One of my best vacation memories is of ending a really long drive across France by checking late at night into a hotel in a small village. We slept fitfully becoming increasingly aware of an extraordinarily heady and intense smell of lavender. With the arrival of dawn we were able to discover that we were next to the village's lavender processing factory.

I love lavender. But perhaps it doesn't love me. In 2008, I read en passent somewhere that lavender oil in cosmetics may kill skin cells. Then I found a study - by strange coincidence conducted at my alma mater, University of Westminster in London - that makes the cell killer case pretty convincingly. Dr I C Locke found that in concentrations of 0.25%, lavender oil can be cytotoxic to human skin cells. He thinks that linanool, a component of lavender oil and an ingredient that crops all the time in cosmetics and shampoos, is the culprit along with linalyl acetate.

I did this research in 2008 and oddly, at the time, this seemed to be the only study of its kind. I couldn't find another that either corroborated or disproved it. There was, however, additional evidence that lavender can irritate the skin and cause allergic reactions. And at least two studies linked lavender (and tea tree oil) to the development of breasts in young boys and  premature breast development in young girls. Discontinuation of shampoos and lotions with these oils results in a rapid reversal (source).

18 months later in January 2010, I felt that I hadn't really got to the bottom of whether I should love lavender or relegate it to those sachets that freshen linen closets.

PubMed (the US Library of Medicine) has a long entry on lavender and confirms that it can be a skin irritant. However, it says nothing about the Westminster study. In fact, it cites "small Phase I human trials" of the lavender constituent perillyl alcohol (POH) as a treatment for cancer, although says efficacy has not been demonstrated.

As far as I can tell Dr Locke has not repeated his 2004 study on lavender oil. He has, however, performed tests on cloves and concluded that they are cytotoxic as well. Eventually, I found a 2007 study on a different plant's essential oil that was also composed of linalool and which also proved to be cytotoxic. A similar study from 2004 (testing another non-lavender linalool source) came to the same conclusion.

Actually, there are numerous studies that show linalool to be cytotoxic. This can be a good thing if you want to get rid of cancerous cells, but you wouldn't want it zapping the cells that you are spending a fortune on trying to preserve.

The important conclusion that I have come to is that lavender oil is probably best avoided in direct contact with the skin given that it is mostly comprised of linalool and linalyl acetate. Many cosmetics simply list linalool, without giving its source. However, it seems that linalool, wherever it comes from, is cytotoxic. Of course, we don't know what quantities are used, but the Westminster study suggests you don't need a lot.