?::'embryonic tissue. Another causal ?i connection between estrogenic hot- ' .: mone action and skin condition is :i!•}:" evident from the frequent occurrence ::11!i/øf atrophic skin changes during the j ! menopause. ARE HORMONE COSMETICS •': ':' DANOEROUS ? ß Occasionally one encounters in the literature warnings as to the poten- ::"tially carcinogenic action of estro- . genic hormones. Their origin ap- pears to stem from the experiments of Lacassagne (19) who showed that :!:: the administration of estrogenic hormones to young male mice from . a cancer susceptible strain, increased the rate of cancer incidence in later . life, as compared with the spontane- ous cancer incidence of the controls. In the case of mice free from suscep- ':" tibility to spontaneous mammary ::i tumors, the treatment did not elicit such a reaction. Incidentally, in order to produce this phenome- non, Lacassagne had to employ doses which are many times higher than would correspond to those ad- ministered in the most intensive es- trogen therapy, let alone in cos- metic usage. The following random quotations are given in further reference to this matter: Short (20): "Analysis of the data leads to the conclusion that no evi- dence exists that estrogenic hor- mones, given in physiological doses, have led to the development of car- cinoma in man." Geist and Salmon (21): "It is ob- viously impossible in human beings COSMETIC ASPECTS OF ESTROGENIC HORMONES 415 to administer the huge doses of es- trogens over the long periods of time that would justify comparison with the experimental production of car- cinoma in rodents. Howeve:r, the conclusion seems warranted, on the basis of these studies, that, within the limits of the dosage used in this investigation (up to 53,400,000 I.U.) there appears no evidence to justify the fear that carcinoma of the geni- tal tract may result from the thera- peutic use of estrogens." Hawkinson (22): "It is true that carcinoma can be produced in sus- ceptible animals with estrogen. This would seem significant were it not for the fact that the work has been done chiefly in the rodent, with rela- tively huge doses, and with animals having a high hereditary tendency to the development of carcinoma." Emge (23): "Evidence is accumu- lating to prove that the action of estrogenic hormones is controlled by definite biological patterns, and that their cancer-provoking faculty in small laboratory animals is strictly limited by hereditary tendencies .... We are not convinced, because es- trogen favors spontaneous mam- mary cancer in mice highly suscept- ible to this malignancy, that other species of mammalia are likewise af- fected." ß Goldzieher (3): "Warnings against the application of estrogens to the skin, lest they stimulate la- tent carcinogenic tendencies, are speculative, and not based on solidly established evidence. As a matter of fact, senile hyperkeratosis, in- cluding the pigmented variety, has

JOURNAL OF THE SOCIETY been classified with the precancerous lesions of the skin, yet it is favorably influenced by estrogens and may completely regress in the course of prolonged topical application." Dodds (24:): "(Lacassagne's re- sults) do not constitute a contrain- dication to the clinical use of estro- gens, since the doses given to human beings are fractional compared with those administed by Lacassagne to mice." Eller and Eller (5): "Particular attention was paid to the cytologic character of the epidermal cells of the patients showing a response to estrogen. It can be stated unequiv- ocally that no abnormalities were observed. There was no increase in the number of mitoses, or any suggestion of changes other than those of simple regeneration." Davis (18): "An estrogen be- comes a carcinogenic factor only when used on a strain of mice that have a strong hereditary tendency toward the formation of cancer." Mazer and Israel (25): "The authors have observed the develop- ment of uterine cancer in only 2 of 1000 climacteric women who had re- ceived as much as 10,000 rat units of estrogen every fourth day for pe- riods varying from six months to two years. According to the law of averages, more of these patients will eventually develop breast or uterine cancer, but its relationship to estro- gen treatment of two or more years previously would be more than doubtful." Dunbar (26): "There has been a considerable amount of work on the OF COSMETIC CHEMISTS carcinogenic properties of these ma- terials. In highly susceptible strains of rats under very special conditions, the administration of the estrogens has apparently produced carcinoma, but with ordinary laboratory ani- mals and in all studies so far on hu- man beings there is no evidence whatsoever that these products are carcinogenic. "We have in- vestigated the files of a number of firms producing these products searching for evidence of injuries. We have not found such evidence." The latter reference deserves par- ticular attention since it is based upon information gathered by the Food and Drug Administration. It should be noted, in this con- nection, that no quantity of estro- gen administered for therapeutic purposes can produce a concentra- tion in the blood stream comparable to that during the second half of ges- tation. As shown by Salter, Hunam, and Oesterling (27), also by Jailer (28), the excretion of natural estro- gen in the urine of the normal fe- male, at its peak, amounts to an es- trone equivalent of 600 to 1000 I.U. daily. (Between the peaks the vari- ation is from 50 to 200 I.U. daily.) Nevertheless uterine and mammary carcinomas do not occur in child- bearing women more often than in nulliparous women of the same age and hereditary background. In any case, because of the great difference in the comparative sizes of dosage, it is hardly admissible to con- sider the relevance of the carcino- genic aspects of estrogenic hormone