Myriophyllum Heterophyllum Michx. (Haloragaceae): Control and Vegetative Reproduction in Southwestern Maine Jacolyn E

The University of Maine DigitalCommons@UMaine

Electronic Theses and Dissertations Fogler Library

5-2007 Myriophyllum heterophyllum Michx. (Haloragaceae): Control and Vegetative Reproduction in Southwestern Maine Jacolyn E. Bailey

Follow this and additional works at: http://digitalcommons.library.umaine.edu/etd Part of the Botany Commons, and the Water Resource Management Commons

Recommended Citation Bailey, Jacolyn E., "Myriophyllum heterophyllum Michx. (Haloragaceae): Control and Vegetative Reproduction in Southwestern Maine" (2007). Electronic Theses and Dissertations. 373. http://digitalcommons.library.umaine.edu/etd/373

This Open-Access Thesis is brought to you for free and open access by DigitalCommons@UMaine. It has been accepted for inclusion in Electronic Theses and Dissertations by an authorized administrator of DigitalCommons@UMaine. MYRIOPHYLLUM HETEROPHYLLUM MICHX. (HALORAGACEAE):

CONTROL AND VEGETATIVE REPRODUCTION

IN SOUTHWESTERN MAINE

Jacolyn E. Bailey

B.A. University of Maine Farmington, 1995

A THESIS

Submitted in Partial Fulfillment of the

Requirements for the Degree of

Master of Science

(in Ecology & Environmental Science)

The Graduate School

University of Maine

May, 2007

Advisory Committee:

Dr, Aram J.K. Calhoun, Professor of Wetland Ecology, Advisor

Dr. Ann C. Dieffenbacher-Krall, Assistant Research Professor, Climate Change Studies

11 MYRIOPHYLLUM HETEROPHYLLUM MICHX. (HALORAGACEAE):

CONTROL AND VEGETATIVE REPRODUCTION

IN SOUTHWESTERN MAINE

By Jacolyn E. Bailey

Thesis Advisor: Dr. Aram J.K. Calhoun

An Abstract of the Thesis Presented in Partial Fulfillment of the Requirements for the Degree of Master of Science (in Ecology & Environmental Science) May, 2007

Native to the southeastern United States, variable-leaf watermilfoil

(Myriophyllum heterophyllum) is an invasive species in the Northeast and has been documented in Maine lakes for twenty years. Variable-leaf watermilfoil is targeted by the Maine Department of Environmental Protection as a species of grave concern as it has aggressively colonized twenty-six water bodies in Maine. This aquatic invasive plant grows in dense mats and outcompetes native vegetation. It is causing both ecological and economic disruption to Maine's lakes and ponds. The plants clog boat motors and deter people from swimming and other water related activities.

Allofragmentation and autofragmentation occur quite extensively in this species, and contribute to its ease of dispersal. During implementation of management techniques, further fragmentation of the plants can occur. Although natural resource

in managers commonly assume a 2.5 cm fragment size as being the smallest size that can regenerate, we were unable to find any research documenting this assertion.

My research focused on two key areas for variable-leaf watermilfoil: (1) determining which control methods are most effective for removing variable-leaf watermilfoil from lakes and (2) vegetative regeneration.

We looked at three management techniques for variable-leaf watermilfoil, hand removal, cutting, and benthic mats, to determine the most effective management strategy.

Our study showed that all three methods reduced plant growth significantly. However there were no significant differences among the three management methods. Differences

were present in time and cost required to implement the strategies between benthic mats

and hand removal, as well as benthic mats and cutting. Although less expensive than benthic mats, cutting was found to be unrealistic to implement in practice because of difficulties in implementation. Determining the most effective management technique for an area depends on the extent and density of the infestation. Benthic mats provided

an excellent option for thick, large infestations, whereas hand removal was more efficient

for lighter infestations. Hand removal is best used in areas with small, high density

infestations or for selective removal in sparsely infested stands of mostly native

macrophytes. This method would also be useful during management surveys when

individual plants or small clusters of variable-leaf watermilfoil are detected. Based on

our study we suggest that the benthic barrier and hand removal methods are the most effective non-mechanical management techniques for lake associations and state agencies to incorporate into their management plans.

iv In a twenty-two week greenhouse investigation, variable-leaf watermilfoil vegetative fragments were observed to determine smallest size for regeneration. Four fragment sizes were collected from the plants (a leaf, a single whorl, 2.5-cm stem with whorls, and 5-cm stem with whorls) and two substrates, sand and top soil, were used. All fragment sizes regenerated buds with the exception of the individual leaves. Evidence that fragment regeneration from any plant fragment containing a stem node is very useful for developing long-term management strategies for Myriophyllum heterophyllwn.

Managers need to emphasize the removal of fragments generated during removal processes as well as after heavy recreational use of an infested lake in order to reduce the potential spread of M. heterophyllum.

v ACKNOWLEDGMENTS

This work would not have been possible if not for all of the volunteer divers and field help: Brad Agius, Sylvia Bailey, Pete & Lorna Boilard, Aram Calhoun, Jim

Chandler, Glenn Dixon, Michael & SueEllen Glover, Stefany Gregoire, Paul Gregory,

Karen Hahnel, Chris Hemenway, Jeremy Judd, Peter Leach, TR Morley, Jarrett Poisson,

Gabby Rigaud, Jeff Timm and particularly Chris Rigaud. Chris donated countless hours to diving in milfoil-infested waters and coming up with the Rigaud Retrieval system for the benthic barriers. Thank you to Glenn Dixon for his video and photographic expertise on Lake Arrowhead and Hogan Pond, as well as his creative solution to the greenhouse aquaria netting that kept falling in the water. Thank you to the lake associations and individuals who provided me with access to the research lakes: Lake Auburn & The

Basin - Auburn Water District (Mary Jane Dillingham), Friends of Lake Arrowhead

(Dave Sanfasen), Hogan & Whitney Pond Association, Little Sebago Lake Association,

Messalonskee Lake (Marilyn Eccles), Pleasant Pond - Friends of Cobbossee, Shagg Pond

(Jim Chandler and Scott & Thelrna Kendrick) and Thompson Lake Environmental

Association. Thank you to the Maine Department of Environmental Protection Invasive

Species Group (John McPhedran, Karen Hahnel, Paul Gregory, Roy Bouchard) for providing information on the Maine lakes with invasive infestations. Thank you to the

Maine Center for Aquatic Invasive Species and Roberta Hill for providing management information and insight on the infestations. I am very grateful to William Halteman for his statistical assistance and Sue Erich for providing insight on my plant nutrient analysis.

I am very appreciative of the support of my fellow graduate students TR Morley, Megan

vi Gahl, and Amanda Shearin for their feedback and support throughout my research.

Thank you to Ann Dieffenbacher-Krall and Katherine Webster for their advice and serving on my committee. Thank you is just not enough for Aram Calhoun, my advisor, mentor and friend. Your advice, wisdom, patience and friendship have made my experience all that much richer. This research was funded by the Department of Plant,

Soil, and Environmental Sciences, the Maine Association of Wetland Scientists and the

University of Maine, Association of Graduate Students. I send a big thank you to my mom, dad, sister and friend, Nicki Breton, for their support and providing sympathetic ears during my frustrations and successes. Finally, and most importantly, I thank my husband, Jarrett Poisson, for his support and willingness to brave leeches on Little

Sebago Lake.

