Cropwatch Newsletter 4 - Red Alert Issue


§2. Alleged Estrogenic Effects: Lavender Oil Goes Tits-Up.


If we can remember this correctly, a statistician once showed an apparently convincing statistical relationship between the number of divorces annually in the US and the size of the apple crop in China, as being a good example of a non-causal relationship. Cause and effect we are more used to getting heads ‘round - but the less direct tendency effects of causality are a bit more tricky to comprehend, and now we might an example of the latter to worry about.


The development of enlarged breasts (idiopathic prepubertal gynecomastia) in five boys aged four to seven who had been exposed to tea tree oil and lavender oil in OTC health products (soaps, shampoos), was reported by Derek Henley & Edward Reiter of the National Institute of Environmental Health Sciences, Research Triangle Park, N.C. & Bay State Children's Hospital, Springfield, Mass. USA at a presentation to the Endocrine Society’s 20th Annual Meeting at Boston on 25th June 2006. The authors also tested both oils and found an estrogenic effect on cultured female human breast cancer cells. The story is carried at various locations on the Internet: for example at,,8122-2259179,00.html & at The latter article reports that the author Derek Henley described exposure to lavender & tea tree oils and gynecomastia in young boys as ‘an association’, but stopped short of describing the relationship as ‘cause & effect’. The estrogenic effects in boys (presumably with low circulating testosterone), which also prevented male hormone responsiveness, are claimed to recede if the stimulus is withdrawn. Elsewhere in a second-hand report  at Henley is reported to say that “the finding is the first to implicate ‘essential oils’ from plants in gynecomastia.”


Cropwatch mailed the Endocrine Society (July 2006), who kindly sent us the abstract of the Henley & Reiter paper. We subsequently asked for permission to reproduce the abstract here, but this hasn’t come through as we go to press with this newsletter. Basically we can reveal that the human breast cancer cell line MCF-7 was used to determine the estrogenic activity of lavender & tea tree oils in vitro, and MDA-kb2 cells were used to evaluate the antiandrogenic activity in vitro, the results from which the authors claim are clinically relevant to the presented cases of males with prepubertal gynecomastia. Several Cropwatch supporters have written in enquiring how the authenticity of the lavender & tea tree essential oils used in the study was established, but we have no information on this at present. Lavender oils are, of course, commonly adulterated, but tea tree oils less so, since prospective adulterants (a-terpinene, g-terpinene) are not cost-effective.


Brief Background to Estrogenic Activity of Essential Oils.

Lets try and put this finding in context - is outcome so very unexpected?


We know that many plants posses sex hormonal activity such that NAPRALERT (1988) had records for 4410 plants as emmenagogues, 2630 as abortives, 1249 as possessing contraceptive properties, 1275 as aphrodisiac actions and 432 to fertility promotion (Farnesworth 1990). In addition we know that a number of classes of compounds possess estrogenic activity – apart from plant steroids like estrone found in apple pips Malus pumilia Mill. or 17b-estradiol in French Beans seeds Phaseolus vulgaris Wall., all isoflavones are estrogenic when injected parentally into animals (Cambie & Brevis 1997). Setchell (1998) notes that isoflavanoids show a weak affinity for the mammalian estrogen receptor at about 10-3 fold compared to estradiol. The weak estrogenic isoflavones formononetin & genistein in clover pastures (Trifolium subterranium L.) were reported to cause serious reductions in lambing percentages (Bradbury & White 1954, through Cambie & Brevis 1997), and the estrogenic effects of daizen & genistein in soy milk Glycine max Merrill. have been suggested as the reason why Japanese women have fewer menopausal symptoms (hot flushes etc.) compared with women in the US (Aldercreutz et al 1992, through Cambie & Brevis 1997). Other classes of compounds possessing estrogenic activity include certain coumestans, coumarins and diterpenoids. Plants used to treat menopausal symptoms containing steroidal saponins or related compounds include False unicorn (Chamaelirium luteum (L.) A. Gray), Sage (Salvia officinalis L.), St. Johns Wort (Hypericum perforatum L.), Black Cohosh (Cimicifuga racemosa (L.) Nutt.) & Chaste Tree (Vitex agnus-castus L.).


The essential oils that are traditionally/anecdotally associated (rightly or wrongly) with weak estrogenic effects include clary sage, sage, fennel & anise oils. Few published studies have been carried out in this area but Howes et al. 2002 showed that at very high concentration some oxygenated terpenes commonly found in essential oils could react with estrogen receptors, the significance remaining unclear. Early studies of Zondek (1938) implicated phenyl methyl ethers as likely estrogenic agents in fennel oil Foeniculum vulgare Mill. The later work of Albert-Puelo (1980) indicated that anethole polymers such as the dianethole & photoanethole were responsible for estrogenic effects in both fennel & anise oils, although Kraus et al. (1980) was unable to confirm the presence of this compounds in fennel oil (apparently these compounds only occur as an impurity in commercial anethole samples). Tabanca et al. (2004) working with various true anise oils (from Pimpinella spp.) and using a recombinant yeast screen found that estrogenic activity was not solely due to E-anethole. As a further reference to studies on the estrogenic effects of fennel, the disogenin content of the seed embryo was considered by both Fazili Fry & Hardman (1968) and Seshadri et al. (1973) to have phyto-hormonal properties and to emulate estrogenic activity in adult human females. In Sudan, the galactoguogic properties of the fennel plant are used in ethnic medicine (Ayoub & Svendsen 1981). Bone (2003) provides an interesting discussion in a monograph on Wild Yam Discorea villosa L. used by Western menopausal women for hormone balancing. Wild Yam contains steroidal saponins, & Bone argues that disogenin may also be formed as a result of bowel flora metabolism of the yam rhizome. Bone further suggests however that there is no proof that disogenin is metabolised in the body to any steroidal hormone, but rather that steroidal saponins may bind to receptors in the hypothalamus and exert a negative feedback effect for estrogen control.


