Contact the CoNPS Team

Administration (virtual office)
1536 Wynkoop Street, Suite 911
Denver, Co 80202

Programs HQ & Store
704 East Boulder Street
Colorado Springs, CO 80903

by Ken Keefover-Ring

Ph.D. Candidate, University of Colorado, Boulder

Sometimes you just have to stop and smell the Wild Bergamot, and that is exactly what I have been doing for the last four years as part of my Ph.D. dissertation research, conducted with Prof. Yan Linhart at CU-Boulder.

Like many species in the mint family, Monarda fistulosa L. (Lamiaceae), commonly known as Wild Bergamot, Bee Balm, or Horse Mint, produces small volatile chemicals called monoterpenes in tiny sacs (trichomes) located on leaves, calyces, and even flower petals. When one rubs bee balm leaves, these trichomes are broken and almost immediately one detects the scent of escaping monoterpenes. Also known as essential oils, monoterpenes have been used by humans for thousands of years as fragrances and flavors; however, plants use them for a variety of functions, including: suppression of plant competitors, repelling herbivores, or attracting pollinators and seed dispersers (Harborne 1993).

Monoterpenes are widely distributed among a variety of plant families and their variation in natural plant populations has been extensively studied. Many labiate species contain individuals with distinct chemical phenotypes, called chemotypes, which are controlled by simple Mendelian genetics (Vernet et al. 1986, Vokou et al. 1993). Plants of a particular chemotype usually produce one monoterpene that dominates their total essential oil composition. This means that individuals of the same species can have very different smells, tastes, and interactions with other organisms, such as herbivores and pollinators.

Prior to my work, three chemotypes had been identifiedin Monarda fistulosa, containing either geraniol, carvacrol, or thymol, as their main monoterpene (Marshall and Scora 1972, Weaver et al. 1995, Johnson et al. 1998). The smell of geraniol is sweet or lemony — it is a major component of lemon oil, whereas carvacrol and thymol remind one of thyme or oregano, both of which have these chemotypes. While the existence of these three chemotypes in M. fistulosa has been known since the early 1970s, only one study presented detailed information on the distribution of chemotypes in the species (Marshall and Scora 1972).

I have discovered that two of the known M. fistulosa chemotypes, carvacrol and thymol, occur in Colorado in pure and mixed populations. Additionally, in one population in southern Colorado, I have found a previously unknown chemotype for this species with essential oil composed almost entirely of the monoterpene linalol. With the support of a Colorado Native Plant Society research grant from the John Marr Fund, I have been characterizing the chemotype variation of M. fistulosa over the landscape and trying to understand how these patterns arose and how they are maintained.

The first step to understanding chemical polymorphism in M. fistulosa was to map chemotype patterns of populations. The methodology I used was simple. Using historic herbarium records, information from locals and by just driving around, I located about 50 populations of wild bergamot throughout the state. At each site I randomly collected a single leaf from an average of 20 plants, soaked them in pure ethanol for one week to extract the monoterpenes and then analyzed the solution by gas chromatography, a technique that allows separation, identification, and quantification of monoterpenes.

Chemical analyses of over 900 Colorado plants revealed variation ranging from populations comprising only carvacrol plants to populations comprising only thymol plants, as well as populations comprising various mixtures of the two. So, what factors are responsible for these patterns? One explanation may be temperature. In Boulder County, where almost half of the populations analyzed occur (Figure 1), it appears that populations high in thymol plants are found at higher elevations or in colder areas, such as deep canyons. These micro site differences may be important in M. fistulosa, since in Colorado it has a large geographical range over most of the state with diverse habitats from prairies to high mountain meadows. Also, the idea that plants of a particular chemotype are excluded from an area due to temperature has been shown for common thyme (Thymus vulgaris) in the south of France. In the case of thyme, certain chemotypes were absent from the floor of an enclosed basin that regularly experienced much colder temperatures than the surrounding uplands (Amiot et al. 2005). To unravel this question in M. fistulosa, I plan to monitor temperature, humidity and other abiotic parameters at sites with different chemotype compositions.

