Page  89

HYBRIDIZATION DYNAMICS OF INVASIVE CATTAIL (TYPHACEAE) STANDS AT PIERCE CEDAR CREEK INSTITUTE: A MOLECULAR ANALYSIS

Kelsey Huisman, Alex Graeff, and Pamela J. Laureto Department of Biological Sciences Grand Rapids Community College 143 Bostwick NE Grand Rapids, Michigan 49503 plaureto@grcc.edu

ABSTRACT

Three cattail taxa are recognized in Michigan USA: native Typha latifolia (broad-leaf cattail), the invasive Typha angustifolia (narrow-leaf cattail), and the hybrid of these two species Typha ? glauca. Typha angustifolia and T. ? glauca are of special interest because of their ability to aggressively spread and out-compete the native cattail T. latifolia. Typha ? glauca has been shown to out-com- pete both its parental taxa and produce monospecific stands. We surveyed the Pierce Cedar Creek In- stitute (PCCI) property for cattails and located 25 distinct cattail marshes. We determined the total area of cattail marsh at PCCI to be roughly 10% of the 267 ha property. Cattail individuals were sam- pled from each of the 25 stands and RAPD markers were used to identify the individuals to species. We found that 20 of the 25 stands were monospecific for the native cattail, T. latifolia. Five of the stands were mixtures of the native T. latifolia and the introduced T. angustifolia, and T. ? glauca was found in two of the mixed stands. We recommend removal of the invasive T. angustifolia and T. ? glauca individuals and the establishment of a monitoring plan in order to maintain the long-term health of the cattail marshes at PCCI. KEYWORDS: Typha spp., RAPD markers, invasive species

INTRODUCTION

Species of Typha L. (Typhaceae), commonly known as cattails, are highly productive emergent plants that grow in a variety of wetland habitats throughout the world (McManus et al. 2002). In the northern USA and Canada three taxa of cattail have been recognized: Typha latifolia L. (broad-leaf cattail), Typha an- gustifolia L. (narrow-leaf cattail), and Typha ? glauca Godr. (white cattail). Typha latifolia is a native plant species that is a keystone emergent in marsh communities throughout North America (Smith 2000). In the United States, broad-leaf cattail is native to all 50 states. In Canada, it occurs in all provinces and territories except Nunavut (USDA, NRCS 2012). Typha angustifolia is thought to have been introduced to the eastern seaboard from Europe in the early 19th century (Stucky and Salamon 1987; Selbo and Snow 2004), although Shih and Finkelstein (2008) suggest it may be present in pollen cores dating back 1,000 years. Following its colonization of the Atlantic coast, T. angustifolia began to move inland slowly—but by the early 20th century it had begun a rapid westward expansion (Mills et al. 1993). Today, T. angusti-