The berry oil of the Chaste tree Vitex agnus-castus is sometimes also used for hormonal balancing by aromatherapists: especially in post- and peri-menopausal women. There is evidence that diterpenes in the oil cause circulating female hormone levels to change, sometimes dramatically. The oil should therefore only be used under medical monitoring & supervision (Lucks et al. 2002; Lucks B. 2003-4; Sorenson J. 2003).


The ethanolic extract & essential oil of Spanish Sage Salvia lavandulifolia Vahl was shown to possess estrogenic activity (Perry 2000) who had separately had looked at the in vitro estrogen receptor binding properties of the major components of the oil (Perry et al. 1996). Only geraniol (0.1-2mM <1% of essential oil) of the five terpenoids tested was found to have any estrogenic activity. Houghton (2004) reviewing the work concluded that the estrogenic properties of the extract was not shown up in the compounds tested and warned of the limitations of in vitro testing – for example an in vivo biotransformed metabolite  may be responsible for the estrogenic activity.


So…. any estrogenic activity of tea tree & lavender oils is not much supported in the literature by historical or anecdotal observations (apart from some very loose association with lavender), and no active substances found in lavender & tea tree oils causing these alleged associations readily spring to mind. It will be interesting therefore to see if the findings of Henley & Reiter can be independently confirmed with pure essential oils 100% derived from the named botanical source, and a direct biological mechanism derived for the process. Further, since other common natural phytochemicals such as linoleic acid have also been suggested as weakly estrogenic agents, it will be interesting to see how widespread the causes of this medical phenomena might potentially be - we are aware for example of the use of the seeds of Black Cumin Nigella sativa L. to plump out women’s breasts in Egyptian times (Genders 1986), but we have no information on whether the essential oil of Black Cumin has similar properties.




Albert-Puelo M. (1980) “Fennel & Anise as Estrogenic Agents” J. Ethnopharmacol. 2(4), 337-44.


Aldercreutz H., Hamalainen E., Gorbach S. & Golden B. (1992) Lancet 339, 1233.


Ayoub S.M.H. & Svendsen A.B. (1981) Fitoterapia 52, 245.


Bone K. (2003) A Clinical Guide to Blending Liquid Herbs Churchill Livingstone (Elsevier).


Bradbury R.B. & White D.E. (1954) Vitamins & Hormones 12, 207.


Crambie R.C. & Bevis A.A. (1997) Anti-Fertility Plants of the Pacific CSIRO Australia 1997.


Fazli Fry & Hardman R. (1968) Trop Sci 10,66.


Farnesworth N.R. (1990) Bioactive Compounds from Plants Ciba Foundation Symposium 154 J. Wiley & Sons Chichester 1990.


Genders R. (1986) Natural Beauty: The practical Guide to Wildflower cosmetics. Webb & Bower.


Henley, D.V., C.A. Bloch, and K.S. Korach (2006) “Components of health care products associated with male prepubertal gynecomastia possess estrogenic and antiandrogenic activities.” Endocrine Society Meeting. June 24-27. Boston.


Houghton P.J. (2004) Activity & Constituents of Sage Relevant to the Treratment of Symptoms of Alzheimer’s Disease” Herbalgram 61, 48-53.


Howes MJ., Houghton PJ., Barlow DJ., Pocock VJ., Milligan S.R. (2002). “Assessment of estrogenic activity in some common essential oil constituents.” J. Pharm. Pharmacol. 54(11), 1521-8.


Kraus A. et al (1980)  “Untersuchungen an Fenchelolen.” Dragoco Report 27: 31-40.


Lucks, Barbara (2003-4): various aromatherapy newsgroup mails.


Lucks B., Sorenson J., Veal L. (2002) “Vitex agnus-castus oil & Menopausal Balance: A Self-Care Survey ” Complementary Therapies in Nursing & Midwifery 8(3), 148-154. 


Perry N., Court G., Bidet N., Court J., Perry E. (1996) “European Herbs with cholinergic activities: potential in dementia therapy.” Inst. J. Geriat. Psychiatry 11, 1063-9.


Perry N.S.L., Houghton P.J., Theobald A.E., Jenner P., Perry E.K. (2000) “In-vitro inhibition of human erythrocyte acetycholine esterase by Salvia lavandulifolia essential oil & constituent terpenes.” J. Pharm. Pharmacol. 52, 895-902. 


Seshadri T.R. et al. (1973) Curr Sci 42, 421


Setchell K.D. (1998) “Phytoestrogens: the Biochemistry, Physiology & Implications for Human Health of Soy Isoflavones.” Am. J. Clin. Nutr. 68, 1333-1346.


Sorenson J. (2003): various aromatherapy newsgroup mails.


Tabanca N., Khan S.I., Bedir E., Annavarapu S., Willet K., Khan I.A., Kirimer N., Baser K.H. (2004) “Estrogenic activity of isolated compounds and essential oils of Pimpinella spp. from Turkey, evaluated using a recombinant yeast screen.” Planta Med. 70(8), 728-735.


Zondek B., Bergmann E (1938) “Phenl methyl ethers as oestrogenic agents” Biochem J. 32: 641-645


Back to Contents                                                                                                                                                                                                                                    Next Article