<%BANNER%>

Physiological characteristics of herbicides and management of Old World climbing fern (Lygodium microphyllum).

Permanent Link: http://ufdc.ufl.edu/UFE0041621/00001

Material Information

Title: Physiological characteristics of herbicides and management of Old World climbing fern (Lygodium microphyllum).
Physical Description: 1 online resource (157 p.)
Language: english
Creator: Hutchinson, Jeffrey
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: areas, c14, florida, gametophytes, glyphosate, herbicide, invasive, loxahatchee, lygodium, metsulfuron, microphyllum, natural, nontarget, plants, spores, translocation
Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: PHYSIOLOGICAL CHARACTERISTICS OF HERBICIDES AND MANAGEMENT OF OLD WORLD CLIMBING FERN (LYGODIUM MICROPHYLLUM) Jeffrey T. Hutchinson Absorption and translocation patterns were determined in Old World climbing fern Lygodium microphyllum (Cav.) R. Br.; OWCF using the three herbicides most commonly used for management of this plant by land managers in Florida. Using 14C-labeled herbicides, absorption and translocation were evaluated for glyphosate, metsulfuron, and triclopyr in OWCF using five different application scenarios (cut and spray, basal spray, 25% foliar spray, 50% foliar spray, and 100% foliar spray). Triclopyr was absorbed to the greatest extent (60.3%) of applied radioactively compared to glyphosate (31.2%) and metsulfuron (19.8%). The majority of radioactivity remained in treated leaves for all herbicides with only small percentages of the absorbed radioactivity being detected in other plant parts. All three herbicides translocated acropetally and basipitally to some extent. Radioactivity, for the most part, translocated evenly throughout the plants but the greatest amount of radioactivity derived from triclopyr occurred in rhizomes when the cut and spray and basal applications were used. The radioactivity in rhizomes derived from glyphosate was greater in those treated using cut and spray. Based on autoradiographs, there was limited horizontal movement of any herbicide in the rhizomes of OWCF which may explain why re-sprouts are observed several weeks following treatment. OWCF gametophytes were highly susceptible to metsulfuron with ? 1.4% survival at concentration ? 27 mg a.i./l. Inhibition concentrations of OWCF gametophytes exposed to metsulfuron were 6.1 and 26.4 mg a.i./l for the I50 and I95 values, respectively. Survival of treated gametophytes that developed into sporophytes was ? 0.014% at concentration ? 27 mg a.i./l, but no sporophytes developed at concentrations ? 432 mg a.i./l. Inhibition concentrations of OWCF sporophytes were 5.6 and 24.1 mg a.i./l for the I50 and I95 values, respectively. Based on the total number of OWCF gametophytes treated that developed into sporophytes, the tolerance level for a concentration of 216 mg a.i./l metsulfuron is 2.0 x 10-5. This indicates that the potential for tolerant OWCF populations is high using lower concentrations as no gametophytes developed into sporophytes at the standard field treatment concentration of 432 mg a.i./l. Spores of OWCF were highly susceptible to metsulfuron, but exhibited tolerance to imazapyr, glyphosate, fluroxypyr, asulam, and triclopyr exposed in bathing solution. Spore germination rates were 0.4% for spores exposed to 0.1 g a.i./l of metsulfuron, but 0% for rates ? 0.2 g a.i./l. Reduction in spore germination was observed with all other concentrations of herbicides tested, ranging from 10.4% with triclopyr (40.0 g a.i./l) to 42.6% with asulam (4.2 g a.i./l) compared to 47.9% germination for untreated checks. Metsulfuron also exhibited residual activity on OWCF gametophyte development, growth and survival on agar containing a concentration of 0.4 g a.i./l of metsulfuron compared to untreated checks. These results may explain why greater control has been achieved with metsulfuron for control of OWCF. Aerial and ground applications of metsulfuron on Everglades tree islands were more effective in reducing OWCF cover than aerial and ground application of glyphosate. Both herbicides resulted in significant non-target damage to ground cover. There was a reduction in native climax species ground cover and an increase in ruderal species ground cover at the end of the study. High tree survival and limited reductions in canopy cover were observed on tree islands treated with metsulfuron, while low tree survival and large decreases in canopy cover were observed on tree islands treated with glyphosate. Based on the results of this study, metsulfuron should be aerially applied to for initial control of OWCF and followed up with ground treatments using glyphosate or herbicides with other modes of action.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jeffrey Hutchinson.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Langeland, Kenneth A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041621:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041621/00001