« TABLE OF CONTENTS

ACKNOWLEDGEMENTS vi

LIST OF TABLES x

LIST OF FIGURES xi

Chapter

1. COMPARISON OF THREE PHYSICAL MANAGEMENT TECHNIQUES

FOR VARIABLE-LEAF WATERMILFOIL IN MAINE LAKES 1

Abstract 1

Introduction 2

Materials & Methods 5

Study Area .....5

Experimental Design , 6

Control Methods 9

Hand Removal 9

Cutting 9

BenthicMat 10

Assessment of Management Technique Effects 10

Statistical Analyses 10

Time and Cost Determination 11

Results & Discussion 11

Hand Removal 12

vni Cutting 14

BenthicMats 14

Management Recommendations 16

Chapter References 18

2. EFFECTS OF FRAGMENT SIZE ON VEGETATIVE REGENERATION

IN MYRIOPHYLLUM HETEROPHYLLUM MICHX. (HALORAGACEAE)

IN A GREENHOUSE EXPERIMENT 20

Abstract 20

Introduction 21

Materials & Methods 23

Results 25

Bud Regeneration 25

New Growth 27

Water Chemistry 28

Discussion 28

Management Implications 29

Chapter References 31

REFERENCES 32

APPENDIX: Greenhouse experiment water chemistry results 36

BIOGRAPHY OF THE AUTHOR 38

IX LIST OF TABLES

Table 1.1 Comparison of eight research lakes infested with variable-leaf

watermilfoil in southwestern Maine 6

Table 1.2 Time and cost comparison for three management techniques of variable-

leaf watermilfoil invasions in 12 m2 experimental plots in eight Maine

lakes 13

Table 1.3 Advantages and disadvantages of hand removal, cutting, and benthic mats

management techniques 16

Table 2.1 Total number of buds per fragment size and length of new growth by

substrate for Myriophyllum heterophyllum 26

Table A.l Greenhouse fragment vegetative regeneration experiment chemistry

results for top soil and sand substrate water (tested at week 11) and tap

water (tested on day 1) 37

x LIST OF FIGURES

Figure 1.1 Map of invasive aquatic plant infestations in Maine , 7

Figure 1.2 Schematic of 3 by 4 m experimental plots , 8

Figure 1.3 Comparison of plant dry weight for hand removal, cutting, benthic

mats, and control experimental plots 12

Figure 2.1 Schematic of fragment sizes of Myriophyllum heterophyllum

used in a greenhouse study 24

Figure 2.2 Average bud length by substrate and fragment size of

Myriophyllum heterophyllum in a greenhouse study evaluating

vegetative regeneration 27

XI Chapter 1 COMPARISON OF THREE PHYSICAL MANAGEMENT TECHNIQUES

FOR VARIABLE-LEAF WATERMILFOIL IN MAINE LAKES

Abstract

Variable-leaf watermilfoil, (Myriophyllum heterophyllum), native to the southeastern United States, is an invasive species in the Northeast. This invasive aquatic grows in thick, dense mats and outcompetes native vegetation. The plants clog boat motors and deter people from swimming and other water-related activities. Quantitative evaluation of control methods to determine the effectiveness of hand removal, cutting and benthic mats on variable-leaf watermilfoil is not available in the literature. We looked at these three management techniques on eight infested lakes in Maine to determine the most effective management strategy. Our study showed that all three treatments resulted in significantly lower plant re-growth than the control. No significant differences were found among the three treatments in plant re-growth or among lakes in percent re-growth of variable-leaf watermilfoil. Differences were present in time and cost required to implement the strategies between benthic mats and hand removal, as well as benthic mats and cutting. Although less expensive than benthic mats, cutting was found to be unrealistic to implement in practice because of difficulties in implementation.

Determining the most effective management technique for an area depends on the extent and density of the infestation. Benthic mats provided an excellent option for thick, large infestations, whereas hand removal was more efficient for lighter infestations. Hand removal is best used in areas with small, high density infestations or for selective removal in sparsely infested stands of mostly native macrophytes. This method would also be useful during management surveys when individual plants or small clusters of variable- leaf watermilfoil are detected. Based on our study we suggest that the benthic barrier and hand removal methods are the most effective non-mechanical management techniques for lake associations and state agencies to incorporate into their management plans.

Introduction

Variable leaf watermilfoil (Myriophyllum heterophyllum) is an invasive aquatic plant of concern in New England lakes (Moody and Les 2002). In the battle against invasive aquatic plant species, eradication is the ultimate goal. The reality is that eradication is seldom achieved (Madsen 2000). Once an introduced plant has invaded a lake, an ongoing management effort is necessary. In areas where plant removal is achieved, continuing surveillance is needed to watch for re-infestations. In the United

States research on management techniques for aquatic invasive plants has been conducted on widespread species. Hydrilla {Hydrilla verticillata), native to Africa, Australia, and parts of Asia (Madeira and others 2000), was established in the United States in Florida in 1950 (Shearer and Jackson 2006). It is now found throughout the south and west and as far north as Maine (Shearer and Jackson 2006). Despite limited success controlling hydrilla with herbicides, this species has developed herbicide resistance (Michel and others 2004). Water hyacinth (Eichornia crassipes) (Holm and others 1969), water chestnut (Trapa natans) (Madsen 1993), and Eurasian watermilfoil {Myriophyllum spicatum) (Boylen and others 1996) are also invasive species widespread in the United

States.

2 Herbicides are often used in conjunction with other management techniques such as hand removal (Helsel and others 1996). Aquatic herbicides are applied to the water where the target plant is located (Madsen and others 2000). Use of this management technique often causes public concern due to the potential of bio-magnification, environmental persistence, and minimal understanding of long-term effects (Charudattan

2001). The use of herbicides is no guarantee that eradication of the target species will

occur. In Florida, despite an extensive herbicide management program, hydrilla continues to spread to more waterbodies every year (Koschnick and others 2006).

Biological controls, using either introduced or naturalized organisms, require

extensive research and time to find an appropriate biocontrol agent and ensure that it will

not become invasive, or influence native species. For aquatic plant control, fish and

invertebrates are often the key species studied as potential biocontrol agents (Madsen and others 2000; Pipalova 2006). They are not, however, a silver bullet and are often used in

conjunction with other control methods (Nelson and Shearer 2005). There are also concerns regarding the large populations needed to achieve control, the amount of time

required for control to be achieved, and potential introduction of new pathogens to local

invertebrate populations (Madsen and others 2000).

Physical management techniques include a variety of mechanical and non-

mechanical methods. Non-mechanical methods are usually more economical than herbicides and biocontrol agents and can be put into action without rigorous controls,

however they are time and labor intensive (Madsen 2000). Mechanical methods can be costly due to machinery maintenance. They also can spread numerous plant fragments

and have negative effects on the ecosystem, such as large amounts of sediment

3 disturbance, re-suspending chemicals from the substrate, and injury to organisms

(Madsen 2000). There currently is no research evaluating the effectiveness of non- mechanical physical management techniques on variable-leaf watermilfoil.

Variable-leaf watermilfoil is native to the southeastern United States and was introduced in New England in the early 1900s (Les and Mehrhoff 1999). Although it is not as widespread as Eurasian watermilfoil, variable-leaf watermilfoil is known to be aggressive locally (Crowe and Hellquist 2000). Eurasian watermilfoil has been studied extensively for invasive traits (Galatowitsch and others 1999; Grace and Wetzel 1978), dispersal capacity (Madsen and Smith 1997), and management and eradication techniques (Helsel and others 1996; Madsen and others 2000; Nichols 1972). Whereas, variable-leaf watermilfoil has few studies on management techniques (Bugbee and others

2003).

