152 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS list of "safe" diluents for cosmetic use. The report of the European Committee on Chronic Toxicity Hazards (Eurotox) (22) lists lanolin as completely acceptable in the category of vehides and solvents. It should be pointed out that this review of the literature of reported instances of lanolin sensitivity yields a total of about 100 odd cases-- reported over a period of 30 years and from several countries. In spite of all these data lanolin continues to be suspect in some quarters and with some dermatologists. PRETESTING OF COSMETICS AND INGREDIENTS Under these circumstances it seems unnecessary that the cosmetic manufacturer should be greatly concerned with the problem of cutaneous reaction to lanolin because the reported instances of such reactions are so infinitesimally low. However, no reputable manufacturer of cosmet- ics and toiletries would want to place on the market a product that was unsafe for use. With the exception of 1960 legislation requiring the pretesting of the class of cosmetics defined as color additives, Federal legislation does not require the pretesting of cosmetics for safety. The law as it stands provides for the seizure of dangerous cosmetics, but only after the Government has proved they are dangerous. As pointed out by Miller (23), the Food & Drug Administration has urged that, just as in the case of new drugs and color additives, there be a requirement for pretesting all cosmetics. The Harris Bill now pending before Congress would require such pretesting. Levenstein, Draize et al. (24-26) describe several of the animal tests currently in use to screen cosmetics and ingredients for possible toxicity and irritation. Levenstein feels there is enough correlation between animal pretesting and human use experience to warrant such tests. Rieger and Battista (27) point out that it is not always possible to cor- relate between animals and human tests but urge that both be used. The following are basic screening tests considered to be a reasonable mini- mum for most topically used ingredients. Primary Derreal Irritation Primary irritation, or a skin reaction following a single contact with a substance, is usually determined by patch testing the material on both the intact and abraded skin of rabbits. A more realistic approach is the use of human subjects where possible. In the instance of lanolin and the oil soluble liquid fraction of lanolin,* Shelanski (28) reported that * Lantrol© Maimstrom Chemical Corp.

LANOLIN ALLERGY? 153 tests on 50 human subjects indicated they are not primary irritants. Kligman (29) reported another test for irritancy before the SOCIETY OF COSMETro CHEMISTS. Tested with this procedure both lanolin and the liquid lanolin showed "a remarkably low order of irritancy... ex- traordinarily innocuous for human skin" (30). Acute Oral Toxicity This is a basic test for any cosmetic or ingredient which may be in- gested, even if accidentally. Rats are fed successively increasing doses of the materials to determine the minimum amount, if any, required to kill 50% of the animals. In the instance of lanolin, its liquid fraction and other lanolin derivatives relatively massive doses are tolerated (31). Rabbit Eye Irritation The usual method of test for eye irritation is that described by Draize (26). Such tests on lanolin and some of its derivatives consistently show a very low order of irritation (32). Goldemberg (33) pointed out the possibility of formulating with anti-irritants, that is, substances which might reduce the irritancy of other materials. Russell and Hoch (34) reported that the addition of the liquid fraction of lanolin to deter- gent systems seems to reduce the eye irritation of the detergent system. Ser•sitization and Allergenicity One test for sensitization involves the intradermal injection of the test material in guinea pigs over a period of days, followed by a challeng- ing injection. Areas of the wheels resulting from the final injection are compared with average responses in the previous ones. A substantial increase in response to the final injection is considered evidence that the material is a sensitizer. Human testing is generally considered more desirable. Klauder (35) feels that such predictive patch testing using as many as 200 human subjects would only begin to detect allergenicity at about the mid-region of a scale of materials "where at one end of which allergenicity is very rare (lanolin) and at the other end it is fre- quent (paraphenylenediamine)." Regardless of this feeling, the repeated insult test is considered desirable. In the case of lanolin and the liquid fraction of lanolin such tests on 25 human male and 25 human female subjects (36) indicate these materials are not sensitizers. Silverman (37) studying a suppository base containing 40% of the liquid fraction of lanolin reported a similar result.