EVALUATION OF ANTIDANDRUFF FORMULATIONS 139 cocci). Obviously, no product should be considered of medical value until it has proved itself to be clinically effective in seborrheic individuals. In past years a useful yardstick for measuring the effectiveness of anti- dandruff agents has been their effect on yeast like organisms cultured from the scalp. The "bottle bacilli" are difficult to cultivate artificially unless the acid, oily, partially anaerobic conditions under which they grow naturally are simulated. Unfortunately, it has been our observation that even though the organism certainly occurs on the majority of normal male and many female scalps, it is not associated with seborrhea and dandruff in high enough frequency to establish causal relationship. The most apparent correlation we could find after studying 100 young men was that an increase in adherent scale was associated with an increased frequency of positive culture but that many of the subjects with no evidence of hair loss, dandruff or seborrhea showed positive cultures for the organism (3). A broader bacteriological base than just Pityrosporum ovale cultures was needed. Therefore, we studied the scalp bacteriology of a group of patients with seborrhea to determine flora patterns. All individuals had routine scalp cultures made on three media: brain-heart, Sabouraud's realrose and Benham's wort agar. The dominant organisms were Staphylo- coccus aureus, Staphylococcus albus and Pityrosporum ovale yeasts. All individuals harbored the Staphylococci but there was an apparent male dominance of yeast infection susceptibility (2). The next approach to the problem of evaluating antidandruff agents was to screen materials in the laboratory for their potency against the particular bacteria and yeasts isolated from seborrheic individuals. Typical results obtained on a laboratory screening are shown in Table I. Here measurements against the Staphylococci and several of the Pityrosporum groups are shown. The behavior of chlorinated phenols, quaternary ammonium compounds and anionic detergents indicates their failure to TABLE I--LABoKATOKY COMPARISON OF GEKMICIDAL AGENTS TESTED AGAINST OKGANISMS ASSOCIATED WITH SEBOKKHEA Agent Concentration Cup Plate Assay--Inhibition Zones, mm. Pityrosporum ovale Staph. Staph. 'Iype Type Type aureus albus 1 2 4 Hexachlorophene (G-11) Quaternary ammonium type Sodium alkyl sulfate (Duponol) Merthiolate Phenyl mercuric acetate Sodium sulfacetamide (SEN- zon) Actamer type compound Selenium sulfide (Selsun) Polysulfide compound 0.35% lotion 24 0.1% aqueous 30 1.0% aqueous 22 0.01% aqueous 23 0.01% aqueous 23 10% aqueous Slight 1.0% glycol 38 2.5% aqueous susp. 34 3% alcohol 56 30 Slight 0 0 32 23 Slight Slight 17 Slight Slight Slight 20 25 ...... 26 39 ...... 20 28 Slight Slight 42 52 33 26 27 38 37 30 41 40 43 35

140 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS control adequately the Pityrosporum ovale. To obtain the broadest spec- trum either a mercurial or some type of sulphur containing compound is desirable. This, of course, confirms clinical experience, because am- moniated mercury or sulphur preparations have been used widely for many years. During the past two years we have found it worthwhile to culture the specimens taken from seborrheic scalps on an anaerobic bacteriological medium in addition to the other media. We have not yet screened anti- dandruff agents against the anaerobic Staphylococcus that dominates this medium. In the past, we have relied heavily upon such laboratory comparisons. Laboratory study will separate the more effective antidandruff agent from the less, but it cannot guarantee clinical value. We have developed a panel type of comparative use test, wherein under controlled conditions subjects may compare clinically several products. This panel type of study has been sufficiently selective to differentiate between antidandruff agents that could not be separated by laboratory criteria. Normal male subjects are used because first, their short hair permits greater efficiency in scalp and hair sampling and secondly, the measurable factors found in dandruff conditions are more apparent in men. Details of this method have been published together with some general examples of the results (4), but further extensions of the work with some specific data on antidandruff formulations will now be given. Comparative data are collected and quantitated during clinical examinations of the subjects at definite periods after use of the products. During the test periods (weeks), the test products are randomized to minimize the effects of uncontrollable factors such as temperature, humidity and dust in the environment. A standard method of evaluation has been evolved through experience. The examina- tion measures the more general hair and scalp features, such as alopecia, visible loose dandruff, adherent dandruff and scale, grooming and cleanli- ness, as well as specific hair factors, namely, color, texture, brittleness, oil content and others. Scalp skin conditions, e.g., tightness, seborrhea, irritation or induced sensitization, are recorded. Laboratory work on bacterial flora, hair fragility and oil content are also completed. Quantita- tion is obtained by grading all observations numerically in terms of 1 to 4. Real numbers assigned to grade clinical observations make subsequent data comparison possible. Table II gives comparative data obtained for an individual subject. Shown are the examination values found after a control period during which no shampoo was used, after a similar two week trial period during which a germicidal shampoo was used, then again after a period during which a nongermicidal shampoo was used. It may be noted that use of the shampoo containing 1 % G-11 increased the apparent oiliness of the hair,