This study evaluated the effectiveness of three physical management techniques: hand-removal, cutting (leaving roots below ground), and benthic barriers on variable-leaf watermilfoil in lakes in southwestern Maine. The Maine Department of Environmental

Protection has taken a conservative approach to herbicide use in Maine lakes and currently there are only three lakes in which herbicides have been used. Herbicides were used in these cases based on the aggressive nature of the species in the lake or the relative isolation of the lake itself. There are currently no waterbodies with variable-leaf watermilfoil being treated with herbicides in Maine, which leaves physical management techniques as the only immediate options for resource managers. Management techniques for this study were chosen for evaluation based on their ease of implementation for resource managers and lake associations, as well as their minimal

4 impact to the surrounding lake system (Nicholson 1981). Removal and benthic mat techniques are commonly used physical management methods for many invasive aquatic plant species and have had favorable results. Benthic mats block sunlight from reaching the invasive plants and eventually the plant material decomposes but they are nonselective and all covered plants including natives are killed (Madsen 2000). Once the mat is removed, native species can re-colonize the area. We additionally evaluated the effectiveness of cutting the variable-leaf watermilfoil plant at the water-substrate line, leaving the roots intact. Cutting using large mechanical mowing apparatus is commonly used for aquatic weed management (Madsen 2000) and causes large amounts of fragmentation of the plants. We were interested in developing a hand cutting technique that could be easily implemented by SCUBA divers and would minimize substrate disturbance while speeding up the removal process. Our objectives for this study were to determine which physical management technique was most effective in controlling variable-leaf watermilfoil, which technique was most cost and time effective, as well as which technique was most suitable for dense infestations versus patchy infestations.

Materials & Methods

Study Area

We selected eight lakes (Table I) in Maine, USA, from the Maine Department of

Environmental Protection's list of lakes with confirmed invasions of variable-leaf watermilfoil (http://www.maine.gov/dep/blwq/topic/invasives/doc.htm) to evaluate the efficacy of three non-mechanical control techniques. Maine has 29 lakes that have been invaded by four species of invasive aquatic plants. Variable-leaf watermilfoil has been

5 found in 26 waterbodies and the remaining three aquatic invaders (hydrilla, curly-leaf pondweed (Potamogeton crispus) and Eurasian watermilfoil) were each found in only one waterbody (Figure 1). Invaded lakes are located in the southwestern area of Maine where tourist activity and boating traffic is high. The eight research lakes represent the northern and southern extent of the invaded lakes in Maine and vary in substrate composition, surface area, and number of boat access points (Table 1). Lakes were chosen that could accommodate the experimental plots and having no prior management, as well as, no current management occurring in the experiment plots at the time of the study.

Table 1.1: Comparison of eight research lakes infested with variable-leaf watermilfoil in southwestern Maine.

Research Lake Lake Surface Area Public Launches Observed (hectares) Substrate Composition Lake Arrowhead 407 2 Sand Lake Auburn 928 3 sandy/rocky Hogan Pond 72 0 Sand Little Sebago Lake 768 2 sand/organic Messalonskee Lake 1,420 3 clay/sand Pleasant Pond 302 2 Sand Shagg Pond 26 1 leaf litter Thompson Lake 1,791 3 Organic

Experimental Design

Experimental plots were established on each study lake based on accessibility to the plots, minimal boating traffic around the study area, and having an infestation of at least 60% variable-leaf watermilfoil. On each lake, four 3 by 4 meter plots were established along a perpendicular transect extending out 20 meters from the shoreline

6 (Figure 2), in 2 to 3 m of water depth. To visually identify plot boundaries while

SCUBA diving, the corners of each plot were marked with 0.6 m orange stakes. A red buoy marked the lower right corner of each plot at the water's surface. A 3-m buffer was left between each plot. The three treatments (cutting, hand removal, and benthic mat) and control were randomly assigned to the experimental plots.

Figure 1.1: Map of invasive aquatic plant species infestations in Maine.

^XS^

• Variable-leaf watermilfoil B Eurasian watermilfoil • Hydnlla ^ Curly-leaf pondweed

7 gure 1.2: Schematic of 3 by 4 m experimental plots.

. i Plot 4 o o • :

Marker Poles

/\ Buoys

Plot 2

Plot 1

20 m

Visual marker for transect j Shoreline e.g. tree stump

8 Control Methods

We set up our experimental plots in the summer and fall of 2004, implemented the three management techniques in spring and early summer of 2005, and collected all plant matter in late summer 2006. We monitored our plots bi-weekly during the summer and fall of 2005 and spring and summer of 2006. We canoed or kayaked over each plot and used an Aquascope™ viewer to check the plots for any disturbance.

Hand Removal

We removed variable-leaf watermilfoil plants by hand, including roots, from plots by SCUBA diving to the lake-bottom. Plant matter was collected in mesh bags, then stored in plastic tubs with lake water, and transported to the laboratory for drying and weighing. We waited approximately 30 minutes for the sediment to settle after the initial removal and then conducted a sweep to locate any plants that may have been missed.

Native species were not removed from the experimental plots. This process was continued until every variable-leaf watermilfoil plant was removed from the plot.

Cutting

We cut the vegetative portion of each variable-leaf watermilfoil plant with anvil pruners at the sediment-water interface. Plants were collected in mesh bags, stored in plastic tubs with lake water, and transported to the laboratory for drying and weighing.

After initial cutting, divers waited approximately 30 minutes for the sediment to settle and did a sweep to locate additional plants. This process was continued until all variable- leaf watermilfoil plants were removed from the plot. Native species were left intact in

9 the experimental plots. The cutting method was repeated throughout the management season (summer / fall 2005) whenever any re-growth was identified.

Benthic Mat

We placed a 4 by 3 m fabric mat over the assigned plot of variable-leaf watermilfoil on each research lake. The mat was constructed of a six ounce non-woven geotextile that had six 2.4 m sections of rebar placed at 0.76 m intervals to add weight.

The rebar was held in place using zip ties. We cut small (3-5 cm) holes into every meter section to allow gases from degrading plant material to escape. The benthic mats were installed during the fall of 2005 and removed in the spring of 2006.

Assessment of Management Technique Effectiveness

We collected all plant matter in the summer of 2006. During the spring and early summer of 2006 experimental plots were not manipulated. Any re-growth of variable- leaf watermilfoil that occurred was collected during the final collection phase in the summer of 2006. Native species were not removed. Plant matter dried on screens in the

sun for 30 days, and was then placed on racks in a drying room at 30°C for an additional

30 days to provide adequate time for complete drying. We then weighed the dried plant material.

Statistical Analysis

Data analysis was performed on SAS® 9.1 using ANOVA followed by mean separation tests (LSD, a = 0.05) to determine plant weight differences among the four

10 treatments, percent of variable-leaf watermilfoil re-growth differences among the study lakes and observed substrate type, and plant weight differences among plots based on the distance from shore. Percent re-growth was estimated using plant dry weight. For each lake, the control plot plant dry weight was the baseline of 100 percent growth and each management technique plot dry weight was calculated for percent re-growth based on the control plot plant dry weight.

Time and Cost Determination

Average time of management technique per site was based on the amount of time it took two divers to implement the method. Cost per site is based on the average wage for invasive watermilfoil SCUBA divers in Maine ($35/hour/diver) multiplied by the time required to implement the management technique and cost of materials. Dive time was computed based on average time for implementation of management techniques including 30 minutes for gear set-up/break-down. Equipment costs are based on the prices of two dive bags for plant material collection ($12 / bag), two anvil pruners

($3/pruner), rebar ($6 / 2.4 m section), and benthic mat material ($10 / 12 m2).

Results & Discussion

Plant dry weight was lower in all three treatments compared to the control (p <

0.01) (Figure 3). However, plant dry weight among the three treatments did not differ (p

= 0.62). During final plant collection, some re-growth of variable-leaf watermilfoil was found in the interior portions of the experimental plots, but the majority (>60%) of plant matter was collected along the edges. The re-growth along the edges of the experimental

11 plots was likely influenced by variable-leaf watermilfoil plants immediately outside the plots. Because the study lakes varied in observed substrate type and composition we looked at percentage of re-growth in experimental plots to see if the substrate differences influenced the amount of re-growth, however they did not differ (p = 0.79), There was also no difference in plant dry weight (p = 0.77) or percentage of re-growth (p = 0.91) for plots based on distance from the shore. Comparison of percent re-growth among lakes also proved not to differ (p = 0.57).