Page  90

folia is found in 42 of the 50 United States; it is absent from Florida, Georgia, Alabama, Mississippi, Texas, Utah, Arizona, Hawaii and Alaska. In Canada it is found in all provinces except Labrador and Newfoundland and is absent in the Yukon, Northwest, and Nunavut Territories (USDA, NRCS 2012). Because of its aggressive spread, T. angustifolia is considered an invasive species. It often out- competes native wetland species, including T. latifolia, to produce very dense monospecific stands (Grace and Harrison 1986). Typha latifolia and T. angustifolia are obligate wetland species, meaning that they are always found in or near water. Both species generally grow in flooded areas; however, T. latifolia is typically found in waters that do not exceed 0.8 m while T. angustifolia prefers deeper water, usually greater than 0.75 m. Where the two species are sympatric, they typically segregate by water depth (Travis et al. 2010). According to Grace and Harrison (1986), both T. latifolia and T. angustifolia are self-compatible, wind pollinated, clonal species. The reproductive shoots of both species are monoecious, with staminate flowers occurring above pistillate flowers, and protogynous (pistillate flowers produced prior to staminate flow- ers). The pistillate flowers remain receptive to pollen for four weeks (Kuehn et al. 1999). While the protogynous condition would seem to facilitate out crossing (Smith 2000), Krattinger (1975) showed that cattails are largely self-fertilized and that vegetative reproduction occurs more frequently than sexual reproduc- tion. In part, the two species can be distinguished by the presence or absence of a spike gap between the staminate and pistillate flowers (T. latifolia generally has no spike gap, whereas T. angustifolia has a spike gap ranging from 0.5 to 4 cm). T. latifolia often flowers later than T. angustifolia but overlap in flower- ing times can lead to hybridization between the species (Selbo and Snow 2004). In fact, T. latifolia and T. angustifolia appear to hybridize wherever the two occur sympatrically (Galatowitsch 1999; Kuehn et al. 1999; Olsen et al. 2009). The third taxon, Typha ? glauca, is the hybrid of T. angustifolia (maternal) and T. latifolia (paternal) (Grace and Harrison 1986; Kuehn et al. 1999); how- ever, its hybrid status has long been disputed. Typha ? glauca has been identified as a distinct species; a stabilized hybrid taxon; an introgressed taxon with T. ? glauca representing a series of intermediates in a hybrid swarm; and also as a sterile F1 hybrid (Kuehn et al. 1999 and references therein). Taxonomic de- scriptions of T. ? glauca frequently identify the plant as a sterile F1 hybrid; how- ever, recent studies have revealed the presence of backcrossed and later genera- tion hybrids indicating that T. ? glauca has at least some degree of fertility (Snow et al. 2010; Kirk et al. 2011). The spread of invasive taxa are of particular interest to evolutionary biolo- gists and ecologists because of their ability to alter community structure and ecosystem function (Horvitz et al. 1998). Typha ? glauca is considered to be a highly invasive species due to its aggressive range expansion and ability to dom- inate wetland habitats. According to Galatowitsch et al. (1999), several hypothe- ses have been advanced in an effort to explain how introduced plant species be- come invasive species. The Introgression/Hybrid Speciation hypothesis suggests that interspecific hybridization between an introduced taxon and a native taxon results in novel phenotypes with selective advantages. Therefore, hybridization

Page  91

between native and introduced species is considered to be one of the driving forces behind the evolution of invasiveness (Ellstrand and Schierenbeck 2000; Schierenbeck and Ellstrand 2009). Hybrids between native and introduced species have frequently been shown to have increased fitness with respect to their parental species as they possess greater genetic and phenotypic diversity than their parents (Kuehn et al. 1999; Ellstrand and Schierenbeck 2006; Kirk et al. 2011). This appears to be the case for T. ? glauca which can colonize the en- tire range of water depths in which the parental species segregate (Travis et al. 2010). Several researchers (e.g., Zedler and Kercher 2004; Travis et al. 2010) have documented the competitive superiority of T. ? glauca indicating its po- tential as a highly invasive taxon. F1 hybrids are generally expected to be morphologically intermediate to their parental taxa but this is not necessarily the case for the hybrid T. ? glauca. The leaf width of the hybrid T. ? glauca is believed to range from 6 mm to 16 mm; although leaf widths up to 21 mm have been reported. Kuehn et al. (1999) found considerable phenotypic variation in leaf width for each of the parental species. The leaf width of T. angustifolia ranged from 4.5 mm to 12 mm and the leaf width of T. latifolia ranged from 7.5 mm to 22 mm. Therefore, the leaf width of T. ? glauca overlaps with the parental species making this trait unreliable for hy- brid identification. Flowering in cattails is also an unreliable trait for taxonomic identification because cattails often exhibit poor flower production. Dickerman (1982) found that over a three-year period only three of 1,779 marked shoots flowered at Lawrence Lake in Barry County, Michigan. The degree of shading, including self-shading in dense stands (Grace and Wetzel 1982), and the depth of rhizome submergence (Grace 1989) affect flowering success. In addition to poor flowering and the overlap in morphological traits, backcrossed and advanced generation hybrids are phenotypically more similar to one of the parental taxa further complicating the identification of hybrid individuals through morpholog- ical traits (Kuehn et al. 1999; Selbo and Snow 2004; Snow et al. 2010). Because morphological traits can be highly variable, DNA markers are considered to be more reliable for the identification of cattail species and their hybrids (Kuehn et al. 1999; Selbo and Snow 2004). This study examined 25 discrete marsh populations at Pierce Cedar Creek In- stitute in Hastings, Michigan, USA in order to gain an understanding of the dis- tribution and abundance of T. latifolia, T. angustifolia, and T. ? glauca. We used random amplified polymorphic DNA (RAPD) markers coupled with intensive field sampling of the 25 populations to identify individual cattails to species. Kuehn et al. (1999) developed species-specific RAPD markers for T. latifolia and T. angustifolia. RAPD analysis of genomic DNA is expected to produce species-specific RAPD banding patterns while the hybrid, T. ? glauca, is ex- pected to display the banding patterns of each parent. Because of the potential for T. angustifolia and T. ? glauca to invade the marsh communities at PCCI and out-compete the native T. latifolia, baseline data on the presence and distribution of Typha spp. is imperative to the development of an effective wetland manage- ment plan for the institute.