Material Information

Title: Physiological characteristics of herbicides and management of Old World climbing fern (Lygodium microphyllum).
Physical Description: 1 online resource (157 p.)
Language: english
Creator: Hutchinson, Jeffrey
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: areas, c14, florida, gametophytes, glyphosate, herbicide, invasive, loxahatchee, lygodium, metsulfuron, microphyllum, natural, nontarget, plants, spores, translocation
Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: PHYSIOLOGICAL CHARACTERISTICS OF HERBICIDES AND MANAGEMENT OF OLD WORLD CLIMBING FERN (LYGODIUM MICROPHYLLUM) Jeffrey T. Hutchinson Absorption and translocation patterns were determined in Old World climbing fern Lygodium microphyllum (Cav.) R. Br.; OWCF using the three herbicides most commonly used for management of this plant by land managers in Florida. Using 14C-labeled herbicides, absorption and translocation were evaluated for glyphosate, metsulfuron, and triclopyr in OWCF using five different application scenarios (cut and spray, basal spray, 25% foliar spray, 50% foliar spray, and 100% foliar spray). Triclopyr was absorbed to the greatest extent (60.3%) of applied radioactively compared to glyphosate (31.2%) and metsulfuron (19.8%). The majority of radioactivity remained in treated leaves for all herbicides with only small percentages of the absorbed radioactivity being detected in other plant parts. All three herbicides translocated acropetally and basipitally to some extent. Radioactivity, for the most part, translocated evenly throughout the plants but the greatest amount of radioactivity derived from triclopyr occurred in rhizomes when the cut and spray and basal applications were used. The radioactivity in rhizomes derived from glyphosate was greater in those treated using cut and spray. Based on autoradiographs, there was limited horizontal movement of any herbicide in the rhizomes of OWCF which may explain why re-sprouts are observed several weeks following treatment. OWCF gametophytes were highly susceptible to metsulfuron with ? 1.4% survival at concentration ? 27 mg a.i./l. Inhibition concentrations of OWCF gametophytes exposed to metsulfuron were 6.1 and 26.4 mg a.i./l for the I50 and I95 values, respectively. Survival of treated gametophytes that developed into sporophytes was ? 0.014% at concentration ? 27 mg a.i./l, but no sporophytes developed at concentrations ? 432 mg a.i./l. Inhibition concentrations of OWCF sporophytes were 5.6 and 24.1 mg a.i./l for the I50 and I95 values, respectively. Based on the total number of OWCF gametophytes treated that developed into sporophytes, the tolerance level for a concentration of 216 mg a.i./l metsulfuron is 2.0 x 10-5. This indicates that the potential for tolerant OWCF populations is high using lower concentrations as no gametophytes developed into sporophytes at the standard field treatment concentration of 432 mg a.i./l. Spores of OWCF were highly susceptible to metsulfuron, but exhibited tolerance to imazapyr, glyphosate, fluroxypyr, asulam, and triclopyr exposed in bathing solution. Spore germination rates were 0.4% for spores exposed to 0.1 g a.i./l of metsulfuron, but 0% for rates ? 0.2 g a.i./l. Reduction in spore germination was observed with all other concentrations of herbicides tested, ranging from 10.4% with triclopyr (40.0 g a.i./l) to 42.6% with asulam (4.2 g a.i./l) compared to 47.9% germination for untreated checks. Metsulfuron also exhibited residual activity on OWCF gametophyte development, growth and survival on agar containing a concentration of 0.4 g a.i./l of metsulfuron compared to untreated checks. These results may explain why greater control has been achieved with metsulfuron for control of OWCF. Aerial and ground applications of metsulfuron on Everglades tree islands were more effective in reducing OWCF cover than aerial and ground application of glyphosate. Both herbicides resulted in significant non-target damage to ground cover. There was a reduction in native climax species ground cover and an increase in ruderal species ground cover at the end of the study. High tree survival and limited reductions in canopy cover were observed on tree islands treated with metsulfuron, while low tree survival and large decreases in canopy cover were observed on tree islands treated with glyphosate. Based on the results of this study, metsulfuron should be aerially applied to for initial control of OWCF and followed up with ground treatments using glyphosate or herbicides with other modes of action.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jeffrey Hutchinson.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Langeland, Kenneth A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041621:00001


This item has the following downloads:


Full Text





PHYSIOLOGICAL CHARACTERISTICS OF HERBICIDES AND MANAGEMENT OF
OLD WORLD CLIMBING FERN (LYGODIUM MICROPHYLLUM)















By

JEFFREY T. HUTCHINSON


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2010































2010 Jeffrey T. Hutchinson
































To my wife, Thea, thanks for the love and support









ACKNOWLEDGMENTS

First and foremost, I thank my wife, Thea, my most trusted friend, companion,

supporter, and cosmic mate for 23+ years for her love and support. I am forever

grateful and indebted to Dr. Kenneth Langeland for offering me the opportunity to

pursue a Ph.D. in weed science. Dr. Langeland allowed me a unique opportunity to

work independently, but always maintained an open door policy.

My committee members all greatly contributed to the knowledge I learned over the

last 5+ years. Dr. Robert Stamps took several days of his time to show me the basic

culture techniques of fern propagation. Dr. Doria Gordon provided very insightful

comments that I hope are exhibited throughout much of this dissertation. Dr.

MacDonald offered great insight and comments on the design of experiments as well as

allowing me to use his lab. Dr. William Haller offered insightful comments on different

herbicides to screen.

It is nearly impossible to recall all the assistance and support I received over the

last 5+ years for field sites. Dr. Laura Brandt, Stefani Melvin, Gayle Martin, Dr. Mark

Barrett, Jeremy Conrad, Lisa Jameson, Jon Wallace and all staff members at A.R.M.