Figure 1.3: Comparison of plant dry weight for hand removal, cutting, benthic mats, and control experimental plots.

Cutting Hand Removal Benthic Mat Control

Management Technique

Hand Removal

Hand removal had a similar site per hour cost ($97.44) as the cutting ($96.79) technique, but was considerably lower than benthic mats ($314.40). Although this is a

12 fairly inexpensive technique to implement, it is time and labor intensive (Table 2). There are different options for implementing this technique including wading into shallow areas, SCUBA diving in deeper areas, and diver-assisted suction devices. Each of these methods adds a degree of expense to the process.

Table 1.2: Time and cost comparison for three management techniques of variable-leaf watermilfoil invasions in 12 m2 experimental plot in eight Maine lakes.

Average Time / 12 m2 Site Cost/12 m2 Site Cost/Hour

Hand Removal 2 hours 10 minutes $209.50 $97.44

Cutting 2 hours 50 minutes $271.00 $96.79

Benthic Barrier 20 minutes $104.80 $314.40

Hand removal is an effective management technique for waterbodies with small, high density stands of variable-leaf watermilfoil or for selective removal in stands of mostly native macrophytes with sparse numbers of variable-leaf watermilfoil interspersed among the natives. This method would also be useful during follow up surveys of management areas when individual or small clusters of variable-leaf watermilfoil are detected. Immediate removal would decrease the opportunities for further spread of the plant.

13 The removal of invasive plants by hand is a fairly low impact management technique (Nicholson 1981). There is some disturbance to the substrate causing resuspension of sediments, however not to the same degree as mechanical methods

(Madsen 2000). During the hand removal process it is important to remove the entire root system below the substrate. An incompletely removed root system may be able to

regenerate a plant based on our field observations.

Cutting

Cutting was a slower management technique and sediment was resuspended in the

water column causing decreased visibility and making it difficult for divers to find the

substrate-water line to cut the plants. Once the initial disturbance occurred there was a 15 to 20 cm layer of disturbed sediment hovering over the substrate (personal observation).

This disturbed sediment layer made it difficult for divers to see any shorter stems that

were above the substrate. This technique was initially tested because we hypothesized that by not removing the rooted material of the variable-leaf watermilfoil plant the

substrate would be less disturbed and divers would be able to more efficiently remove the upper vegetation. There is no advantage to using this method over hand removal

techniques because sediment disturbance does occur.

Benthic Mats

Benthic mats were the most costly technique although they took the least amount of time to implement (Table 2). The mats can be put in place relatively quickly even with just two divers. Some re-colonization by watermilfoil in the benthic mat experimental

14 plots occurred, but these plants were individuals that were easily removed by hand.

During final variable-leaf watermilfoil plant matter collection, we observed that native species had also re-grown in the benthic mat sites.

Typically, a benthic mat is left over an infested area for 45-60 days during the macrophyte growing season (Madsen 2000). We left the benthic mats in place over one winter (fall 2005 to spring 2006) to determine if this timing was effective. In areas where the number of times benthic mats can be moved and placed over new variable-leaf watermilfoil areas is limited due to winter freeze of lakes, this could be a useful way to

"extend" the benthic mat placement season. By being able to add another round of benthic mat installation in the fall and removing them the following spring more area can be managed annually. Since there was a difference between the benthic mat experimental plots and the control plots we feel that this over winter usage is an effective tool, although we cannot assess whether growing season usage would have been more effective.

Gases accumulating under benthic mats may be problematic (Madsen 2000).

Rebar and sand bags are often used to counter the effect of the gases. Typically, a woven geotextile is used as benthic mat material (Eakin 1990; Eichler and others 1995). We chose a similar costing non-woven material because it had a higher water flow through rate (110gpm/ft2) as opposed to the woven material (6gpm/ft2), which might also mean a better release of gases through the fabric. In the experimental plots, there was still some lifting that occurred with the non-woven mat material. We also observed native and variable-leaf watermilfoil plants that settled on top of the benthic mats with roots that grew into the non-woven fabric. When we removed the benthic mats and tried to clean

15 them, it was difficult and in some cases impossible. Material for benthic mats is fairly expensive and the ability to re-use the material helps lower that cost. The lifespan of the mat is dependent on the type of material used. Using a material that could easily be cleaned when removed and reused for a number of installations would be much more cost effective.

Management Recommendations

Based on our findings we suggest that the benthic barrier and hand removal methods are the most effective techniques (Table 3). Hand removal would be most effectual in sparsely infested sites where selective removal is needed in order to minimize impacts to native plants. However, in areas with dense populations of invasive plants, benthic barriers were the most effective.

Table 1.3: Advantages and disadvantages of hand removal, cutting, and benthic mats management techniques.

Advantages Disadvantages

• Relatively inexpensive • Resuspension of Hand Removal • Quick implementation sediment • Low tech • Time intensive • Selective removal • Labor intensive • Relatively inexpensive • Difficult to implement Cutting • Low tech • Resuspension of • Selective removal sediment • Time intensive • Labor intensive • Quick installation • Cost Benthic Barrier • Effective for dense infestations • Nonselective • Low tech

16 Although eradication is seldom achieved, we believe variable-leaf watermilfoil infestations can be managed effectively by incorporating the use of hand removal and benthic barriers in management plans. We observed reduced variable-leaf watermilfoil plant numbers both in the current study and in Maine lakes that implemented these methods (personal observation). A longer-term study to monitor re-colonization of the experimental plots by variable-leaf watermilfoil and native macrophytes would provide managers with a better idea of the efficacy of these three management techniques.

17 Chapter References Boylen CW, Eichler LW, Sutherland JW. 1996. Physical control of Eurasian watermilfoil in an oligotrophic lake. Hydrobiologia 340:213-218. Bugbee GJ, White JC, Krol WJ. 2003. Control of Variable Watermilfoil in Bashan Lake, CT with 2,4-D: Monitoring of Lake and Well Water. Journal of Aquatic Plant Management 41:18-25. Charudattan R. 2001. Are we on top of aquatic weeds? Weed problems, control options and challenges. International symposium on the World's Worst Weeds. United Kingdom: British Crop Protection Council. Crowe GE, Hellquist CB. 2000. Aquatic and Wetland Plants of Northeastern North America: University of Wisconsin Press. Eakin HL. Effects of benthic barriers on aquatic habitat: preliminary results. In: Station UAEWE, editor; 1990 26-30 November 1990; Orlando, Florida, p 100-102. Eichler LW, Bombard RT, Sutherland JW, Boylen CW. 1995. Recolonization of the littoral zone by macrophytes following the removal of benthic barrier material. Journal of Aquatic Plant Management 33:51-54. Galatowitsch SM, Anderson NO, Ascher PD. 1999. Invasiveness in wetland plants in temperate North America. Wetlands 19(4):733-755. Grace JB, Wetzel RG. 1978. The production biology of Eurasian watermilfoil (Myriophyllum spicatum L.): A Review. Journal of Aquatic Plant Management 16:1-11. Helsel DR, Gerber DT, Engel S. 1996. Comparing spring treatments of 2,4-D with bottom fabrics to control a new infestation of Eurasian watermilfoil. Journal of Aquatic Plant Management 34(JULY):68-71. Holm LG, Wheldon LW, Blackburn RD. 1969. Aquatic weeds. Science 166:699-709. Koschnick TJ, Haller WT, Netherland MD. 2006. Aquatic plant resistance to herbicides. Aquatics 28(1 ):4-9. Les DH, Mehrhoff LJ. 1999. Introduction of nonindigenous aquatic vascular plants in southern New England: a historical perspective. Biological Invasions 1(2-3):281- 300. Madeira PT, Jacono CC, Van TK. 2000. Monitoring hydrilla using two RAPD procedures and the nonindigenous aquatic species database. Journal of Aquatic Plant Management 38:33-40. Madsen JD. 1993. Waterchestnut seed production and management in Watervliet Reservoir, New York. Journal of Aquatic Plant Management 31:271-272. Madsen JD. 2000. Advantages and disadvantages of aquatic plant management techniques. Environmental Laboratory, US Army Corps of Engineers. Report nr ERDC/ELMP-00-1. Madsen JD, Crosson HA, Hamel KS, Hilovsky MA, Welling CH. 2000. Management of Eurasian watermilfoil in the United States using native insects: state regulatory and management issues. Journal of Aquatic Plant Management 38:121-124. Madsen JD, Smith DH. 1997. Vegetative spread of Eurasian watermilfoil colonies. Journal of Aquatic Plant Management 35:63-68. Michel A, Arias RS, Scheffler BE, Duke SO, Netherland M, Dayan FE. 2004. Somatic mutation-meidationed evolution of herbicide resistance in the

