FG-501A function generator with modified offset potentiometer. g) Control for translation parallel to GBE axis. Cable drives one axis of Ralmike's 023- 700R positioning table. h) Ralmike's #060-2-2 right-angle lead screw slide. i) GBE head mounted in a swivel lock attached to the lead screw.