HAIR BREAKAGE: REPEATED GROOMING EXPERIMENTS 451 SOME COMMENTS ON THE NATURE OF BROKEN FIBER FRAGMENTS In their studies, Robbins and Kamath (6,7) focused heavily on the size of broken hair fi - bers in repeated grooming experiments, distinguishing between short ( 2.54 cm) and long ( 2.43 cm) fragments while also proposing mechanisms for their occurrence. So- called “short segment breakage” was hypothesized to occur by the “end wrapping” of fi - bers during grooming, while “long segment breakage” was proposed to occur by “impact Figure 10. Models and experimental data for repeated grooming experiments on bleached hair both with and without conditioner at 60% RH. Figure 11. Models and experimental data for repeated grooming experiments on virgin, 1´ bleached, and 3´ bleached Caucasian hair at 60% RH.

JOURNAL OF COSMETIC SCIENCE 452 loading” on looped fi bers and entanglements. In short, the two mechanisms represent different ways by which suffi ciently-high localized stress could be generated to induce breakage. It is not disputed that snags and tangles can lead to direct breakage however, we argue that fatigue theory dictates hair fi bers can (and will) break upon repeated ap- plication of forces considerably less than “the break force.” In short, there is another breakage mechanism that involves progressive propagation of fl aws within the fi ber, and it does not require the presence and occurrence of tangles. While we did not specifi cally focus on the size of the broken fi bers, top-line observations from our experiments suggested the highest proportion occurring as extremely short fragments (≤ 1-2 mm in length). In fact, the shortest of these appear as dark, dust-like specks, although microscopic investigation confi rms them to be hair pieces. Obviously, such short fragments would not be noticed by the wearer, although the other sides of the break (that on the hair which remains attached to the head) would presumably propagate equally well to form a split end. These short breaks must represent fractures at the very tips of fi bers. This observation appears instinctive, as fi ber ends are oldest, have sustained more wear and tear, have be- come most damaged, and will have increased friction with a comb or brush. Therefore, in terms of the fatiguing principles discussed here, one would expect the highest density of potentially catastrophic fl aws to be in this region. Moreover, the density of these fl aws may be expected to progressively decrease upon moving towards the root end, where the hair is younger and has received less wear and tear. Therefore, one may expect a distribu- tion in the length of broken fi bers that is commensurate with the likelihood of fl aws. Specifi cally, we may anticipate a high population of short fragments due to more fl aws being present near the tips, with progressively lower populations of longer fi bers as the result of fewer fl aws being present nearer the root. To test this theory, we performed additional experiments on 20 virgin Caucasian hair tresses at 60% relative humidity. This time both the number and the size of the broken fi bers were recorded. Weibull parameters that describe the formation of differently sized Figure 12. Models and experimental data for repeated grooming experiments on virgin medium brown Caucasian hair at 20%, 60%, and 90% RH.