18 nonindigenousinvasive plant hydrilla (Hydrilla verticilata). Molecular Ecology 13:3229-3237. Moody ML, Les DH. Evidence of hybridity in invasive watermilfoil (Myriophyllum) populations; 2002. p 14867-14871. Nelson LS, Shearer JF. 2005. 2,4-D and Mycoleptodiscus terrestris for control of Eurasian watermilfoil. Journal of Aquatic Plant Management 43:29-34. Nichols SA. Myriophyllum problems and harvesting controls in three Wisconsin Lakes.; 1972 1972. p 62-63. Nicholson SA. 1981. Effects of Uprooting on Eurasian Watermilfoil. Journal of Aquatic Plant Management 19:57-59. Pipalova I. 2006. A review of grass carp use for aquatic weed control and itm impact on water bodies. Journal of Aquatic Plant Management 44:1-12. Shearer JF, Jackson MA. 2006. Liquid culturing of microsclerotia of Mycoleptodiscus terrestis a potential biological control agent for the management of hydrilla. Biological Control 38:298-306.

19 Chapter 2 EFFECTS OF FRAGMENT SIZE ON VEGETATIVE REGENERATION IN

MYRIOPHYLLUM HETEROPHYLLUM MICHX. (HALORAGACEAE)

IN A GREENHOUSE EXPERIMENT

Abstract

Myriophyllum heterophyllum has aggressively colonized waterbodies throughout

New England replacing native aquatic plants and negatively affecting recreational uses in lakes and rivers. Allofragmentation and autofragmentation occur quite extensively in this species and contribute to its spread. During implementation of management techniques, further fragmentation of the plants can occur. Currently managers focus on collecting fragments larger than 2.5 cm citing this as being the smallest fragment size that can regenerate. In a twenty-two week experiment in aquaria, Myriophyllum heterophyllum vegetative fragments were observed to determine smallest size for regeneration and whether substrate type affected growth of the regenerated buds. Four fragment sizes were tested: a leaf, a single whorl, 2.5-cm stem with whorls, and 5-cm stem with whorls and two substrates: sand and top soil. All fragment sizes regenerated buds with the exception of the single leaves. Evidence that fragment regeneration from any plant fragment containing a stem node, no matter how small, is very useful for developing long-term management strategies for Myriophyllum heterophyllum. Managers need to emphasize the removal of fragments generated during removal processes as well as after heavy recreational use of an infested lake in order to reduce the potential spread of M. heterophyllum.

20 Introduction

Variable-leaf watermilfoil {Myriophyllum heterophyllum) is an aggressive invasive macrophyte that easily spreads and is a challenging problem for resource managers. This plant is a member of the Haloragaceae family, submerged aquatic plants which have different submergent and emergent leaf forms (Aiken 1981). Myriophyllum heterophyllum is native to the southeastern United States and is considered invasive in the Northeast and Northwest. There are at least two other species of watermilfoil that are invasive in the United States and New England, M. spicatum and M. aquaticum.

Typically found in shallow littoral zones, M. heterophyllumjnay reach lengths of four meters in Maine (personal observation) and can grow in dense mats out-competing native aquatic vegetation (Cameron and Berg Stack 2005). The plant is quite prolific and can grow up to 2.5 cm per day in optimal conditions (Les and Mehrhoff 1999).

Myriophyllum heterophyllum reproduces both sexually and vegetatively, however, vegetative regeneration is a dominant mode of reproduction (Crowe and Hellquist 2000;

Les and Mehrhoff 1999).

A number of management strategies have been employed to manage

Myriophyllum heterophyllum as an aquatic weed, including chemical and physical removal techniques. Physical management techniques include both mechanical

(harvesters and suction dredge) and non-mechanical (hand removal and benthic barriers, fabric blankets laid over the plants on the lake bottom) methods. However, these practices may be contributing to the spread of M. heterophyllum during implementation.

Myriophyllum heterophyllum is a brittle plant and fragments are easily broken off by wind and wave action as well as boating and other recreational activities. These

21 fragments are then readily moved around by people, animals, and water currents.

Fragments that wash up along shorelines and get stranded often form a terrestrial morph that is a much smaller compact version of the plant (JEB personal observations). Natural resource managers generally assume that a fragment size of 2.5 cm is required for M. heterophyllum regeneration. However, we were unable to find studies in the literature determining the smallest size necessary for regeneration. Previous studies have shown vegetative regeneration via propagule production and fragmentation in other aquatic plants (Barrat-Segretain and Bornette 2000; Kane and others 1991; Madsen and Smith

1997) and even regeneration of fragments in watermilfoil species (Barrat-Segretain and

Bornette 2000; Barrat-Segretain and others 1998; Madsen and others 1988), but they have not established a minimum size for regeneration.

The impact of substrate type on rooted submerged aquatic species has been studied fairly extensively (Aiken and Picard 1980; Barko 1991; Spencer and Ksander

1995). In previous studies of rooted Myriophyllum plants, plant height varied over different substrate types and suggested that nutrient levels and other substrate characteristics are important controls on the growth of the plants (Aiken and Picard 1980;

Barko and Smart 1986). Fragment growth in low nutrient waters has been studied by

Madsen et al.(l 988) but we were unable to find literature on the influence of substrate type on available nutrients in the water for fragment regeneration. By understanding how substrate affects nutrient availability in the water and thereby regeneration of fragments, we may be able to inform optimal management techniques for use in infested lakes based on sediment characteristics.

22 The objectives of this study were to experimentally determine the minimum regenerative length of M. heterophyllum fragments and to determine if substrate type affected growth of the regenerated buds. We used glass aquaria in a greenhouse, to look at four fragment sizes and determine which would develop roots and buds. We also compared two substrate types to determine whether rooted fragments grew more robustly in one over the other.

Materials & Methods

Approximately 50 M. heterophyllum plants were collected by hand from Lake

Auburn, Auburn, Maine, USA, in September 2005. Lake Auburn is located within 64 km of ten other M. heterophyllum infested lakes. The plants were stored in open containers and transported in lake water to the lab. They were then stored for a day at room temperature (20°C) to acclimate to greenhouse conditions.

Twenty-four glass aquaria (36 cm x 24 cm x 40 cm) were set up with 5 cm of sediment placed on the bottom. We used generic bagged sand and unaugmented bagged top soil in the tanks with 12 aquaria for each sediment type. Sand and top soils were used to represent two.extreme types of lake substrates present in Maine infested lakes.