Loxahatchee N.W.R. provided the best support possible from 2005-2009. Kevin Maier,

George Pelt and Don Napier were very helpful with the initial setup of plots on tree

islands. Gayle Martin was a constant contact and went well beyond the call of duty

many times when I needed help. I can't think of a better place to do field research than

A.R.M. Loxahatchee N.W.R.

Dr. John Volin (now at the University of Connecticut) and his staff at Florida

Atlantic University provided valuable insight on culture methods of OWCF spores. Dr.

Michael Kane and his staff at the University of Florida provided valuable insight on









various methods that could be used to handle and count OWCF spores. Dr. Mete

Yilmaz at the University of Florida gladly allowed me to use his lab on a moments

notice, going well beyond the call of duty expected of him. Dr. Frank Chapman allowed

me to use about every piece of equipment in his lab without ever making a minor

grievance, and at the same time offering me very informative advice along the way. I

am very thankful for the advice Dr. Chapman gave me.

Bob Querns, David Mayo, Aimee Copper and Pedro Mendez at the University of

Florida provided assistance in the lab and field, and I am grateful for their support. Bob

Querns was very helpful with the translocation project and his help was greatly

appreciated. Thea Hutchinson provided assistance in the field multiple times digging up

OWCF plants in south Florida. Many staff and faculty members in the University of

Florida Fisheries Department allowed me to use their equipment and provided practical

advice, and I am grateful to all of them.

I would like to thank the U.S. Fish and Wildlife Service, Florida Fish and Wildlife

Conservation Commission's Bureau of Invasive Plants, South Florida Water

Management District, St. Johns River Water Management District, and the Florida

Exotic Pest Plant Council for providing funding. Without the financial support of these

agencies, these projects would not have been possible.

Finally, I want to thank my parents, Melvia Hall and Cecil Hutchinson for their

support over the years. My in-laws, now deceased, Helen and Amurat Bogushevitz

were a constant source of moral and financial support, which made life easier. My

crazy dogs, Maggie and Lola, made life fun and stress-free. I will always be grateful for

the time I got to spend with Phillip Myers, a great friend and a big source of inspiration.









TABLE OF CONTENTS

page

A C K N O W LE D G M E N T S ......................................................................................... ... 4

LIST O F TA BLES ............... ................................................................................... 9

LIST O F FIG U R ES ........................................................................................................ 12

A BSTRA CT ................................................................................................. ............... 14

CHAPTER

1 IN T R O D U C T IO N .................................................................................................... 1 7

2 ABSORPTION AND TRANSLOCATION OF GLYPHOSATE, METSULFURON,
AND TRICLOPYR IN OLD WORLD CLIMBING FERN (LYGODIUM
M IC R O P H Y LLU M ) ................................................................................................ 27

Introduction ...................................................................................... ............... 27
M materials and M ethods .......................................................................................... 30
Plant Material ................................................. 30
Absorption and Translocation......................................................................... 31
Experimental Design and Data Analysis ........................................................ 32
Results and Discussion ............................................. 33
Visual Effects of Herbicides on OW CF........................................................... 33
A utoradiography ............................................................................................. 34
A absorption ........................................................................................................ 35
T ra n s lo c a tio n .................................................................................................... 3 6

3 TOLERANCE OF OLD WORLD CLIMBING FERN SPORES AND
GAMETOPHYTES TO HERBICIDE EXPOSURE .................................. ............... 46

Introduction ...................................................................................... ............... 46
M materials and M ethods .......................................................................................... 48
S ource of S pores ............................................................................................ 48
S pore P reparation .......................................................................................... 49
Spore C ounts .......................................................................................... .. 49
G erm nation M edium ...................................................................................... 49
S pore Inoculation ........................................................................................... 50
G erm nation R ates ......................................................................................... 5 1
Sporophyte Developm ent ................................................................................. 51
Experiment 1 Tolerance of Gametophytes to Metsulfuron ............................. 52
Experiment 2 Tolerance of Spores to Selected Herbicides ............................ 53
Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium ...... 54
R e s u lts ...................................... .............................................................................. 5 5
Experiment 1 Tolerance of Gametophytes to Metsulfuron ............................. 55









Experiment 2 Tolerance of Spores to Selected Herbicides .......................... 56
Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium ...... 57
Discussion ................. .. .. ......................... .. .. ............... 58
Experiment 1 Tolerance of Gametophytes to Metsulfuron ........................... 58
Experiment 2 Tolerance of Spores to Selected Herbicides .......................... 60
Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium ...... 62

4 REPEATED HERBICIDE APPLICATION FOR CONTROL OF OLD WORLD
CLIMBING FERN AND THE EFFECTS ON NON-TARGET VEGETATION ON
EV ER G LA D E'S TR EE ISLA N D S ......................................................... ............... 83