Page  92

FIGURE 1. Cattail stands identified at Pierce Cedar Creek Institute in Hastings, Michigan. Numbers correspond with the coding assigned to sampled individuals. (See Table 1 for the species composi- tion of each stand.)

METHODS

Study Site This research took place at Pierce Cedar Creek Institute (PCCI) in Hastings, Michigan, USA. PCCI is an environmental field station and nature center whose property consists of forest, field, open water, and a variety of wetland habitats. Of the 267 ha site, a total of 103 ha is considered wet- land. Brewster Lake and approximately 1.9 km of Cedar Creek are contained on the property and many of the wetlands are associated with these bodies of water. Most of the property was previously farmed and many of the wetland habitats along Cedar Creek appear to be recovering from distur- bance. In addition, the creek has also been dammed by beavers. This has led to flooding and the cre- ation of shallow water habitat which is ideal for cattail growth. The cattail composition at PCCI has not been previously studied, but a 2003 vascular plant inventory of the property indicated that only T. latifolia was present (Slaughter and Skean 2003).

Page  93

Mapping of Cattail Stands In April of 2012, we visually surveyed the property by traveling on foot along the trails, and also by canoeing Cedar Creek. After visual determination of the locations of cattail stands, each stand was mapped using a hand-held GPS unit (Garmin e-Trex 30). Mapping consisted of traveling the perime- ter of each cattail stand with the GPS unit creating an electronic track file. This data was entered into a geographical information system (ArcGIS 10) to create a digital map depicting the location of each cattail stand on PCCI property (Fig. 1). Each track on the GPS, with the exception of the Brewster Lake track, was created by walking the perimeter of the cattail stand. The Brewster Lake track was created by rowing around the perimeter of the lake in a boat.

Taxon Sampling Typha samples were collected from each identified cattail stand. Within each stand, we began collecting at the edge and traveled straight-line transects, by foot, through the stand, stopping at least every 10 m to collect leaf tissue from the nearest ramet (an individual shoot of a clonal organism). In larger stands, the distance between sampled ramets often exceeded 10 m. For stands that were too small to establish transects, or in cases where traveling in a straight line was impractical, sampled ramets were separated by a distance of no less than 10 m. Having a minimum distance of 10 m be- tween samples has been shown to decrease the likelihood of sampling clone mates (Kuehn et al. 1999; Olsen et al. 2009; Snow et al. 2010). No a priori species assignments were made at the time of collection since T. latifolia, T. angustifolia and T. ? glauca display significant overlap in leaf width (Kuehn et al. 1999, Olsen et al. 2009) and all samples were collected prior to flowering; however we did sample from a representative range of morphological variants within each stand. We sampled by clipping approximately 7 cm of leaf tissue from the youngest leaf. The tissue was immediately placed into a plastic bag containing silica gel desiccant. Each sampled ramet was flagged and num- bered along with the corresponding tissue sample.

