52 JOURNAL OF COSMETIC SCIENCE ml was used. The average recovery from five samples was 92.0%, with a 0.7% CV. This data indicates that the assay method was reproducible. Precision and accuracy were evaluated by determining the MP concentration of six samples of the emulsion to which 0.5% to 3.0% additional MP was added in 0.5% increments. A linear relationship (R 2 = 0.9597) (Figure 3) between the theoretical MP concentration and the area ratio of the MP and BP peaks was obtained. In addition, analysis of a blank emulsion did not produce a peak corresponding to MP. Thus, it was concluded that the SPE extraction and HPLC analysis of MP in o/w emulsions is free of interference. CONCLUSIONS The use of the Oasis HLB cartridge (1 cc) effectively separated MP from other compo- nents of an o/w emulsion. This technique was simple, rapid, precise, and accurate, used a small volume of solvent, and required no drying stage. The determination of the amount of MP by SPE followed by HPLC analysis yielded well-resolved peaks of MP and the corresponding internal standard (BP) with excellent recovery (92.0%) and good precision (0.7%). ACKNOWLEDGMENTS This study was supported in part by the NSF Industry/University Cooperative Research Center in Pharmaceutical Processing. The authors gratefully acknowledge the technical assistance provided by Mr. Allen Anderson of Waters Corporation. REFERENCES (1) A. Wade and P.J. Weller, Handbook of Pharmaceutical Excipients, 2nd ed. (Pharmaceutical Press, London, 1994), p. 310. (2) D. Kollmorgen and B. Kraut,J. Chromatogr., B707, 181-187 (1998). (3) M. A. Abuirjeie, M. E. Abdel-Hamid, A. A. Abdel-Aziz, and E. A. Ibrahim, Anal. Letters, 23, 67-82 (1990). (4) L. Gagliardi, A. Amaro, A. Basili, G. Carazzutti, E. Gattavecchia, and D. Tonelli,J. Chromatogr., 315, 465•i69 (1984). (5) N. D. Brown, L. L. Hall, and H. K. Sleeman, J. Chromatogr., 166, 316-319 (1978). (6) F. A. Fitzpatrick, A. F. Summa, and A.D. Cooper, J. Soc. Cosmet. Chem., 26, 377-387 (1975). (7) Y. Maeda, M. Yamamoto, K. Owada, S. Sato, and T. Masui,J. Chromatogr., 410, 413-418 (1987). (8) U. Leuenberger, R. Gauch, and E. Baumgarmer, J. Chromatogr., 173, 343-348 (1979). (9) N. Pongcharoenkiat, G. Narsimhan, R. T. Lyons, and S. L. Hem, J. Pharm. Sci. 91,559-570 (2002).

j. Cosmet. sci., 54, 53-62 (January/February 2003) Scalp hair length. I. Hair length in Florida theme parks: An approximation of hair length in the United States of America CLARENCE ROBBINS and MARJORIE GENE ROBBINS, Clarence Robbins Technical Consulting, 12425 Lake Ridge Circle, Clermont, FL 3 4711. Accepted for publication May 15, 2002. Synopsis Ten hair length studies were conducted (January through May of 2001) at theme parks in central Florida, by observing and counting people with different lengths of hair with reference to specific anatomical sites. We observed 13.20% of the 24,300 males and females with hair shoulder-length or longer and 2.43% with hair down to the bottom of the shoulder blades or longer. We observed only 77 persons or 0.32% with hair waist-length or longer. Only three of these 77 persons were men. We observed 22 men with hair shoulder- length or longer or about 0.18%, among men only. Only four women, representing 0.0165% of this population, had hair reaching to the bottom of the buttocks. Hairstyles such as buns, topknots, or other styles of wearing hair bound to the head did interfere with our estimate of shorter lengths but not the longer ones (waist-length or longer). Head covers, e.g., caps, hats, scarves, etc., did not interfere with these estimates. INTRODUCTION The first part of this paper summarizes important variables affecting scalp hair length and explains why hair length is so variable. In addition, highlights from a 1972 hair length survey among 2,400 women in the USA are presented along with literature data on abnormally long hair (global) from a few different sources. The second part then summarizes data from our recent studies on hair length, taken at ten different times in Florida theme parks during the period January through May of 2001. It is believed that these data represent a good approximation of the percentages of persons in the USA with hair of the indicated lengths however, only a limited amount of evidence is available to support this conclusion. The focus of these studies was on long hair, shoulder-length or longer, for the total population studied. Therefore, the only attempts to record the hair length of men versus women were on those longer lengths. Since the focus was on longer hair, no specific attempt was made to separate the data of men versus women on the shorter-length hair into subsegments. 53