In tro d u c tio n .......................................................................................................... .. 8 3
M materials and M ethods .................................................................. .............. 87
S tu d y S ite .................................................................................................... 8 7
Evaluation P eriods ................................................................................... 88
H e rb icide T reatm e nt ..................................................... ................ ........... 8 8
G round C over .. ...................................................................... .. .......... 89
C a no py C ove r ........................................................................................... 89
P la nt S pe cie s ........................................................................................... 9 0
N ative S hrubs and T rees .............................................................. ........... 9 1
S tatistica l A na lysis .................................................................. .. .......... 9 1
O W C F ............................................................ .................... ... ............ 9 1
C over, R ichness, and Evenness ................................................... ............... 93
Shrub and Tree Survival ............................................................... .............. 93
R e s u lts ............................................................................................................... 9 4
OW CF G round Cover ................................................................... .............. 94
N ative plant ground cover ....................................................... ............... 95
S hrub and tree ground cover .................................................. ............... 96
C a nopy cove r...................................................................................... 96
Shrub and Tree Survival ............................................................... .............. 97
Ilex ca ssine .. ................................................................... .. .......... 9 7
P e rse a p a lu stris ...................................................... ............... .......... 9 7
M yrica ce rife ra ................................................................................... 97
Rapanea punctata .................................................................. .............. 98
P la nt R ichness D yna m ics .............................................................. ............... 98
G enera l trends ................................................................................... 98
Total species richness ........................................................... .............. 98
Late succession species richness........................................... ............... 99
R udera l species richness ........................................................ ............... 99
O their non-native species richness.......................................... ............... 99
Species evenness ................................................................ .............. 100
D is c u s s io n ....................................................................................................... 1 0 0
Ground Cover .................. ....... ................................. 100
OW CF m anagem ent ...... ............... ............... .................... 100
P lant diversity dynam ics ...................................................... .............. 103
N ative ground cover............................................................. .. .......... 105
Tree Survival and Canopy Cover ................ ........................ 108









Management Implications............ ........... .................. 111

5 S U M M A R Y ..................................................................................... ............... 130

Management Recommendations...... ...... ....................... 132
A additional Research ................................................................................. 133

APPENDIX

PLANT SPECIES DOCUMENTED ON TREE ISLANDS AT A.R.M.
LOXAHATCHEE N.W.R. TREATED WITH METSULFRUON (0.08 AND 0.16
KG A.I.) AND GLYPHOSATE (2.80 AND 5.60 KG A.I.).................... .............. 135

LIST O F R EFER EN C ES ................................................................. .............. 144

BIO G RA PH ICA L SKETC H ..................................... .......................... ............... 157









LIST OF TABLES


Table page

2-1 Sources of material used in OWCF 14C translocation study........................... 42

2-2 Radioactivity (percent of applied) and standard error for total recovery, leaf
wash, absorbed, treated leaf, and translocation following application of 14C-
labeled metsulfuron, glyphosate, and triclopyr to OWCF (all treatments
c o m b in e d )...................................................................................................... 4 3

2-3 Percent 14C-metsulfuron absorption, translocation, and distribution in OWCF
for five treatment methods. Values represent the means of 10 replications ...... 43

2-4 Percent 14C-glyphosate absorption, translocation, and distribution in OWCF
for five treatment methods. Values represent the means of 10 replications ...... 44

2-5 Percent 14C-triclopyr absorption, translocation, and distribution in Old World
climbing fern for five treatment methods. Values represent the means of 10
re p lic a tio n s ........................................................................................................ 4 4

3-1 Site, county, managing agency and geographical pre-positioning system
(GPS) coordinates of locations where OWCF spores were collected.............. 65

3-2 Collection site and age of OWCF spores in weeks for each experiment ......... 66

3-3 Herbicide concentrations used to evaluate effects on Old World climbing fern
spore germ nation ................................................................... .............. 66

3-4 Effects of various concentrations of metsulfuron on germination of Old World
clim bing fern spores and sporophyte survival................................ ............... 67

3-5 Inhibition rates (15o and 195) for germination of Old World climbing fern spores
following exposure to six different herbicides based on non-linear regression... 68

4-1 Herbicide treatments and treatment frequency for aerial and ground herbicide
application on tree islands in the A.R.M. Loxahatchee National Wildlife
R e fu g e .......................................................................................................... .. 1 1 3

4-2 Effects of aerial herbicide applications on Old World climbing fern ground
cover one year after application with pre-treatment cover as the covariate...... 113

4-3 Effects of aerial and a single ground treatment of herbicide on Old World
climbing fern ground cover at one year following ground treatment and two
years following aerial application with one year post treatment cover as the
c o v a ria te ...................................................................................................... 1 1 4









4-4 Effects of aerial and two annual ground treatments of herbicide on Old World
climbing fern ground cover at year 4 with year 1 cover as the covariate and
tim e ............................................................................................................ 1 1 4

4-5 Effects of aerial and two annual ground treatments of herbicide on Blechnum
serrulatum (swamp fern) ground cover at year 4 with year 1 cover as the
cova riate a nd tim e ........................................................ ............... ......... 1 15