Molecular Identification of Typha Species and Hybrids Total genomic DNA was extracted from approximately 3 cm2 of dried leaf tissue following a variation of the 2 ??CTAB (cetyl trimethylammonium bromide) extraction method for high polysac- charide plants described by Nickrent (2006). DNA extractions were stored at –20°C for use in ge- netic identification of species and hybrids. We used RAPD molecular markers to identify T. latifolia, T. angustifolia, and T. ? glauca. Total genomic DNA was amplified with PCR using RAPD primers A2 (TGCCGAGCTG) and A8 (GT- GACGTAGG) (Eurofins mwg/Operon Huntsville, AL, USA). Each 25 µL reaction mixture contained 19.7 µL ultra-sterile H2O, 2.5 µL 1 ? PCR buffer (200 mM Tris-HCl [pH 8.4], 500 mM KCl), 1 µL 50 mM MgCl2, 1 µL of either RAPD - A2 or RAPD - A8 primer, 0.25 µL 10 mM PCR Nucleotide Mix (USB Corp. Cleveland, OH, USA), 0.05 µL Platinum® Taq DNA Polymerase (Invitrogen, Carlsbad, CA, USA), and 0.5 µL total genomic DNA. The 25 µL reaction mixtures were incubated in an iCycler Thermal Cycler (Bio-Rad Laborato- ries, Inc., Hercules, CA, USA) programmed for 2 min at 94°C (initial denaturation); 25 cycles of 1 min at 94°C, 1 min 30 sec at 34.5°C, and 2 min 20 sec at 72°C followed by 20 cycles of 1 min at 94°C, 1 min 30 sec at 34.5°C, and 2 min 20 sec at 72°C with a 10 sec increase after each cycle. RAPD products were separated by gel electrophoresis on 1.2% agarose gels for 1 h at 90 V. Gels were stained in an ethidium bromide bath for 15 min and photographed under a UV light source. Molecular weights of amplification products were estimated using a 1 kb plus DNA ladder (Invitro- gen, Carlsbad, CA, USA). Primer A2 yields 5 species-specific bands with bands for T. latifolia at 1.5 kb, 1.0 kb, and 0.6 kb and bands for T. angustifolia at 1.2 kb and 0.8 kb; primer A8 produces 3 species-specific bands with T. latifolia having a band at 1.0 kb and T. angustifolia having bands at 2.0 kb and 1.8 kb (Kuehn 1999). F1 T. ? glauca individuals are expected to show all parental bands.

RESULTS Twenty-five wetland communities containing Typha spp. were identified from PCCI property (267 ha). These ranged in individual area from 8 m2 to 97,613 m2 (Table 1; Fig. 1). The largest stands were associated with the perime-

Page  94

TABLE 1. Area, number of Typha individuals (n) identified, and species composition of the 25 cat- tail stands identified at Pierce Cedar Creek Institute. Area was calculated in Garmin e-Trex 30. The species composition of each stand (L - Typha latifolia, A - Typha angustifolia, and G - Typha ? glauca) is presented as a percentage of each species per the total number of identified individuals from the stand.

Stand Area (m2) n % Species Composition 100 12181.00 9 L (100%) 200 717.82 4 L (100%) 300 27583.00 12 L (100%) 400 20048.00 15 L (100%) 500 3507.20 4 L (100%) 600 10751.00 8 L (75%), A (25%) 700 752.56 2 L (50%), A (50%) 800 22149.00 16 L (100%) 900 4607.00 16 L (100%) 1000 882.16 4 L (100%) 1200 97613.00 43 L (100%) 1300 14698.00 2 L (100%) 1400 14028.00 13 L (100%) 1500 5260.00 5 L (100%) 1600 3215.8. 8 L (100%) 1700S 207.03 3 L (33%), A (33%), G (33%) 1700N 177.02 3 L (100%) 1800 586.29 3 L (100%) 1900 95.06 1 L (100%) 2000 7133.76 8 L (100%) 2100 7695.90 14 L (77%), A (23%) 2200 2102.80 5 L (40%), A (20%), G (40%) 2300 1781.10 5 L (100%) 2400 371.96 3 L (66%), A (33%) 2500 1725.80 5 L (100%) 2600 7.63 1 L (100%)

ter of Brewster Lake and the wetland complexes adjoining Cedar Creek. The smallest stand was a drainage ditch adjacent to the main road into the PCCI property. This road was constructed between 1999 and 2001 (Brown, Pers. Comm.). The cumulative area of all cattail stands totaled 259,878 m2 (26.0 ha) which is approximately 9.72% of the PCCI property. We collected tissue samples from a total of 370 individuals from across the 25 cattail stands. Of these, DNA was extracted from 252 randomly chosen individ- uals, representing all 25 stands. Photos of two representative electrophoresis gels of RAPD DNA fragments produced by PCR amplification of cattail individuals using primer A8 are shown in Figure 2. The species-specific bands and species identification are indicated on the photos. We identified most individuals to species using primer A8. For approximately 1% of the individuals we confirmed the primer A8 species identification using Primer A2. No difference in species identification was observed between the two primers. RAPD analysis allowed us to identify 212 of these individuals to species: 200 (94.34%) were T. latifolia, 9 (4.25%) T. angustifolia, and 3 (1.42%) were the hybrid T. ? glauca. Of the 25 cattail stands at PCCI, 20 were monospecific, consisting of only the

Page  95

B.

