542 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS also, was it a finished product, was it a primary irritant, was there a soap reaction, were all skin tests negative, and what was the reaction time? Finally, lists of positive and/or negative reactions are prepared in terms of the number of people affected. If the clinician had a pertinent com- inent, it would also be included. These data are fed to the computer, which updates the record and generates our printout. The printout is in formulation number order and arranged by date. Two copies of the printout of clinical results are sent to the clinical office and one is retained in Technical Information Services (T.I.S.). When these data processing procedures were initiated, the Clinical Cosmetic Department requested only one copy of output per update, but because of the many telephone requests to their office for this data, an additional copy was supplied, giving both the research office and clinical area look-up capabilities. This printout has proven to be a very important desk-top tool inasmuch as it is not necessary to contact T.I.S. for the .data in question. Questions generally asked concern the status of specific formulations, whether they were tested, and with what results. At present, we update the clinical data base monthly and at the same time generate the necessary printouts for the Clinical Cosmetic Depart- ment and T.I.S. The following is an example of a typical search request processed in our office: Have the following ingredients been combined to formulate a lipstick and has it been tested in skin irritation studies? Oleyl alcohol, Lantrol, Lanogene, Carbowax 400, methylparaben, butylated hydroxy- anisole, lipstick oil, D & C Red 7, D g• C Yellow 7, D & C Orange 8, F D g: C Blue 1 and titanium dioxide? We first referred to our Specific Ingredient Dictionary, a portion of which is shown in Table VI. This look-up table contains an alphabetized list of ingredients with their corresponding Termatrex card color num- bers. We pulled the 12 Termatrex cards corresponding to the 12 desired ingredients and overlayed them on the shadow box. Light did not show through, thus indicating the ingredients in question had not been previ- ously formulated (i.e., there were no formulations common to all 12 in- gredients). Rather than settle for a negative answer, we made an attempt to find a lornrelation similar to the one requested. The four pigment cards, D gc C Red 7, D g: C Yellow 7, D & C Orange 8, and F D g: C Blue 1, were removed from the shadow box leaving only the base formulation. Then, by alternately replacing the pigments, it was found that D 8c C Red 7 ex- isted in combination with the desired base in formulation 6321. The

FORMULATION-CLINICAIJ DATA RETRIEVAI. SYSTEMS 5.1.3 Table VI Specific Ingredient Dictionary Ingredient Name Termatrex Card Identification Acetophosphate Red 85 Acetulan Yel 00 Acetylated lanolin dervt Grn 78 Acridine Orange RRR Grn 50 Acrysol A1 Yel 69 Adipic acid Grn 51 Albagen No 4439 Grn 79 Alipal-All Red 86 Allantoin Yel 70 Alon C Yel 01 Alrosol C-All Yel 02 Alrosperse 1 IP Grn 80 Alunfina hydrate C-3I Grn 52 Aluminum chlorhydrate Red 75 Aluminum chlorhydroxide Grn 81 Aluminum chlorhydroxy allantoin Grn 76 Aluminum chloride Red 87 Aluminmn powder Grn 82 [ortnulation number was then checked against the clinical data printout for data obtained in skin irritation studies. SUMMARY We have presently in operation two systems which can be used sep- arately or in combination with each other: the "Specific Ingredient" Termatrex system and "Clinical Sensitivity Test" desk-top look-up table. They have worked well to answer the more common questions related to research in these areas. (Received November 12, 1971) REFERENCES (1) Starker, Lee N., Owen, Katherine Crawford, and Batson, Betty Cooper, A multi-level retrieval system. II. Medium-sized collections, J. Chem. Doc., 9, 161 (19•9).