4-6 Effects of aerial and two annual ground treatments of herbicide on Smilax
laurifolia (green brier) ground cover at year 4 with year 1 cover as the
cova riate a nd tim e ........................................................ ............... ......... 1 15

4-7 Effects of aerial and two annual ground treatments of herbicide on Osmunda
cinnamomea (cinnamon fern) ground cover at year 4 with year 1 cover as the
cova riate a nd tim e ........................................................ ............... ......... 1 16

4-8 Effects of aerial and two annual ground treatments of herbicide on
Woodwardia virginica (Virginia chain fern) ground cover at year 4 with year 1
cover as the covariate and time...... ..... .... ..................... 116

4-9 Effects of aerial and two annual ground treatments of herbicide on Vitis
rotundifolia (muscadine) ground cover at year 4 with year 1 as the covariate
a n d tim e ........................................................................... .. ............... 1 1 7

4-10 Effects of aerial and two annual ground treatments of herbicide on Osmunda
regalis (royal fern) ground cover at year 4 with year 1 as the covariate and
tim e ............................................................................................................ 1 1 7

4-11 Effects of aerial and two annual ground treatments of herbicide on Cladium
jamaicense (sawgrass) ground cover at year 4 with year 1 cover as the
cova riate a nd tim e ........................................................ ............... ......... 1 18

4-12 Effects of aerial and two annual ground treatments of herbicide on Myrica
cerifera (wax myrtle) ground cover at year 4 with year 1 cover as the
cova riate a nd tim e ........................................................ ............... ......... 1 18

4-13 Effects of aerial and two annual ground treatments of herbicide on Persea
palustris (swamp bay) ground cover at year 4 with year 1 cover as the
cova riate a nd tim e ........................................................ ............... ......... 1 19

4-14 Effects of aerial and two annual ground treatments of herbicide on Ilex
cassine (Dahoon holly) ground cover at year 4 with year 1 cover as the
cova riate a nd tim e ........................................................ ............... ......... 1 19

4-15 Effects of aerial and two annual ground treatments of herbicide on Rapanea
punctata (myrsine) ground cover at year 4 with year 1 cover as the covariate
a n d tim e ........................................................................................................ 1 2 0









4-16 Effects of aerial and two annual ground treatments of herbicide on tree
canopy cover at year 4 with year canopy cover as the covariate and time....... 120

4-17 Survival rates (%) of the four most common shrub and tree species (llex
cassine, Persea palustris, Myrica cerifera and Rapanea punctata) at three
years post aerial herbicide treatment.................................... 121

4-18 Summary statistics for logistic regression analysis of tree variables as model
predictors of Ilex cassine (Dahoon holly) survival following herbicide
tre a tm e n t. ................................................................................................... ...... 1 2 1

4-19 Summary statistics for logistic regression analysis of tree variables as model
predictors of Persea palustris (swamp bay) survival following herbicide
tre a tm e n t. ................................................................................................... ...... 1 2 2

4-20 Summary statistics for logistic regression analysis of tree variables as model
predictors of Myrica cerifera (wax myrtle) survival following herbicide
tre a tm e n t. ................................................................................................... ...... 1 2 2

4-21 Summary statistics for logistic regression analysis of tree variables as model
predictors of Rapanea punctata (myrsine) survival following herbicide
tre a tm e n t. ................................................................................................... ...... 1 2 3

4-22 Total number of late succession, ruderal, non-native (including OWCF) and
all plant species recorded pre-treatment (year 0) to year three post treatment
for metsulfuron and glyphosate treated tree islands and untreated checks...... 124

4-23 Effects of aerial and two annual ground treatments of herbicide on total
species richness (including non-native species) at year 4 with year 1 as the
cova riate a nd tim e ........................................................ ............... ......... 12 5

4-24 Effects of aerial and two annual ground treatments of herbicide on late
succession species richness at year 4 with year 1 as the covariate and time .. 125

4-25 Effects of aerial and two annual ground treatments of herbicide on ruderal
species richness at year 4 with year 1 as the covariate and time.................. 126

4-26 Effects of aerial and two annual ground treatments of herbicide on non-native
species (other than Old World climbing fern richness at year 4 with year 1 as
the covariate and tim e .................................................. ............... .. ......... 126

4-27 Effects of aerial and two annual ground treatments of herbicide on species
evenness at year 4 with year 1 as the covariate and time............................. 127









LIST OF FIGURES


Figure page

1-1 Number of OWCF herbarium records per decade from the University of
Florida, Florida Sate University, the University of South Florida, and Fairchild
T ropical G ardens .................................................................. .............. 26

2-1 Dried plants and autoradiographs of OWCF treated basally with metsulfuron,
g lyphosate and triclopyr ................................................ ................ ........... 45

3-1 Location of collection sites of Old World Climbing Fern spores in Florida ....... 69

3-2 Germination rates of Old World climbing fern spores of different age groups .... 70

3-3 Germination rates (%) of Old World climbing fern spores collected from
different location in Florida ............................................................ .. ......... 7 1

3-4 Survival of Old World climbing fern gametophytes 30 days after (60 total
days in agar + nutrient medium) exposure to metsulfuron in agar medium ........ 72