Figure 2. Two exemplary gel photos showing ran- dom amplified polymorphic DNA (RAPD) pheno- types for cattail individuals identified using primer A8. The left lane contains a 1-kb plus DNA ladder size standard. A species-specific band for Typha lat- ifolia (?) occurs at 1kb and two species-specific bands for T. angustifolia (•) occur at 2.0kb and 1.8kb. A. One T. ? glauca (G) and two T. latifolia (L) individuals identified by RAPD analysis. B. Eight T. latifolia (L) and four T. angustifolia (A) in- dividuals identified by RAPD analysis. A.

native cattail, T. latifolia, four were mixed for T. latifolia and T. angustifolia, and two stands consisted of a mixture of both parental species and their hybrid T. ? glauca (Table 1).

DISCUSSION

This study documents the extent to which the native cattail Typha latifolia, the introduced and widespread cattail T. angustifolia, and the hybrid of these two species, T. ? glauca occur at Pierce Cedar Creek Institute. Approximately 40% of the PCCI property is characterized as wetland habitat based on its soil type and hydrology (Brown, pers. comm.). However, many of these wetlands are vis- ibly dry throughout much of the year. Because cattails prefer standing water, we were surprised to find that nearly 10% of the property consisted of cattail marshes. The density and species composition within each sampled wetland var- ied from what appeared to be a monospecific stand of cattails with one ramet im- mediately adjacent to another, to stands in which individual cattail ramets were separated by distances of approximately 10 m. Diversity appeared to be much higher in these stands with a variety of sedges, forbs, and shrubs occurring in be- tween the Typha ramets. On our initial observation most stands appeared to be the native T. latifolia because the individual ramets consisted of broad leaves; however, variation in

Page  96

leaf width led us to suspect that genotyping would reveal mixed stands of both parental species and their hybrid. Within the large stand that surrounded Brew- ster Lake (stand 1200; Fig. 1) a 50 m ? 3 m patch of narrow-leaved ramets grew out into the lake to a water depth of about 1 meter. A second patch of narrow- leaved ramets occurred in stand 2100 in an area of deeper water. We suspected that individuals from these two patches of narrow-leaved, deeper water ramets would genotype to T. angustifolia. Our results demonstrate that the native cattail, T. latifolia, is the dominant cat- tail species at PCCI; 20 of 25 cattail stands were monospecific for T. latifolia. Based on variation in leaf widths we believed that many of the cattails stands at PCCI would be mixed for both parental species and their hybrid. We found evi- dence of the widespread, introduced species, T. angustifolia, in only six of the 25 stands and in two of these mixed stands the hybrid T. ? glauca was present. We had suspected that a patch of narrow-leaved individuals from the Brewster Lake stand (stand 1200; Fig. 1) was T. angustifolia but genotyping did not reveal any T. angustfolia individuals from Brewster Lake. This is likely due to random sam- pling of our collected tissue samples for genotyping which may have excluded individuals from this patch. Genotyping did however reveal the patch of narrow- leaved individuals from stand 2100 to be T. angustifolia. It should be noted that for most individuals we only analyzed a single RAPD locus and with only a sin- gle analyzed locus it is possible that individuals identified as T. latifolia could in fact be T. angustifolia if introgression has occurred. Therefore, there is a chance that our species assignments could be more complex if introgression is rampant throughout the cattail stands at PCCI. Additionally, our % species composition (Table 1) was calculated as a percentage of each species per the total number of RAPD identified individuals per stand. Since several stands had limited RAPD genotyping the % species composition should be considered tentative with re- gard to the entire stand. The homogenous T. latifolia stands were primarily associated with the wet- lands surrounding Brewster Lake and Cedar Creek. It is likely that these stands experience lesser disturbance from changes in hydrology and water depth. Typha latifolia is reported to have superior growth and competitive ability over T. an- gustifolia in shallow, undisturbed, water habitats (Grace and Wetzel (1989). By contrast, T. angustifolia was shown to be competitively superior to T. latifolia in shallow water habitats when they were highly eutrophic (Weisner 1993). We found that mixed parental stands were located in areas that are likely experienc- ing greater disturbance because of their proximity to the PCCI trail system, roads, and other anthropogenic influences such as culverts which may be acting to alter hydrology and increase sedimentation. Stand 1700S was of particular in- terest because it is a small retention pond that was constructed on the PCCI prop- erty sometime between 1999 and 2001 (Brown, Pers. Comm.). The very nature of a retention pond suggests fluctuating water levels and sedimentation. Since the hybrid T. ? glauca can exist throughout the range of water depths occupied by its parental species it is not surprising that we identified a hybrid individual from this site. It is well documented that when the native T. latifolia occurs sympatrically with the introduced T. angustifolia they can produce the hybrid T. ? glauca

