152 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Present methods for the detection of prospective allergens in the guinea pig, such as the Landsteiner-Draize test, are successful with strong sensitizers however, sensitizers of weak or intermediate strength are often missed (6, 7). For instance, Marzulli and coworkers, in a study of allergic contact dermatitis in the guinea pig and in man, reported that they were unable to sensitize any of several hundred guinea pigs to the topical anaesthetic '%enzocaine," where- as they could sensitize a substantial number of men in a human test popula- tion to this compound (5). Clinical]y, benzocaine is a fairly common cause of allergic contact dermatitis in man (8). We here propose a new method for rating, in the guinea pig, the allergic potential of prospective sensitizers of man. Means are taken to render the guinea pig more agreeable to sensitization, so that weaker allergens are recog- nized. The method derives from the "split-adiuvant" technique, in which chemical allergen and Freund's complete adiuvant are administered separate- ly to the skin, rather than as an emulsion (9, 10). MATERIALS AND METHODS Animals The subiects of these experiments are closed-colony, controlled randomly- bred, Hartley guinea pigs weighing 350-550 g. The animals are maintained on fresh Purina guinea pig chow supplemented by thrice weekly lettuce, and water ad libitum. Males or nonpregnant females are used. The colony is clini- cally free of infection with Streptococcus C we exclude animals of question- able health (skinny, poor coat, etc. ). The toe nails and distal portion of both rear feet are wrapped with water-proof adhesive tape so as to prevent skin damage from scratching or clambering by cage mates (Fig. 1). These "boots" are renewed as they are shed. Sensitization The test animal is clipped in an area of skin on the back lust behind the right shoulder girdle. The area to be clipped is delimited by a cloth frame whose square center measures about 2 x 2 cm. We use an Oster animal clipper with a size "0000" head the lower teeth of the head are adiusted so as to even- ly, and only slightly, over-ride the upper teeth-this minimizes skin irritation. Long hairs at the edges of the clipped patch, which might protrude into the site, are cut away with a scissor. The clipped patch is then shaved so as to re- move a good portion of the loose keratin in practice, a single-edge adiustab]e safety razor was found to be the best. We shave iust short of bleeding and, in place of shaving cream, saturate the site with water. The animal is now wrapped in a "window dressing" which consists of an outer layer of adhesive

BIOASSAY OF CONTACT ALLERGENS 153 .v ..,,. •:.-'•-..•%., :-,.. :: .•:.:..:..::j:' :..'•/. ::,• .. .• . .-el •':::'::*t'*':: •. . .., .... :½.'-'? -•.:, i•,, .....-:'• ,..•.: ß ...,. • .'•.' ::.:...:..: .-:- .• . •.:... "•'.• .. 7' ' • ' • ..-,:. .. •.. ':.'.!.' •-',' :7' ' ".•.-:•/• . •:•-•.i.' •-'•'•'i• • •:".' •:¾.i u' ',.. •.:',': • -. •.:•.. ß .. • •c' "• • "' . ........... ,,. . ß .• : ' '•.'1•' ¾.•::•½• /. . •...-- .- ?•' ... =- . : .... . •.. stretch fabric (Elastoplast½*) and an inner layer of loose mesh gauze (Kling©t). A 2 x 2-cm opening has previously been cut in the dressing this opening is placed over the sensitization site, but overlapping slightly forward. The dressing is fixed with encircling adhesive tape fore and aft and is allowed to settle for a few hours during this time it usually slides a % cm or so caudad of its initial placement. To the site, dry ice is applied for 5 sec with firm pres- sure. Then 0.2 ml of test ointment is applied, gently spread, and this is cov- ered with a double thickness of Whatman #2 filter paper (Fig. 2). (In the case of liquids, 0.1 ml is usually used. ) The filter paper has been cut so that its edges fit under the dressing. Then the area is covered with occlusive tape (Blenderm©•) and this, in turn, fixed in place with adhesive tape. On Day 2, the dressing that covers the window is removed, 0.2 ml of test material is ap- plied, and the window is reclosed with the same dressing. Day 4, the window is again opened and 0.1 ml of Freund's complete ad]uvant (H37Ra, Difcoõ) is * Duke Laboratories, Inc., South Norwalk, Conn. ? Johnson and Johnson, New Brunswick, N.J. •+ 3M Company, St. Paul, Minn. õ Difco Co., Detroit, Mich.