Sediment was covered with 25 cm of tap water and left to acclimate for 21 days prior to adding fragments. To maintain constant water levels, tap water was regularly added to the aquaria. Because algal growth occurred during week one of the experiment, we added 6 ml of Aquarium Pharmaceuticals Algal Destroyer™ algaecide to all tanks biweekly.

23 We tested bud regeneration in four fragment sizes: (1) a single leaf, (2) a single whorl, (3) a 2.54 cm section of stem and leaves, and (4) a 5 cm section of stem and leaves

(Figure 1). Ten fragments were placed on the water's surface of each tank. We set up three replicates for each fragment and sediment type and randomly assigned their placement in the greenhouse. A mesh fabric was placed over each aquarium and held in place with an elastic cord to limit outside debris entering into the aquaria. Surface water was gently mixed with two clockwise stirs to simulate wave and wind action each week.

The greenhouse was maintained at ~30°C and with natural light on an 8.5 h light: 15.5 h dark cycle.

Figure 2.1: Schematic of fragment sizes of Myriophyllum heterophyllum used in a greenhouse study. (Plant drawings from Britton, N.L., and A. Brown. 1913. Illustrated flora of the northern states and Canada. Vol. 2: 616)

We monitored fragment root and bud development and location of fragments in the tank over a 22-week period. At the completion of the experiment, we removed each

24 fragment and noted number of buds, length of new growth, and rooting. At weeks 1 and

11, we tested pH, alkalinity, nutrient content (ammonium nitrogen, nitrate, sulfate, orthophosphate), and major ions (calcium, chloride, potassium, magnesium, sodium, aluminum, iron, and manganese) in the water. Samples were analyzed by the Maine

Agricultural and Forest Experiment Station Analytical Laboratory at the University of

Maine.

Data analysis was performed on SAS® 9.1 using ANOVA followed by mean separation tests (LSD, a = 0.05) to determine differences of number of buds developed among the four fragment sizes. A Wilcoxon-Mann-Whitney statistical test was done to analyze new growth lengths between top soil tank water and sand tank water.

Results

Bud Regeneration

Myriophyllum heterophyllum plants produced buds from all fragment sizes with the exception of a single leaf (Table 1). Regeneration data indicated that the number of buds that grew differed significantly among fragment sizes (p < 0.01). Single leaf fragments slowly disintegrated and disappeared with no regeneration. The smallest fragments that grew buds were single whorls, which were 0.2 to 0.3 cm long. On many of the whorls, the leaves fell off and only the stem, and in some cases growing buds, were left. Buds did not begin growing on the whorl fragments until week five and in only one tank. By week seven, whorl fragments in five of the six tanks were regenerating. Five of the 2.5 cm fragment tanks had bud regeneration starting in week five. By week seven, all 2.5 cm fragment tanks had bud regeneration occurring. Buds grew on the 5-cm

25 fragments in five of the six tanks within three weeks. By week five, all 5-cm fragment tanks had buds growing on fragments.

Table 2.1: Total number of buds per fragment size and length of new growth by substrate for Myriophyllum heterophyllum.

New Growth Length Total Number (cm) of Buds Mean Range TOP SOIL 100 4.8 0.1-29.6

Leaf 0

Whorl 26 6.10 0.9-29.6

2.5cm 33 2.62 0.6-5.5

5cm 41 5.82 0.1-21.9

SAND 96 1.2 0.1-2.7

Leaf 0

Whorl 7 1.58 0.9-2.1

2.5cm 24 1.05 0.1-2.7

5cm 65 1.16 0.1-2.5

Roots appeared on whorl fragments within one week and on 5 -cm fragments within two weeks. By week four, roots were growing in eight tanks (2-5 cm, 1-2.5 cm,

1-whorl). None of the fragments rooted into the substrate, although some did settle to the bottom of the aquarium but became re-suspended over time. New roots did not grow in additional tanks after week four, which coincided with the time bud growth began to occur more vigorously.

26 New Growth Length

Fragments growing in top soil aquaria had both greater average new growth length and overall longest new growth length than fragments growing in the sand aquaria

(Figure 2). There was a significant difference (p < 0.01) between new growth lengths in top soil tanks versus sand tanks.

Figure 2.2: Average bud length by substrate and fragment size of Myriophyllum heterophyllum in a greenhouse study evaluating vegetative regeneration.

8 7 ft 6 H5cm SE5 i •n ^ • 2.5cm 3 • Whorl 2 1 0 Sand To•p SoiJl Substrate

27 Water Chemistry

Alkalinity, pH, calcium, potassium, and magnesium concentrations were lower in sand treatment tanks than in top soil treatment tanks and were all lower in fresh tap water samples than in aquaria water from either treatment. Alkalinity was 2 to 2.5 times higher in the top soil tanks than in the sand tanks. Sulfate was the only nutrient lower in the top soil treatment than the sand treatment and tap water samples. Ammonium and phosphorus levels were below detection limits in all samples. Concentrations of the remaining nutrient (nitrate) and major ions (chloride, sodium, aluminum, iron, and manganese) were similar for both treatments, however there was a difference in concentrations between the two substrate treatments and the tap water samples. No further analysis was performed on the water chemistry data since they were inconclusive.

Discussion

This study demonstrated that a fragment composed of a single stem node has the ability to regenerate. The inability of single leaves to regenerate was expected because the meristematic tissue required to develop buds is normally located in the nodes

(Sculthorpe 1967). Whether or not whorl sized fragments typically regenerate in the field is unknown. However, knowing that such a small size has the ability to regenerate underlines the importance of fragment removal in management efforts.

We expected that the fragments would settle to the bottom of the aquaria and root into the substrate. Although the fragments did settle to the substrate, they would often become re-suspended in the water column. Many fragments continued this up and down movement throughout the experiment. We speculate that the fragments required some

28 sort of structure to attach to in order to root. We observed this phenomenon in the lake environment as fragments caught in native grasses formed roots into the substrate, whereas in open bottom areas fragments were floating along the substrate surface.

The expectation of fragment rooting was the impetus for the differing substrate types to determine if there were any differences in growth. Although the rooting did not occur, we still observed differences in new growth lengths between the two substrate treatments. Because the fragments were not rooted into the substrate, we speculate they absorbed the necessary nutrients from the water. In M. spicatum plants, there are structures associated with the leaves which are thought to be major sites of mineral ion absorption (Grace and Wetzel 1978). New growth in tanks with top soil had greater average lengths than new growth over a sand substrate (Table 1). The maximum length new growth in top soil was 26.5 cm longer than the longest new growth over the sand. It is probable that both substrate treatments released important nutrients into the water that provided for bud development and that top soil had some additional factor that provided for the more robust new growth lengths.

Management Implications

The challenges of managing and preventing the spread of an invasive aquatic plant such as M. heterophyllum can be frustrating for managers. Understanding how and why M. heterophyllum is invasive can lead to new management methods that can help manage and potentially eradicate this species from lakes and rivers. The ability of M. heterophyllum to regenerate from a whorl fragment may be a strategy that contributes to its invasiveness. This finding helps to underline the importance of fragment collection

29 during management technique implementation and after recreation activities. Current management techniques both mechanical and non-mechanical may be contributing to the spread of M. heterophyllum. During the removal processes fragments are generated and surface crews need to be on hand to collect stray fragments. Although this seems like it would be straight forward, when removal is done on a large area of a lake, many individuals are needed to collect the stray fragments. Optimally, every last fragment is collected but this can be difficult due to sedimentation of the water column which makes seeing fragments below the surface challenging. An alternative to fragment collection by surface crews entails setting up a fragment barrier to surround the work area and prevent fragments from floating away. This process requires additional cost in materials and takes extra time to set up, disassemble and move during the plant removal processes.

