338 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS unknown tension to skin in the measurement area and actually increased panelist move- ment due to discomfort. Contact between the dynamometer and the skin was achieved with double-sided tape (3M, No. 666©). A circular disk of tape of diameter 4 mm was placed on the skin and a bushing insulator (Amphenol 31-1099), a plastic disk with a hole drilled through the center, was placed on top of the tape. A wire probe (from the armature) was bent 90 ø to form a hook which fits into the bushing insulator. The possible influence of tape occlusion at the test site was ignored since the viscoelastic response appeared to be independent of time. The elasticity values for the first experiment represent averages of three traces from one site of the lower left leg. In the glycerol treatment study, the GBE measurements represent averages of two traces obtained from three sites on each leg, approximately 10, 15, and 20 cm below the knee joint, centered on an axis through the knee and ankle joints. Four fiducial marks on the calf outlining the rectangular treatment area were used with a template to place the GBE probe to within 1 cm of each of the three measurement spots (see Figure 2). No template was used to position the probe in the first study. The force coil driving amplitude was adjusted at each measurement site to provide a maximum armature displacement of only _ 0.75 mm from equilibrium in an effort to minimize the contribution of nonlinear viscoelastic response, as will be discussed later. The oscilloscope was operated in left-versus-right mode, displaying stress-strain ellipses which varied slightly with time due to panelist movement and respiration. The operator would examine a series of traces with time and save only the traces which formed closed loops with symmetric shapes. Data Analysis. An oscillating stress (force) is applied to the skin and the resulting strain (displacement) follows with a certain lag or phase shift (8). For a linear response, one writes the stress as: and the strain as o' = o' o sin (cot ß = ßo sin (cot) If the stress is written as cr = o' o cos 8 sin cot + o'o sin 8 cos cot, the elastic or shear storage modulus is defined as (21) (1) (2) (3) (T O E' = -- cos 8 ß0 and the viscous or shear loss modulus is defined as (4) (T O E" - sin 8 (5) ß0 Thus, from a plot of ß and o' vs. time, one can determine O'o, ßo, and 8. This type of analysis can be done with use of the time base generator to obtain stress and strain as a function of time.

MECHANICAL PROPERTIES OF SKIN 339 Figure 2. Placement of GBE probe for measurement on lower leg sites. a) Positioning of 4-mm double- sided adhesive tape circle by means of a template. b) GBE attached at the central site. Three of the four fiducial marks which outline the treatment site and which are used to position the template are visible.