3-5 Survival of Old World climbing fern sporophytes 120 days (60 days in agar
+ nutrient medium and 60 days in potted soil) after exposure to metsulfuron..... 73

3-6 Old World climbing fern sporophyte dry weight (grams) at 120 days (60 days
in agar + nutrient medium and 60 days in potted soil) after exposure to
metsulfuron on agar medium and transplanted to potting soil ......................... 74

3-7 Effects of metsulfuron on Old World climbing fern spore germination 30 days
p o st e x p o s u re ..................................................................................................... 7 5

3-8 Effects of glyphosate on Old World climbing fern spore germination 30 days
p o st e x p o s u re ..................................................................................................... 7 6

3-9 Effects of imazapyr on Old World climbing fern spore germination 30 days
p o st e x p o s u re ..................................................................................................... 7 7

3-10 Effects of triclopyr on Old World climbing fern spore germination 30 days
p o st e x p o s u re ..................................................................................................... 7 8

3-11 Effects of fluroxypyr on Old World climbing fern spore germination 30 days
p o st e x p o s u re ..................................................................................................... 7 9

3-12 Effects of asulam on Old World climbing fern spore germination 30 days post
e x p o s u re ........................................................................................................ .. 8 0









3-13 Germination rates of Old World climbing fern spores placed on an agar
mixed with metsulfuron (0.4 g a.i. / 1) or agar alone (untreated checks) 15 and
45 days post-treatment from spores collected at 12 locations......................... 81

3-14 Germination of Old World climbing fern spores collected Indian River
Community College (IRCC) and Jonathan Dickinson State Park (JDSP)
placed on an agar media mixed with metsulfuron (0.4 g a.i. / 1) or agar alone
(untreated checks) 45 days post-treatment ................................... ............... 82

4-1 Sampling design used to evaluate responses of ground cover (5 1.5 m),
canopy cover (> 1.5 m), and tagged trees on northern Everglade's tree
islands treated with herbicides to control Old World climbing fern................. 128

4-2 Conceptual model on the effects of OWCF on tree islands and sequence of
events following herbicide treatment. ...... ... ....................................... 129









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PHYSIOLOGICAL CHARACTERISTICS OF HERBICIDES AND MANAGEMENT OF
OLD WORLD CLIMBING FERN (LYGODIUM MICROPHYLLUM)

By

Jeffrey T. Hutchinson

August 2010

Chair: K. A. Langeland
Major: Agronomy

Absorption and translocation patterns were determined in Old World climbing fern

[Lygodium microphyllum (Cav.) R. Br.; OWCF] using the three herbicides most

commonly used for management of this plant by land managers in Florida. Using 14C-

labeled herbicides, absorption and translocation were evaluated for glyphosate,

metsulfuron, and triclopyr in OWCF using five different application scenarios (cut and

spray, basal spray, 25% foliar spray, 50% foliar spray, and 100% foliar spray). Triclopyr

was absorbed to the greatest extent (60.3%) of applied radioactively compared to

glyphosate (31.2%) and metsulfuron (19.8%). The majority of radioactivity remained in

treated leaves for all herbicides with only small percentages of the absorbed

radioactivity being detected in other plant parts. All three herbicides translocated

acropetally and basipitally to some extent. Radioactivity, for the most part, translocated

evenly throughout the plants but the greatest amount of radioactivity derived from

triclopyr occurred in rhizomes when the cut and spray and basal applications were

used. The radioactivity in rhizomes derived from glyphosate was greater in those

treated using cut and spray. Based on autoradiographs, there was limited horizontal









movement of any herbicide in the rhizomes of OWCF which may explain why re-sprouts

are observed several weeks following treatment.

OWCF gametophytes were highly susceptible to metsulfuron with 5 1.4% survival

at concentration > 27 mg a.i./l. Inhibition concentrations of OWCF gametophytes

exposed to metsulfuron were 6.1 and 26.4 mg a.i./l for the 150 and 195 values,

respectively. Survival of treated gametophytes that developed into sporophytes was 5

0.014% at concentration > 27 mg a.i./l, but no sporophytes developed at concentrations

> 432 mg a.i./l. Inhibition concentrations of OWCF sporophytes were 5.6 and 24.1 mg

a.i./l for the 150 and 195 values, respectively. Based on the total number of OWCF

gametophytes treated that developed into sporophytes, the tolerance level for a

concentration of 216 mg a.i./l metsulfuron is 2.0 x 10-5. This indicates that the potential

for tolerant OWCF populations is high using lower concentrations as no gametophytes

developed into sporophytes at the standard field treatment concentration of 432 mg

a.i./l. Spores of OWCF were highly susceptible to metsulfuron, but exhibited tolerance

to imazapyr, glyphosate, fluroxypyr, asulam, and triclopyr exposed in bathing solution.

Spore germination rates were 0.4% for spores exposed to 0.1 g a.i./l of metsulfuron, but

0% for rates > 0.2 g a.i./l. Reduction in spore germination was observed with all other

concentrations of herbicides tested, ranging from 10.4% with triclopyr (40.0 g a.i./l) to

42.6% with asulam (4.2 g a.i./l) compared to 47.9% germination for untreated checks.