However, this measure could significantly reduce the spread of plants from fragments generated during removal processes.

30 Chapter References

Aiken SG. 1981. A Conspectus of Myriophyllum Haloragaceae in North America. Brittonia33(l):57-69. Aiken SG, Picard RR. 1980. The influence of substrate on the growth and morphology of Myriophyllum exalbescens and Myriophyllum spicatum. Canadian Journal of Botany 58:1111-1118. Barko JW. 1991. Sediment interactions with submersed macrophytes. US Army Engineer Waterways Experiment Station. Barko JW, Smart M. 1986. Effects of sediment composition on growth of submersed aquatic vegetation. Department of the Army. Barrat-Segretain M-H, Bornette G. 2000. Regeneration and colonization abilities of aquatic plant fragments: effect of disturbance seasonality. Hydrobiologia 421:31 - 39. Barrat-Segretain M-H, Bornette G, Hering-Vilas-Boas A. 1998. Comparative abilities of vegetative regeneration among aquatic plants growing in disturbed habitats. Aquatic Botany 60:201-211. Cameron D, Berg Stack L. 2005. Maine Invasive Plants Bulletin #2530. University of Maine Cooperative Extension. Canfield DE, Jr., Hoyer MV. 1992. Aquatic macrophytes and their relation to the limnology of Florida lakes. Final report. Tallahasse, Florida. Crowe GE, Hellquist CB. 2000. Aquatic and Wetland Plants of Northeastern North America: University of Wisconsin Press. Grace JB, Wetzel RG. 1978. The production biology of Eurasian watermilfoil (Myriophyllum spicatum L.): A Review. Journal of Aquatic Plant Management 16:1-11. Kane ME, Gilman EF, Jenks MA. 1991. Regenerative capacity of Myriophyllum aquaticum tissues cultured In Vitro. Journal of Aquatic Plant Management 29:102-109. Les DH, Mehrhoff LJ. 1999. Introduction of nonindigenous aquatic vascular plants in southern New England: a historical perspective. Biological Invasions 1(2-3):281- 300. Madsen JD, Eichler LW, Boylen CW. 1988. Vegetative spread of Eurasian watermilfoil in lake George, New York. Journal of Aquatic Plant Management 26:47-50. Madsen JD, Smith DH. 1997. Vegetative spread of Eurasian watermilfoil colonies. Journal of Aquatic Plant Management 35:63-68. Sculthorpe CD. 1967. The Biology of Aquatic Vascular Plants. London: Edward Arnold (Publishers) Ltd. Spencer DF, Ksander GG. 1995. Influence of propagule size, soil fertility, and photoperiod on growth and propagule production by three species of submersed macrophytes. Wetlands 15(2): 134-140.

31 REFERENCES

Boylen CW, Eichler LW, Sutherland JW. 1996. Physical control of Eurasian watermilfoil in an oligotrophic lake. Hydrobiologia 340:213-218. Bugbee GJ, White JC, Krol WJ. 2003. Control of Variable Watermilfoil in Bashan Lake, CT with 2,4-D: Monitoring of Lake and Well Water. Journal of Aquatic Plant Management 41:18-25. Charudattan R. 2001. Are we on top of aquatic weeds? Weed problems, control options and challenges. International symposium on the World's Worst Weeds. United Kingdom: British Crop Protection Council. Crowe GE, Hellquist CB. 2000. Aquatic and Wetland Plants of Northeastern North America: University of Wisconsin Press. Eakin HL. Effects of benthic barriers on aquatic habitat: preliminary results. In: Station UAEWE, editor; 1990 26-30 November 1990; Orlando, Florida, p 100-102. Eichler LW, Bombard RT, Sutherland JW, Boylen CW. 1995. Recolonization of the littoral zone by macrophytes following the removal of benthic barrier material. Journal of Aquatic Plant Management 33:51-54. Galatowitsch SM, Anderson NO, Ascher PD. 1999. Invasiveness in wetland plants in temperate North America. Wetlands 19(4):733-755. Grace JB, Wetzel RG. 1978. The production biology of Eurasian watermilfoil (Myriophyllum spicatum L.): A Review. Journal of Aquatic Plant Management 16:1-11. Helsel DR, Gerber DT, Engel S. 1996. Comparing spring treatments of 2,4-D with bottom fabrics to control a new infestation of Eurasian watermilfoil. Journal of Aquatic Plant Management 34(JULY):68-71. Holm LG, Wheldon LW, Blackburn RD. 1969. Aquatic weeds. Science 166:699-709. Koschnick TJ, Haller WT, Netherland MD. 2006. Aquatic plant resistance to herbicides. Aquatics 28(1 ):4-9. Les DH, Mehrhoff LJ. 1999. Introduction of nonindigenous aquatic vascular plants in southern New England: a historical perspective. Biological Invasions 1(2-3):281- 300. Madeira PT, Jacono CC, Van TK. 2000. Monitoring hydrilla using two RAPD procedures and the nonindigenous aquatic species database. Journal of Aquatic Plant Management 38:33-40. Madsen JD. 1993. Waterchestnut seed production and management in Watervliet Reservoir, New York. Journal of Aquatic Plant Management 31:271-272. Madsen JD. 2000. Advantages and disadvantages of aquatic plant management techniques. Environmental Laboratory, US Army Corps of Engineers. Report nr ERDC/ELMP-00-1. Madsen JD, Crosson HA, Hamel KS, Hilovsky MA, Welling CH. 2000. Management of Eurasian watermilfoil in the United States using native insects: state regulatory and management issues. Journal of Aquatic Plant Management 38:121-124. Madsen JD, Smith DH. 1997. Vegetative spread of Eurasian watermilfoil colonies. Journal of Aquatic Plant Management 35:63-68. Michel A, Arias RS, Scheffler BE, Duke SO, Netherland M, Dayan FE. 2004. Somatic mutation-meidationed evolution of herbicide resistance in the

32 nonindigenousinvasive plant hydrilla (Hydrilla verticilata). Molecular Ecology 13:3229-3237. Moody ML, Les DH. Evidence of hybridity in invasive watermilfoil (Myriophyllum) populations; 2002. p 14867-14871. Nelson LS, Shearer JF. 2005. 2,4-D and Mycoleptodiscus terrestris for control of Eurasian watermilfoil. Journal of Aquatic Plant Management 43:29-34. Nichols SA. Myriophyllum problems and harvesting controls in three Wisconsin Lakes.; 1972 1972. p 62-63. Nicholson SA. 1981. Effects of Uprooting on Eurasian Watermilfoil. Journal of Aquatic Plant Management 19:57-59. Pipalova I. 2006. A review of grass carp use for aquatic weed control and itm impact on water bodies. Journal of Aquatic Plant Management 44:1-12. Shearer JF, Jackson MA. 2006. Liquid culturing of microsclerotia of Mycoleptodiscus terrestis a potential biological control agent for the management of hydrilla. Biological Control 38:298-306. Aiken SG. 1981. A Conspectus of Myriophyllum Haloragaceae in North America. Brittonia33(l):57-69. Aiken SG, Picard RR. 1980. The influence of substrate on the growth and morphology of Myriophyllum exalbescens and Myriophyllum spicatum. Canadian Journal of Botany 58:1111-1118. Barko JW. 1991. Sediment interactions with submersed macrophytes. US Army Engineer Waterways Experiment Station. Barko JW, Smart M. 1986. Effects of sediment composition on growth of submersed aquatic vegetation. Department of the Army. Barrat-Segretain M-H, Bornette G. 2000. Regeneration and colonization abilities of aquatic plant fragments: effect of disturbance seasonality. Hydrobiologia 421:31- 39. Barrat-Segretain M-H, Bornette G, Hering-Vilas-Boas A. 1998. Comparative abilities of vegetative regeneration among aquatic plants growing in disturbed habitats. Aquatic Botany 60:201-211. Boylen CW, Eichler LW, Sutherland JW. 1996. Physical control of Eurasian watermilfoil in an oligotrophic lake. Hydrobiologia 340:213-218. Bugbee GJ, White JC, Krol WJ. 2003. Control of Variable Watermilfoil in Bashan Lake, CT with 2,4-D: Monitoring of Lake and Well Water. Journal of Aquatic Plant Management 41:18-25. Cameron D, Berg Stack L. 2005. Maine Invasive Plants Bulletin #2530. University of Maine Cooperative Extension. Charudattan R. 2001 .Are we on top of aquatic weeds? Weed problems, control options and challenges. International symposium on the World's Worst Weeds. United Kingdom: British Crop Protection Council. Crowe GE, Hellquist CB. 2000. Aquatic and Wetland Plants of Northeastern North America: University of Wisconsin Press. Eakin HL. Effects of benthic barriers on aquatic habitat: preliminary results. In: Station UAEWE, editor; 1990 26-30 November 1990; Orlando, Florida, p 100-102.