Page  97

(Galatowitsch 1999; Kuehn et al. 1999; Olsen et al. 2009). Typha angustifolia it- self can out-compete the native cattail, T. latifolia, and the hybrid T. ? glauca can out-compete both parental species (Grace and Harrison 1986). Because the 2003 survey of vascular plants at PCCI (Slaughter and Skean 2003) did not iden- tify the occurrence of T. angustifolia, we believe it has moved onto the property within the last ten years. Since a large patch of T. angustifolia is directly adjacent to the trail system (stand 2100) it is unlikely that the species was overlooked in the 2003 survey. Therefore, the presence of T. angustifolia is cause for concern regarding the long-term health of the native T. latifolia cattail marshes at PCCI. While the identification of hybrid individuals based on morphological characters alone is not reliable, the leaf width and spike-gap differences between the parental species seem to allow for the accurate identification of T. angustifolia in the field. We recommend the removal of all Typha angustifolia individuals and identified hybrids from the PCCI property in order to ensure the long-term health of the T. latifolia cattail marshes. The removal of T. angustifolia, the maternal parent of the hybrid, should prevent the formation of additional hybrid individu- als on the PCCI property. At the very least, we recommend that PCCI undertake a monitoring program to record the health of, and/or changes to, the cattail marshes on their property. A cattail marsh monitoring program should keep track of fluctuations in hy- drology and water levels, as well as, the numbers and locations of the invasive cat- tail species. Typha angustifolia has been shown to prefer deeper water than T. lat- ifolia (Travis et al. 2010). Therefore, if water levels increase around Brewster Lake or Cedar Creek, such as might be caused by beaver dams, the disturbance could create T. angustifolia appropriate habitat. Additionally, if a cattail marsh regularly receives additional water and nutrients from runoff, such as through the existing culvert systems, they are likely to experience increased growth of the invasive cat- tail species. An appropriate control technique would be to manipulate water levels so that the preferred T. angustifolia habitat is not available for colonization. Optimally, all non-native cattail species would be removed from the PCCI property. According to the USDA NRCS Plant Guide (USDA, NRCS 2012) cat- tails can be removed by mowing the plants after the flowering heads are well- formed but before they have reached sexual maturity. This mowing treatment should be followed by a second mowing after 0.5 m–1 m (2–3 feet) of new growth has occurred (typically 1 month). This treatment opens up habitat for other emergent vegetation and has been shown to be about 75% effective in eliminating undesirable cattail stands. Additionally, herbicides can be applied to the rooted portion of the cut ramets which should kill the submerged rhizome and prevent regrowth. We predict that without intervention, T. angustifolia and T. ? glauca will spread through the marsh complexes at PCCI altering their species composition and the foraging behavior of many of the animal species dependent on them.

ACKNOWLEDGEMENTS

The authors thank the Willard G. Pierce and Jesse M. Pierce Foundation and the Pierce Cedar Creek Institute Undergraduate Research Grants for the Environment program for providing us with

Page  98

financial support and for the opportunity for the first two authors to do independent undergraduate research. We also thank Grand Rapids Community College for use of their laboratory facilities, and the Michigan Botanical Foundation for financial support.

