RABBIT SKIN STUDIES IN EVALUATING COSMETIC SAFETY 379 kept from man, but the weak reactors may not be eliminated but must' be studied employing the human patch test techniques. The laboratory techniques for studying sensitization in the guinea pig follow the exact timing procedure which the dermatologist employs in the human. Instead of a patch technique, the material is injected intradermally: (as an 0.1% solution or suspension) every other day for ten applications. Following a two weeks' rest a challenging dose is injected, and the resulting response compared to the preceding reactions. Sensitization is present if the challenging dose causes a greater skin response than that evoked by any earlier injection. Primary irritation, which denotes a definite skin response to a single application of a cosmetic, can be readily studied in the rabbit. The patch test technique is employed, and a known amount of material is applied to the backs of animals the fur of which has been previously removed with the aid of an electric clipper. Several abrasions are made on one side of each animal. These abrasions penetrate to the dermis but do not go through it, eliminating bleeding. The patch used is a 2"x 2" gauze and a standard dose of 0.5 g or 0.5 ml of the same is applied under each patch. If a powder is being studied, a thick paste is made with water, applied to the skin and, then covered with the patch. Each animal receives one patch on its intact side and one patch over the abraded area. After the patches are applied they are kept in place by means of a rubber dam. This also reduces evaporation. The animals are immobilized in a holder for 24 hours, after' which time the patches are removed and the degree of skin reaction noted. Readings are also made 72 hours after the patches are removed. Scoring is based upon two responses--erythema and edema. Each response is further divided according to degree, and a numerical value given for each stage. Erythema may manifest itself by slight pink to beet red, while edema may be barely perceptible or may produce a raised area more than 1 mm in height at the site of treatment. In order to determine the extent of correlation when this procedure is followed in the rabbit, with the response which the human would show to the same preparations the following experiment was performed. Four prepara- tions were screened for their activity on both the rabbit and the human.. The first was a depilatory in a cream form, the second a revised lornrelation of a nail polish remover, the third a shampoo containing a new anti-dandruff ingredient, and the fourth a cuticle softener. When these four materials were each patch tested on the backs of groups of rabbits the animals res- ponded by showing a marked edema and erythema. After these results were obtained 24-hour patch tests were made with each of these preparations under the direction of a dermatologist. 25 humans received each of these materials as a single patch, and the area was examined 24 and 72 hours after exposure.

380 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS The protected depilatory was a nfild irritant in 9 of the 25 patients, and the nail polish remover produced a severe reaction in all of the patients. The shampoo, when applied to 25 humans, caused a mild inflammation at the site of application in 16 of the 25 patients, while the cuticle softener resulted in a severe dermatitis in all of the patients treated. It appears from this direct comparison between the responses obtained in the rabbit, and the responses obtained in the human, that the reactions in the rabbit can be used to predict the type of response which will occur in the human. The reaction of the rabbit's skin to a primary irritant appears to be a local one which manifests itself at the site of contact. Systemic changes may also occur following a single application of a cosmetic. The evaluation of a systemic reaction which can be judged by the level of a chemical in the blood stream or in an organ of the body can be quantitatively measured, and lends itself to a greater accuracy of determination than does the local dermal reaction. Most preparations intended for the skin do not show any systemic activity. Exceptions are creams, and lotions, containing estrogenic hormones. These are assayed by virtue of systemic absorption which results in vaginal wall changes, and subsequent estrus in the anestrus ovariectomized rat. In systemic toxicity the entire animal responds to the material rather than just the skin at the point of application, with absorption becoming a necessary adjunct to the final response. To study the acute response of a preparation, rabbits are prepared as previously described but the dose used on the exposed area is greater per kg of animal. The hair is removed from the entire back and sides of rabbits weighing from 2 to 3« kg, and abrasions are made as previously described. The test material is applied to groups of 3 rabbits at dose !evels ranging from 4 to 10 g per kg of body weight. Applications of the test material are introduced under a rubber sleeve which fits over the entire immobilized animal, and contact is for 24 hours. After that time the sleeve is removed, the animals' backs are washed clean, their skin reactions recorded, and they are returned to their individual cages to be observed for at least 14 days after the treatment. If the test material is an ointment or a cream, the use of a rubber dam is not required. Before the test is begun, and 1 week and 2 weeks after the application of the cosmetic, a complete blood count and urinalysis is performed. If the cosmetic contains an ingredient concerning which a chemical method for its ana!ysis in blood or urine is known, this analysis should be performed 24 and 48 hours after the application of the test material. At autopsy, sections of the liver, spleen, bone marrow and kidney of each an.•mal are prepared for histopathological examination. A/though death may not result from this single dermal application, physiological changes may occur, together