482 JOURNAL OF COSMETIC SCIENCE conditioners reduce end wrapping and the abrasive damage that occurs to the fiber-ends and in that manner they produce a decrease in short segment breakage. MECHANICAL COMBING Mechanical combing of hair produced unexpected results. We tried to simulate the conditions for hand combing by using the same size hair tresses and the same length of comb stroke as for hand combing. This was a mistake, because as the results show, the hand combing was obviously more vigorous, probably in part due to thumb pressure on the hair to hold it snugly in the comb during hand combing and most likely a faster comb stroke. This hand combing technique was developed to maximize reproducibility rather than to mimic combing hair on the head, however, the comber and technique were tested during the late 1980's and 1990's in a series of consumer tests in Asia, Europe and Latin America where a range of conditioning products were tested for combability/conditioning and the consumer response was in complete agreement with the combing results. The data of Table IV show that for many more comb strokes there were fewer broken hair fragments by mechanical combing than by hand combing. Even though the results show similar effects numerically, because of the large variance for mechanical combing the data of Table 4 are not statistically significant. We intend to repeat these results after making adjustments to the mechanical combing to produce a larger number of broken hairs. WET VERSUS DRY COMBING AND BREAKAGE Wet combing produced interesting and contrasting results compared to dry combing. For example, a larger number of long segment breaks and fewer short segment breaks were produced by wet combing, see the data of Table V. For wet combing, clumping of hairs was readily observed and the tip ends of the fibers were not as free to produce end wrapping. This clumping occurs by a capillary action producing fewer short segment breaks however, crossed hair interactions occur higher up in the tress & because of higher friction more severe snags arise higher up in the tress than in dry combing, producing a larger number of long segment breaks. This snagging higher up in the tress was also readily observed during combing. Table IV Mechanical Combing of Hair Tresses* Number of broken hairs at length (cm) (average of 4 replicas) Non-bleaching+shampoo No. of strokes 2.54 2.54 1000 25.3 0 2000 25 0 * Dry combing of hair at 65% RH. Bleach+shampoo 2.54 2.54 43.5 13.3 44.3 6 Bleach+conditioner 2.54 2.54 19.8 2 20 0 No significant treatment effect, p = 0.62. Significant short vs long segment breakage, p 0.001. No significant comb stroke effect, p = 0.66.

2006 TRI/PRINCETON CONFERENCE 483 Table V Wet Versus Dry Combing of Hair Number of broken hairs at length (cm) (average of 3 replicas) Wet combing data Dry combing data No. of strokes 2.54 2.54 25 16.7 22.7 50 25.7 19.7 75 22.7 23.7 100 23.7 24.3 Short segments only, wet vs dry is significant, p 0.0001. Long segments only, wet vs dry is significant, p 0.0001. No significant comb stroke effect for wet short or long segments. HOW HAIR FIBERS FORM SNAGS AND THE INFLUENCE ON BREAKAGE 2.54 2.54 28 6 44.3 6.7 45.7 6.3 68.7 5.8 Prior to combing, hairs exist in complex interwoven patterns. For dry combing, as the comb descends through the hair, hairs above it are made parallel and those beneath the comb are either made parallel or knotted by, hairs looping around other hairs (5), or hairs looping around comb teeth and other hairs (1) several cm between the comb and the distal tips of the hair. As the comb advances through the looped/knotted hairs long breaks can occur or as the comb descends near the tips wrapped ends (1) can result. End wrapping by inertia & possibly static charge produces short segment breaks which are more numerous if the hair is cut at 90 degrees versus a tapered cut. For wet combing, clumping of hairs by a capillary action produces fewer short segment breaks, by reducing end wrapping however, crossed hair interactions occur higher up in the tress and because of higher friction produces more severe snags higher up in the tress resulting in a larger number of long segment breaks. SUMMARY AND CONCLUSIONS A recent publication (1), provided evidence for two types of hair breakage during combing, short segment breakage (approximately less than 1.27 cm) and longer segment breakage. We have confirmed these results and refined the separation distance between short and long segment breakage at about 2.54 cm. Furthermore, chemical bleaching increased both short and long segment breakage while a commercial hair conditioner decreased both types of breakage. Whether the hair is chemically bleached or conditioned, short segment breakage in- creases with increasing comb strokes, that is, short segment breakage increases as comb- ing damages the ends of the hair, however, long segment breakage does not increase with increasing comb strokes. For wet combing, clumping of hairs by a capillary action produces fewer short segment breaks, by reducing end wrapping however, snags form higher up in the tress because of higher friction producing a larger number of long segment breaks. Mechanical combing of tresses shows similar qualitative results, however the variance