LITERATURE CITED

Dickerman, J. A. (1982). The pattern and process of clonal growth in a common cattail (Typha lati- folia L.) population. East Lansing, MI: Michigan State University. 95 p. Dissertation. Doyle, J. J. and J. L. Doyle. (1987). A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical bulletin 19: 11–15. Ellstrand, N. C. and K. A. Schierenbeck. (2000). Hybridization as a stimulus for the evolution of in- vasiveness in plants. Proceedings of the National Academy of Sciences of the United States of America 97: 7043–7050. Galatowitsch, S. M., N. O. Anderson, and P. D. Ascher. (1999). Invasiveness in wetland plants in temperate North America. Wetlands 19: 733–755. Grace, J. B. (1989). Effects of water depth on Typha latifolia and Typha domingensis. American Journal of Botany. 76: 762–768. Grace, J. B. and R. G. Wetzel. (1982). Variations in growth and reproduction within populations of two rhizomatous plant species: Typha latifolia and Typha angustifolia. Oecologia 53: 258–263. Grace, J. B. and J. S. Harrison. (1986). The biology of Canadian weeds. 73. Typha latifolia, Typha angustifolia L. and Typha X glauca Godr. Canadian Journal of Plant Science 66: 361–379. Horvitz, C. C., J. B. Pacarella, S. McMann, A. Freedman, and R. H. Hofstetter. (1998). Functional roles of invasive non-indigenous plants in hurricane-affected subtropical hardwood forests. Eco- logical Applications 8: 947–974. Kirk, H., C. Connonlly and J. R. Freeland. (2011). Molecular genetic data reveal hybridization be- tween Typha angustifolia and Typha latifolia across a broad spatial scale in eastern North Amer- ica. Aquatic Botany 95: 189–193. Krattinger, K. (1975). Genetic mobility in Typha. Aquatic Botany 1: 57–70. Kuehn, M. M., J. E. Minor, and B. N. White. (1999). An examination of hybridization between the cattail species Typha latifolia and Typha angustifolia using random amplified polymorphic DNA and chloroplast DNA markers. Molecular Ecology 8: 1981–1990. McManus, H. A., J. L. Seago Jr., and L. C. Marsh. (2002). Epifluorescent and histochemical aspects of shoot anatomy of Typha latifolia L., Typha angustofolia L. and Typha glauca Godr. Annals of Botany 90: 489–493. Mills, E. L., J. H. Leach, J. T. Carlton, and C. L. Secor. (1993). Exotic species in the Great Lakes: A history of biotic crises and anthropogenic introductions. Journal of Great Lakes Research 19: 1–54. Nickrent, D. L. (2006). Molecular methods in plant biology. Fourth Edition. Carbondale, IL: South- ern Illinois University. Department of Plant Biology. Olsen A., J. Paul and J. R. Freeland. (2009). Habitat preferences of cattail species and hybrids (Typha spp.) in eastern Canada. Aquatic Botany 91: 67–70. Schierenbeck, K. A. and N. C. Ellstrand. (2009). Hybridization and the evolution of invasiveness in plants and other organisms. Biological Invasions 11: 1093–1105. Selbo, S. M. and A. A. Snow. (2004). The potential for hybridization between Typha angustifolia and Typha latifolia in a constructed wetland. Aquatic Botany 78: 361–369. Shih, J. G. and S. A. Finkelstein. (2008). Range dynamics and invasive tendencies in Typha latifolia and Typha angustifolia in eastern North America derived from herbarium and pollen records. Wet- lands 28: 1–16. Slaughter, B. S. and J. D. Skean. (2003). Annotated checklist of vascular plants in the vicinity of Cedar Creek and Brewster Lake, Pierce Cedar Creek Institute, Barry County, Michigan. The Michigan Botanist 42: 127–148. Smith, G. (2000). Typhaceae. Flora of North America Vol. 22. Oxford, New York. Snow, A. A., S. E. Travis, R. Wildova, T. Fér, P. M. Sweeney, J. E. Marburger, S. Windels, B. Kubá- tová, D. E. Goldbert, and E. Mutegi. (2010). Species-specific SSR alleles for studies of hybrid cat- tails (Typha latifolia X T. angustifolia; Typhaceae) in North America. American Journal of Botany 97: 2061–2067.

Page  99

Stuckey, R. L. and D. P. Salamon. (1987). Typha angustifolia in North America: a foreigner mas- querading as a native. American Journal of Botany 74: 757. Travis, S. E., J. E. Marburger, S. Windels, and B. Kubatova. (2010). Hybridization dynamics of in- vasive cattail (Typhaceae) stands in the Western Great Lakes Region of North America: a molec- ular analysis. Journal of Ecology 98: 7–16. Weisner, S. E. B. (1993). Long-term competitive displacement of Typha latifolia by Typha angusti- folia in a eutrophic lake. Oecologia 94: 451–456. Zedler, J. B. and S. Kercher. (2004). Causes and consequences of invasive plants in wetlands: op- portunities, opportunists, and outcomes. Critical Reviews in Plant Sciences 23: 431–452. United States Department of Agriculture (USDA) NRCS. (2012). The PLANTS Database (http://plants.usda.gov, 27 September 2012). National Plant Data Team, Greensboro, NC 27401- 4901 USA.