52 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS older anatomical evidence would have implied. Among the connections leading into the amygdaloid complex, only olfactory fibers are well defined anatomatically nearly all parts receive either direct or indirect connections (30). The hypothalamus, the most important part of the limbic system as far as behavior is concerned, receives connec- tions from the amygdaloid complex. Hypothalamic activity regulates the secretion of the gonadotropins, follicle-stimulating hormone (FSH) and luteinizing hormone (LH), from the anterior pituitary. In man, the amygdala has been shown to activate olfactory hallucinations and also is implicated in differentiating and identifying odors (31). Electric stimulation of the amygdala and the pyriform cortex of human subjects has been reported to be followed by immediate and excessive incretion of adreno- corticotropic hormone (ACTH) together with a rise of plasma cortisol (32). Of even greater interest, significant increases were observed in plasma cortisol arising from electrical stimulation of the olfactory mucosa (33). No change in plasma cortisol occurred in individuals with post-operative anosmia. The phenomenon was interpreted as demonstrating that intact olfactory pathways allow the electric stimulus to reach hypothalamic structures, which in turn permit the release of ACTH-releasing factor and rise in cortisol level. This interpretation suggests that a specific odorant(s) can cause the same type of effect as observed with electrical stimulation. From the foregoing anatomical and clinical considerations, albeit sketchy, it is evident that an intimate relationship exists between the olfactory and limbic systems in hu- mans. Consequently, pheromonal effects could be involved in various aspects of re- productive physiology in man, similar to those known to occur in non-human animals. 2.3 THE PREOPTIC AREA AND OLFACTORY SENSITIVITY Pfaff (34) compared responses of the normal male rat to urine and non-urine odors by measuring signals from the olfactory bulb and preoptic area, which is an anterior exten- sion of the hypothalamus. A high proportion of preoptic region units responded dif- ferently to estrous female rat urine odor than to ovariectomized female urine, while only a low proportion of olfactory bulb units did so. A reversal was evident for non- urine odors. Approximately 75 per cent of all units, in both the olfactory bulb and the preoptic area, responded differently to urine odors than to non-urine odors. These data suggest that it is theoretically possible for receptor site information, coded by the olfactory bulb, to be "decoded" solely in the preoptic area. Therefore, a volatile chemical, not necessarily detected as an "odorant," might be physiologically active by virtue of the olfactory-preoptic-hypothalamic-pituitary relationship. Also, if the olfac- tory efficiency of the cells in the preoptic area is greater than those in the brain regions responsible for other aspects of odor perception, it is possible that certain odorants, i.e., pheromones, in sub-threshold concentrations for regions in the brain other than the preoptic area, can bring about a response in the preoptic area of mammals. 3. ENDOCRINE FUNCTION AND OLFACTORY SENSITIVITY IN WOMEN Biosynthesis of pheromones and scent gland secretion in mammals usually are influenced by the hormonal states of the body. Pheromones associated with reproduc- tion in lower animals display their effectiveness at or near estrus, the period of ovula- tion and maximum sexual receptivity in the female. Fluctuations in odor detection

PHEROMONES (OLFACTORY COMMUNICATION) 53 performance changes with amounts of circulating estrogen and progesterone in the rat (35). Peak olfactory acuity occurs around the ovulatory period and the fluctuations are not evident in ovariectomized rats. A similar behavior is manifested in women. LeMagnen (36) found that the ability of women to detect the odor of a musk (Pentade- canolide) reached a maximum near or on the day of ovulation. Many materials of both animal and vegetable origin display a musk odor (37) and it has been often stated, with little substantiation, that musk odorants are of biological significance to man. Although the considerable variation in olfactory acuity with regard to Pentadecanolide was reported to be specific for this chemical (36), subsequent investigators (38,42) demonstrated increased sensitivity for at least eight different odorants in the time near ovulation. This cyclic variation correlates with alterations in hormonal levels occurring during the ovarian cycle. Men and ovariectomized women show no variation of olfac- tory acuity with time. Women with decreased gonadal secretion who were not receiv- ing estrogen often displayed poor odor sensitivity. Estrogen treatment significantly enhanced, but testosterone lowered, their acuity to odorants. Since menstruation is triggered by a decrease in estrogen levels, it is not unexpected that odor thresholds in women reach a maximum during the period of menstrual flow. The prepiriform cortex and preoptic area have particularly high uptake of estradiol, an estrogenic hormone, in the female rat (43). It is tempting to speculate that the increased estrogen liberated during woman's ovulatory phase of the menstrual cycle is accompanied by the hor- mone's increase in these regions of the olfactory pathway, thereby causing a heightened olfactory sensitivity to odorants. Despite our ignorance of the precise man- ner by which gonadal hormones influence odor sensitivity, a consistent finding in most studies is the well defined correlation between circulating level and type of hormone with the performance of the sense of smell in the female. In higher animals, the mechanism to ensure mating involves more than one sensory modality and often may require a complex interaction of diverse influence mediated by all the sensory processes. For example, women possess lower thresholds, i.e., greater sensitivities, to visual (44) and auditory (45) stimuli at approximately the time of ovula- tion. From the viewpoint of evolution, such peak attuning to sensory signals would increase the probability of coitus and result in an increased probability of conception. In man's dim past, increased sensitivity to an olfactory signal (acting as an aphrodisiac pheromone) continually emanating from the male might have been perceived by women near the time of ovulation only. Odor quality also may play a role in the olfac- tory behavioral response. The perceived quality of certain odors changes on dilution. Consequently, even if a biologically important odor is detectable throughout the menstrual cycle, the perceived odor quality may change with perceived intensity fluc- tuations. Until recently, anecdotal and indirect observations have indicated that social grouping of the human female can cause synchronization of menstrual timing. Such an effect is not unlike that seen in rodents (7). McClintock (46) conducted a study with young women living in the same college dormitory and demonstrated that menstrual syn- chrony indeed occurred. In addition, the data indicated a parallel with the Whitten ef- fect (8) in mice in which the suppression of estrus in groups of female mice is released by the introduction of a male pheromone. The suppression of the ovulatory cycle in some of the women was released when they were exposed to men. Odor may be impli- cated in these phenomena, but further studies must be conducted in order to de- termine the basis of the intriguing effects.