Metsulfuron also exhibited residual activity on OWCF gametophyte development,

growth and survival on agar containing a concentration of 0.4 g a.i./l of metsulfuron

compared to untreated checks. These results may explain why greater control has

been achieved with metsulfuron for control of OWCF.









Aerial and ground applications of metsulfuron on Everglades tree islands were

more effective in reducing OWCF cover than aerial and ground application of

glyphosate. Both herbicides resulted in significant non-target damage to ground cover.

There was a reduction in native climax species ground cover and an increase in ruderal

species ground cover at the end of the study. High tree survival and limited reductions

in canopy cover were observed on tree islands treated with metsulfuron, while low tree

survival and large decreases in canopy cover were observed on tree islands treated

with glyphosate. Based on the results of this study, metsulfuron should be aerially

applied to for initial control of OWCF and followed up with ground treatments using

glyphosate or herbicides with other modes of action.









CHAPTER 1
INTRODUCTION

Old World climbing fern [Lygodium microphyllum (Cav.) R. Br; OWCF] is native to

Australia, Africa, Asia and Oceania (Pemberton 1998). Within its native range, OWCF is

found in wet tropical and subtropical habitat. OWCF belongs to the order Schizaeales

and the climbing fern family, Lygodiaceae, which is comprised of approximately 25

species (Smith et al. 2006). Fossil records indicate that the family is at least 65-100

million years old (Wikstr6m et al. 2002). The family is characterized by protostelic

subterranean rhizomes, a flexible and twining rachis, leaves that are two to four pinnate,

solitary sporangia on the abaxial surface covered by an indusium, and indeterminate

growth to > 20m (Page 1979; Smith et al. 2006 and 2008). Species of Lygodium are

most common in open swampy forests and forest edges (Page 1979).

Spores of Lygodiaceae are 5 80 pm, tetrahedral in shape with three layers, the

epispore, exine, and intine (Twiss 1910; Clarke 1936). OWCF has an average of 133

sori on each fertile leaflet with an average of 215 spores per sorus (Volin et al. 2004).

Spore production in OWCF occurs throughout the year in south Florida, but is highest

from September to November (Volin et al. 2004). Germination of OWCF spores

requires moisture and occurs in five to seven days following hydration (Twiss 1910).

Upon germination, OWCF spores develop into male, female or hermaphroditic haploid

gametophytes and reach sexual maturity in five to eight weeks (Lott et al. 2003;

Gandiaga et al. 2009). Sexual differentiation among gametophytes is determined by

antheridiogens (Kurumatani et al. 2001) produced by the fastest growing female

gametophytes. This promotes male gametophyte development which facilitates









outcrossing and prevents inbreeding (Schneller 2008). After fertilization, the

conspicuous diploid sporophytes are formed at 13-14 weeks (Gandiaga et al. 2009).

The rachis of Lygodium spp. arise from subterranean rhizomes, and exhibit

continuous apical growth, circumnutation, and indeterminate growth (Clarke 1936;

Mueller 1982). The climbing habit of OWCF is characterized as a "searcher

morphology" in which the rachis expands with minimal pinnae expansion to reduce the

weight of the plant until it finds support (Muller 1983). Once the rachis finds support, the

pinnae expand while the rachis continues to expand acropetally in a twining nature

(Muller 1983). This growth habit allows OWCF to growth horizontally along the ground

and shrub layers and vertically into the canopy. Pinnae along expanding rachis divide

into two pinnules with a dormant bud in between, and the terminal bud along an

expanding rachis will become active if the apical tip is damaged (Clark 1936). Cross

sections of OWCF pinnae include the presence of an upper epidermis, mesophyll cells,

and lower epidermis (Clark 1936), indicating similarity to the leaves of angiosperms.

OWCF has spread rapidly across the landscape of southern Florida and is

increasing its range into central Florida (Ferriter and Pernas 2006; Pemberton and

Ferriter 1998; Pemberton 2003). The distribution of OWCF on Everglade tree islands is

correlated with the density and spatial distribution of tree islands, indicating that the fern

is likely still in an early stage of invasion (Wu et al. 2006).

In a survey of Florida flora from 1927-1930, no reports of Lygodium spp. were

documented (Moldenke 1944). Floral surveys of Water Conservation Areas 2 and 3 in

the central Everglades failed to document any Lygodium spp. in the late 1950's









(Loveless 1959), but presently OWCF is common on many tree islands in the northern

Everglades and coastal prairies in the western Everglades.