Another factor that may shape the chemotype composition of populations is herbivory. At a few sites, I have found plants being fed upon by a small tortoise beetle (Physonota unipunctata) that specializes only on M. fistulosa. The larvae of this beetle have a curious defense mechanism; they accumulate their feces on two projections on the rear end of their bodies, which they curl upward, holding the “fecal shield” over their bodies. Since their diet consists exclusively of M. fistulosa, the fecal shield is rich in plant monoterpenes, adding a chemical dimension to their defense strategy. While the beetle larvae will readily feed on both chemotypes, my preliminary data show that when fed only carvacrol foliage they have lower survival and longer development times. Thus, in populations where these herbivores feed, thymol plants may experience more damage and reproduce less than carvacrol plants.

Another interesting find along my chemical odyssey was the discovery of what appears to be a new chemotype of M. fistulosa. On a lonely stretch of highway west of Trinidad, Colorado, I spotted an isolated roadside population. Initially, the plant leaves I collected all seemed to have the familiar smell of carvacrol or thymol chemotypes, found at many other sites. Suddenly, the scent of one plant was completely different from any other I had previously encountered. Upon returning to my laboratory, I immediately started my gas chromatograph and eagerly watched the monitor as the sample ran. The results showed one main peak, which was the monoterpene linalol. While linalol has been identified in a closely related species, M. didyma, no one has ever reported this compound in M. fistulosa.

As grandiose as it may sound, I felt I was witnessing an evolutionary event, where a new mutant phenotype had appeared, and if it had some heritable advantage over the resident chemotypes, the linalol chemotype may increase. Then again the plant may get mowed by the road maintenance crew before setting seed, a fate for which its new mutation would be useless.

Anyway, the next time you are hiking and encounter wild bergamot in the wild, don’t forget to stop and give it a sniff. You may not find a new chemotype, but you will sample just a little of the olfactory genetic diversity that makes up M. fistulosa in Colorado.

Photographs by Ken Keefover-Ring


Amiot, J., Y. Salmon, C. Collin, and J. D. Thompson. 2005. Differential resistanceto freezing and spatial distribution in a chemically polymorphic plant Thymus vulgaris. Ecology Letters 8:370-377.

Harborne, J. B. 1993. Introduction to ecological biochemistry, 4th edition. Academic Press, London; San Diego.

Johnson, H. A., L. L. Rogers, M. L. Alkire, T. G. McCloud, and J. L. NcLaughlin. 1998. Bioactive monoterpenes from Monarda fistulosa (Lamiaceae). Natural Product Letters 11:241-250.

Marshall, H. H., and R. W. Scora. 1972. New chemical race of Monarda fistulosa (Labiatae). Canadian Journal of Botany 50:1845-&.

Vernet, P., P. H. Gouyon, and G. Valdeyron. 1986. Genetic control of the oil content in Thymus vulgaris L.: a case of polymorphism in a biosynthetic chain. Genetica 69:227-231.

Vokou, D., S. Kokkini, and J. M. Bessiere. 1993. Geographic variation of Greek oregano (Origanum vulgare ssp. hirtum) essential oils. Biochemical Systematics and Ecology 21:287-295.

Weaver, D. K., T. W. Phillips, F. V. Dunkel, T. Weaver, R. T. Grubb, and E. L. Nance. 1995. Dried leaves from Rocky Mountain plants decrease infestation by stored-product beetles. Journal of Chemical Ecology 21:127-142.

Figure 1. The chemotype distribution of 24 populations of Monarda fistulosa in Boulder County, Colorado.

The Research Grants Committee thanks the many Colorado Native Plant Society members and Committee supporters for their donations to the research grants funds.  Your donation to these funds is encouraged and welcomed.  Please make your check to the Colorado Native Plant Society, designate the fund(s) to which you are donating and mail to:
Colorado Native Plant Society
P.O. Box 200
Fort Collins, Colorado 80522

Please contact Jan L. Turner for more information