HAIR BREAKAGE: REPEATED GROOMING EXPERIMENTS 443 EXPERIMENTAL Repeated brushing experiments were performed on a custom-built unit shown in Figure 3. The device consists of a hollow rotating drum-like assembly, where four outer crossbars contain holders for mounting combs or brushes. These outer arms are detachable to allow for different holders to be mounted and experiments to be performed using a variety of combs or brushes. The four combs or brushes are mounted at 90° angles, allowing one complete drum revolution to comb (or brush) a tress four times. This entire setup is duplicated three times in the horizontal direction, allowing four tresses to be combed simultaneously. Collection plates are located under each tress to save broken fi ber frag- ments, while spacer plates on the rotating drum prevent cross contamination. The device contains a variable-speed motor, although experiments generally are performed at 50 strokes/minute. All testing was performed using standard 8-inch, 3-g Caucasian hair tresses obtained from International Hair Importers, while grooming was performed using the head por- tion of Goody Purse Style brushes. All experiments were performed under controlled humidity conditions via the use of a climate-controlled room or a bench-top chamber. On the surface, the performing of such tests would seem rather straightforward, but, in actuality, there are many experimental factors that can contribute to variability and infl uence the outcome. Therefore, when reporting results, it is important to describe conditions as fully as possible. For convenience, these can be categorized into four groups: hair factors, brush/comb factors, environmental factors, and product-usage factors. (I) HAIR FACTORS The propensity for hair breakage increases signifi cantly with hair damage. Therefore, the more damaged the hair, the higher the number of broken fi bers. While there may Figure 2. Survival probability curve for virgin Caucasian hair fi bers exposed to a repeated 0.010–0.011 g/mm2 stress at 60% relative humidity.

JOURNAL OF COSMETIC SCIENCE 444 be multiple sources of hair damage, the easiest way to generate reproducibly in the lab involves the use of chemical treatments. It is recognized that a high proportion of US women utilize some form of chemical treatment on their hair, and so employing such conditions can be considered to represent the norm rather than the exception. The shape and size of tresses would also be expected to have an infl uence, with higher grooming forces being encountered when combing or brushing a thicker mass of hair. In addition, fi ber dimensions and the degree of curvature may also be expected to alter the grooming forces. (II) BRUSH/COMBING FACTORS Grooming forces will be also be dependent on the nature of the comb or brush, with, most obviously, the spacing between teeth or bristles having an infl uence. However, in principle, different comb or bristle materials may also have a different tendency for abra- sion. There is also the need to consider the quality of the combs and brushes—broken or bent teeth and bristles may lead to increased snagging. (III) ENVIRONMENTAL FACTORS Single-fi ber fatigue results suggested substantial differences in breakage as a function of the relative humidity (8). This would seem to stem from changes in the plasticity of the fi bers and the subsequent infl uence on fl aw propagation. However, when dealing with hair arrays in grooming experiments, there is the need to consider additional factors, such as static fl yaway at low humidity and potentially increased friction and snagging at ele- vated humidity due to fi ber swelling. Figure 3. TRI’s custom-built repeat-combing/brushing apparatus.