SPECIFIC LANOLIN ALLERGY 329 positive reaction on patch testing with anything. Out of all 1930 positive reactors, therefore, the total percentage of reactions amounts to 196.25•o, thus each reactor was sensitive, on the average, to 1.96 substances. Baer et al. (31) found the average number of sensitivities per patient, out of 743 in total, to be 4.37 for weak reaction and 2.72 for strong reaction. Cases of this type are not necessarily indicative of true cross-sensitization, of course, but are at least to be regarded as polyvalent sensitivity. Hjorth and Trolle- Lassen (10) found that sensitization to lanolin rarely went alone and that similar symptoms could be provoked by, for example, parabens, or neomy- cin which are far removed in chemical structure from lanolin. They also reported that some forms of dermatitis, particularly varicose eczema of the leg, were more likely than others to be associated with hypersensitivity to lanolin, and that most patients had a long history of eczema before lanolin sensitivity was diagnosed. Some of these findings were corroborated by Wereide (9) whilst Cronin (19) reported that lanolin hypersensitivity could be an aggravating factor rather than a primary condition. In other words, the sensitivity was grafted on to an already existing pattern of dermatitis which did not entirely disappear when contact with lanolin was discon- tinued. Epstein (22) also referred to multiple sensitivities, whilst Stolze (35) dealt in detail with this matter and made a regression analysis showing how the average number of sensitivities per case increased with duration of case history, from 1.6 sensitivities at zero duration to 6.0 at 37 years. On the same theme, a large-scale co-operative European study by Fregert et al. (28) found that, out of a total 4825 patients at skin clinics in seven different countries, whereas 115 showed a positive reaction to lanolin alcohols, 35 of these were simultaneously sensitive to neomycin, 24 to wood tar, 22 to colophony, and so on (private communication to the author). Hence, a reported case of hypersensitivity to lanolin or lanolin alcohols does not automatically mean that either substance was the primary or initiating allergen. Approaching the problem from the opposite direction, tests for allergy to lanolin carried out on normal, apparently healthy skins have not elicited a single report of positive reaction. Sulzberger and Lazar (11) tested three normal subjects, Sulzberger, Warshaw and Hermann (12) tested 120, Newcombe (1) reported on 50 and Norholm-Pederson and Sylvest (32) tested 111, results on all 284 being negative. Even when using an enhanced 'maximization test', in which sodium lauryl sulphate is used to make the skin more sensitive to test substances, Kligman (33) failed to elicit any hypersensitivity to lanolin in 25 human subjects, whilst Magnusson and

330 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Kligman (34), using the same method with a parallel Landsteiner-Draize test on guinea pigs, obtained zero scores throughout. As confirmation of these results, the present author has not personally known of a single case of hypersensitivity to lanolin or wool wax amongst employees engaged in lanolin refining, crude wool wax recovery, or woolsorting, where people are in constant and liberal contact with the substances, out of a total of at least 3000 over a period of 28 years. Primary specific lanolin allergy and secondary hypersensitivity Attempting to identify a specific allergen in lanolin, various workers have tested patients known to be hypersensitive to lanolin with certain individual, known constituents of the substance. The absolute purity of the samples used was not, however, reported and the test results were mixed. Thus, Ellis (36) investigated the effect of cholesterol from two sources and ob- tained positive results to both, on several patients. Sulzberger and Lazar (11), also Sulzberger, Warshaw and Herrmann (12) found mostly negative reactions to two lanosterol samples of different purity and also to cholesterol from two sources (lanolin and cattle spinal cord), only one positive reaction (to lanosterol) being reported. Everall and Truter (37) worked with crude cholesterol, lanosterol, cholestanol and cholesta-3,5-dien-7-one, only the crude cholesterol yielding positive reactions. After purification of this cholesterol reactions were negative, and it does seem unlikely that choles- terol could be the allergen since it is a normal and essential constituent of the human body. Everall and Truter isolated a small quantity of the impurity from the crude cholesterol and found it to give a positive skin response, but no specific chemical identification was made. In the present state of knowledge, no single chemical entity has been certainly identified as the principal allergen of lanolin, but it is now well established in the cited literature that certain sufferers from dermatitis, particularly where some types of eczema such as eczema of long duration, varicose eczema and ulcus cruris are concerned, become polyvalently hypersensitive to a number of chemically unrelated substances, one of which may be lanolin. Although in some of these cases a primary, specific lanolin allergy may be the original cause, in many others lanolin will not be the actual initiating allergen, if a single specific allergen does indeed exist. Since some patients may develop a hypersensitivity to lanolin as a secondary result of some entirely different primary cause, it is important in estimating the incidence of specific lanolin allergy to discount those cases, even though diagnosis of the lanolin hypersensitivity, along with