We work on a diverse set of research projects to facilitate restoration of Pitcher's thistle. These projects are summarized below and focus on three broad areas: Genetics, Demography, and Floral Scent Chemistry.


Five natural populations and four restored populations of Cirsium pitcheri have been monitored demographically by our research group. The demographic data set currently includes a total of 12,754 plants monitored in nine populations for a maximum of 21 years and a minimum of 2 years (Table 1). Population growth rates are quite variable from year to year, but tend to be less than one for most of the natural populations. Factors influencing population growth rates include weather, particularly drought, as well as seed predation, and invasive species.

Research subdivisions and Site names Natural or Restored Annual demographic data collected and total number of plants in demographic study (mean population size) Plants alive in 2009 Flowering 2009 Seedlings 2009 Stochastic Growth Rate (95% CI)
Indiana Dunes NL Research Subdivisions and Indiana Dunes SP in Porter and Lake Counties, IN Ogden Dunes East Restored 1993-2009, 1241(88) 70 3 38 1.003 (0.882, 1.144)
Ogden Dunes West Restored 1993-2009, 2174 (107) 85 8 45 0.639 (0.549, 0.731)
Miller Beach Natural 1988-2009, 2236 (157) 82 6 23 0.887 (0.835, 0.943)
West Beach Natural 1988-2009, 2986 (107) 55 0 23 0.825 (0.778, 0.881)
Big Blowout Natural 1988-2009, 743 (39) 9 3 0 0.914 (0.832, 0.995)
Illinois Beach SP Lake Co., IL Restored 1991-2009, 1615 (179) 137 17 34 0.971 (0.969, 0.973)
Kenosha Dunes SNA Kenosha Co., WI Restored 2004-2009, 423 (133) 129 4 35 1.035 (1.030, 1.039)
Wilderness SP Emmet Co., MI (Sturgeon Bay) Natural 1995-2009 (except 2002, 2004), 2710 (450) 784 76 188 0.908 (0.907, 0.910)
Kohler-Andrae SP Sheboygan Co., WI (Sheboygan) Natural 2007-2009, 209 (147) 161 56 33 0.849 (0.848, 0.850)

Table 1. Study sites for Cirsium pitcheri around Lake Michigan used for demographic analysis. (SP = State Park, NL = National Lakeshore, SNA = State Natural Area).

Floral Scent

More details soon!


We have investigated whether differences in quantitative trait variation correspond with population size, molecular diversity, or observed differences in performance in restorations. Quantitative trait variation is being determined in a common garden study following the methods of Knapp & Rice (1998) using seeds from four natural and four restored populations.

To measure the consequences of inbreeding (genetic load) on Cirsium pitcheri, we grew out plants from two populations which are currently used in our restorations at Illinois Beach and Kenosha Dunes; Kohler-Andrae State Park (WI) and West Beach (IN). These populations also represent the two extremes of Fis (0.25 & 0.13) and gene diversity values (0.13 & 0.20).

As plants in the common garden took only two years to flower, compared to 5-8 yr in wild, we can assume that growing conditions are non-stressful. Nonetheless, plants from Kohler-Andrae had smaller juvenile and adult leaves, slower growth rate, smaller total size, fewer side branches and flowers. In combination with the evidence from neutral markers this might suggest that Kohler-Andrae is heavily inbred, although it is also possible that these differences are a result from adaptations to different local conditions.

We are investigating the genetic structure of Cirsium pitcheri at different spatial and temporal levels. We are using 8 microsatellite markers and screening fourteen natural populations throughout the species range to characterize the genetic diversity and structure (e.g., Fst) within and among these populations. Using this information we will compare statistics from the natural populations to those of the four restored populations.
The second component of the genetic study is to investigate temporal changes in genetic structure. We are comparing samples from nine populations collected in 1997 to samples from the same populations collected in 2007.

Species level diversity

We have collected samples from a total of 26 populations of Cirsium pitcheri; 23 from around Lake Michigan and two populations from Lake Superior and one from Lake Huron. We have screened 21 populations using eight molecular markers, and are currently screening the remaining five. For eight natural and one restored population in the southern end of Cirsium pitcheri’srange we have also screened collections from 1997, 2003 and 2008 to gives us three sampling points over a 11 year period (two generations).

Species level diversity (Noah Sokol NSF-REU)

Previous research by Loveless and Hamrick (1988) had suggested that the Cirsium pitcheri was a depauperate in genetic diversity. To test this further we compared the average genetic diversity (P, A, He, Fis) in a pair of Cirsium pictheri populations to Cirsium species with different life histories (rare, widespread, invasive) and modes of reproduction (asexual or monocarpic) (Table 8).Cirsium pitcheri had some of the lowest genetic diversity (although not as low as some asexual species) but lowest when compared to other monocarpic species. This suggest that as whole this species has lower neutral genetic diversity than other similar species, supporting Loveless and Hamrick (1988) idea that Cirsium pitcheri is a derivative species from the once more widespread species (Cirsium canescens).

Table 8. Percent polymorphic primers, Average number of alleles of polymorphic primers, Gene diversity (He- proportion of polymorphic loci) , Inbreeding co-efficient (Fis) and Genetic differentiation (Fst) of different Cirsium species with different life histories rare, widespread, invasive) and modes of reproduction (asexual (*) or monocarpic). Data for C. acaule, C. heterophyllum and C. arvense was taken from Jump et al (2002) and Jump and Woodward (2003).