33 Eichler LW, Bombard RT, Sutherland JW, Boylen CW. 1995. Recolonization of the littoral zone by macrophytes following the removal of benthic barrier material. Journal of Aquatic Plant Management 33:51-54. Galatowitsch SM, Anderson NO, Ascher PD. 1999. Invasiveness in wetland plants in temperate North America. Wetlands 19(4):733-755. Grace JB, Wetzel RG. 1978. The production biology of Eurasian watermilfoil (Myriophyllum spicatum L.): A Review. Journal of Aquatic Plant Management 16:1-11. Holm LG, Wheldon LW, Blackburn RD. 1969. Aquatic weeds. Science 166:699-709. Kane ME, Gilman EF, Jenks MA. 1991. Regenerative capacity of Myriophyllum aquaticum tissues cultured In Vitro. Journal of Aquatic Plant Management 29:102-109. Koschnick TJ, Haller WT, Netherland MD. 2006. Aquatic plant resistance to herbicides. Aquatics 28(1 ):4-9. Les DH, Mehrhoff LJ. 1999. Introduction of nonindigenous aquatic vascular plants in southern New England: a historical perspective. Biological Invasions 1(2-3):281- 300. Madeira PT, Jacono CC, Van TK. 2000. Monitoring hydrilla using two RAPD procedures and the nonindigenous aquatic species database. Journal of Aquatic Plant Management 38:33-40. Madsen JD. 1993. Waterchestnut seed production and management in Watervliet Reservoir, New York. Journal of Aquatic Plant Management 31:271-272. Madsen JD. 2000. Advantages and disadvantages of aquatic plant management techniques. Environmental Laboratory, US Army Corps of Engineers. Report nr ERDC/ELMP-00-1. Madsen JD, Crosson HA, Hamel KS, Hilovsky MA, Welling CH. 2000. Management of Eurasian watermilfoil in the United States using native insects: state regulatory and management issues. Journal of Aquatic Plant Management 38:121-124. Madsen JD, Eichler LW, Boylen CW. 1988. Vegetative spread of Eurasian watermilfoil in lake George, New York. Journal of Aquatic Plant Management 26:47-50. Madsen JD, Smith DH. 1997. Vegetative spread of Eurasian watermilfoil colonies. Journal of Aquatic Plant Management 35:63-68. Michel A, Arias RS, Scheffler BE, Duke SO, Netherland M, Dayan FE. 2004. Somatic mutation-meidationed evolution of herbicide resistance in the nonindigenousinvasive plant hydrilla (Hydrilla verticilatd). Molecular Ecology 13:3229-3237. Moody ML, Les DH. Evidence of hybridity in invasive watermilfoil {Myriophyllum) populations; 2002. p 14867-14871. Nelson LS, Shearer JF. 2005. 2,4-D and Mycoleptodiscus terrestris for control of Eurasian watermilfoil. Journal of Aquatic Plant Management 43:29-34. Nicholson SA. 1981. Effects of Uprooting on Eurasian Watermilfoil. Journal of Aquatic Plant Management 19:57-59. Pipalova I. 2006. A review of grass carp use for aquatic weed control and itm impact on water bodies. Journal of Aquatic Plant Management 44:1-12. Sculthorpe CD. 1967. The Biology of Aquatic Vascular Plants. London: Edward Arnold (Publishers) Ltd.

34 Shearer JF, Jackson MA. 2006. Liquid culturing of microsclerotia of Mycoleptodiscus terrestis a potential biological control agent for the management of hydrilla. Biological Control 38:298-306. Spencer DF, Ksander GG. 1995. Influence of propagule size, soil fertility, and photoperiod on growth and propagule production by three species of submersed macrophytes. Wetlands 15(2): 134-140.

35 APPENDIX Greenhouse experiment water chemistry results.

36 Table A.l: Greenhouse fragment vegetative regeneration experiment chemistry results for top soil and sand substrate water (tested at week 11) and tap water (tested on day 1).

Top Soil Sand Tap Water Treatment Treatment Samples Tanks Tanks MEAN RANGE MEAN RANGE MEAN RANGE PH 8.5 8.39-8.60 8.18 8.07-8.3 8.03 7.98-8.04

CaCO3 232.25 212-256 72.58 67-82 55.17 53-57 (mg/L) 1.8 1.4-2.5 5.6 5.3-6.3 3.6 3.5-3.6 so4-s (mg/L) Calcium 63.7 58.5-72 19.2 18-21.8 12.4 11.9-12.7 (mg/L) Potassium 40.5 38.3-44.1 30.1 28.1-35 26 25.6-26.3 (mg/L) Magnesium 11 9.3-12.6 3.8 3.3-4.4 3 2.9-3 (mg/L) N03-N 0.01 0.01 0.01 0.01 0.182 0.181-0.184 (mg/L) Chloride 34.1 28.4-45.1 35.2 28.9-40.4 20.5 17.5-33.9 (mg/L) Sodium 16.5 15.7-18.2 16.8 15.5-19.3 11.6 11.4-11.7 (mg/L) Aluminum 0.05 0.05-0.06 0.06 0.05-0.07 0.22 0.20-0.26 (mg/L) Iron 0.08 0.08-0.09 0.08 0.07-0.09 0.17 0.16-0.17 (mg/L) Manganese 0.01 0.01-0.02 0.01 0.01 0.03 0.03 (mg/L)

NH4-N <0.015 <0.015 <0.015 <0.015 <0.015 <0.015 (mg/L) <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 POrP (mg/L)

37 BIOGRAPHY OF THE AUTHOR

Jacolyn E. Bailey was born in Landstuhl, Germany on May 5, 1970. Raised in

California and northern Maine, she graduated from Carrabec High School in 1988. She earned a Bachelor of Arts degree in Environmental Science and Biology with a minor in

Chemistry from the University of Maine at Farmington in 1995, and completed a certificate in Rainforest Studies from Boston University in Yungaburra, Australia. Prior to her graduate studies, Jacolyn worked as an international trade specialist helping

Maine's environmental firms export their products and services overseas. This experience provided her with a look at the global consequences of invasive species and the need to focus on this issue. An avid fan of kayaking and rafting it was a natural for her to focus research on invasive aquatic species. After receiving her M.S. degree,

Jacolyn will be pursuing a Doctor of Philosophy in Ecology and Environmental Science from the University of Maine. She is a candidate for the Master of Science degree in

Ecology and Environmental Science from the University of Maine in May 2007.