The origin of the OWCF population in south Florida is believed to be Queensland,

Australia (Goolsby et al. 2006). OWCF was first reported in the United States in

cultivation from nurseries in Davenport (Polk County), Florida (Florida Division of Plant

Industry, Record #0-3753) and Delray Beach (Palm Beach County), Florida (University

of Florida Herbarium, FLAS P5100) in 1958. It was first reported as escaped and

invading natural areas in the vicinity of Jupiter in northern Palm Beach County, Florida

(Beckner 1968). OWCF was reported to be rare in Florida as late as 1976 when

sporulating specimens were recorded in Martin County (Lakela and Long 1976). By

1978, OWCF was reported to be naturalized and invasive in Palm Beach and Martin

Counties with cover at one site in Palm Beach County estimated to be 7.4 ha (Nauman

and Austin 1978), the first indication of the invasive potential of this fern. Initially, OWCF

was confined to wet, disturbed areas such as canals, rivers, ditches and disturbed

swamps with all sites having standing water for much of the year (Nauman and Austin

1978). By 1984, OWCF had invaded 21% of the wetlands in Jonathan Dickinson State

Park, Martin County (Brown 1984).

In 2003, the area coverage of OWCF in south Florida (all areas south of the north

section of Lake Okeechobee) was estimated to be 48,878 ha (Ferriter and Pernas

2006). In a 2004 aerial survey of Highlands and Polk Counties along the Lake Wales

Region of south-central Florida, 213 locations with Lygodium spp. were documented,

with some infestations of OWCF as large as 40-80 ha (Biehl 2004). By 2005, OWCF

was found as far north as Hillsborough County on the west coast and Volusia County on









the east coast (Wunderlin and Hansen 2008). In December 2008, an unreported

population of OWCF (FLAS 225459) was documented in northern Lake County, Florida,

ca. 25 km south of the Marion County line. This population of OWCF represents a 40

km range increase to the north in central Florida relative to the population reported by

Pemberton (2003) in east Orlando. Based on herbarium records (FLAS 2008; FSU

2008; FTG 2008; Wunderlin and Hansen 2008), there appears to have been a rapid

expansion of OWCF in the 1990's (Figure 1-1).

The herbarium records suggest a 30 year lag time between the introduction of

OWCF and the fern becoming invasive. From 1960-1973, all herbarium records for

OWCF occurrences were from Martin County, except a single herbarium record from

Highland County in 1973 (FLAS 2008; FSU 2008; FTG 2008). From 1975-1979, all

herbarium records were documented from Palm Beach County (FLAS 2008; FSU 2008;

FTG 2008), with a single herbarium record from Polk County (FLAS 2008). In 1981,

OWCF was documented in Collier County (Wunderlin and Hansen 2008) with additional

records in Highlands and Polk Counties (FTG 2008; Wunderlin and Hansen 2008).

From 1991-1996, OWCF was documented in Charlotte, Desoto, Hardee, Manatee, and

Lee Counties (FLAS 2008; FSU 2008; FTG 2008; Wunderlin and Hansen 2008),

indicating a range increase from east to west Florida. In 1997, a range increase was

documented in Brevard County along the southeast coast of Florida (Wunderlin and

Hansen 2008). In 1999, herbarium records indicate OWCF had moved further north

along the southwest coast of Florida with documentation from Hillsborough County

(Wunderlin and Hansen 2008). Additional records of OWCF were recorded in Brevard

County along the east coast of Florida in 2002 (FTG 2008; Wunderlin and Hansen









2008). Documentation of OWCF in Sumter County from 2008 (Wunderlin and Hansen

2008) and from Lake County in 2009 indicates that OWCF is slowly moving northward

in Florida.

In addition to invading further north into central Florida, the potential distribution of

OWCF, based on climate modeling in the New World, indicates that climate and

environmental conditions are suitable for expansion into the southern tip of Texas,

Caribbean islands, Central America, and large parts of South America (Goolsby 2004).

Therefore, OWCF exhibits the potential to invade areas in the new World tropics with

high diversity.

Rapid spread of OWCF across the landscape of southern Florida in less than 52

years since it was first documented is likely due in part to the production of numerous

wind-blown spores that can potentially travel many kilometers (Pemberton and Ferriter

1998). A single fertile pinnae can potentially produce 28,600 spores (Volin et al. 2004),

and a single spore is capable of intragametophytic selfing under laboratory conditions.

Thus, one spore can theoretically establish a population at a distant location far from the

parent plant (Lott et al. 2003). Sampling for airborne OWCF spores at one location in

Martin County, Florida resulted in an average of 724 spores/m3/hour being collected

(Pemberton and Ferriter 1998). The rapid invasion of OWCF in central and southern

Florida can be attributed to multiple factors such as copious spore production, long

distance dispersal of spores, mixed mating system, rapid growth, vertical and horizontal

growth pattern, and a lack of natural enemies.

Intragametophytic selfing appears to be rare in most species of homosporous

ferns (Soltis and Soltis 1992), but may have contributed to the rapid spread of OWCF in









Florida (Lott et al. 2003). Further evidence that selfing may facilitate the spread of

some ferns into new locations is provided by Flinn (2006). Tryon and Vitale (1977) were

the first to observe an antheridiogen system in natural populations of Asplenium

pimpinellifolium Fee. and Lygodium heterodoxum Kze. Antheridiogens have also been

hypothesized to inhibit interspecific competition among other fern species (Schneller et

al. 1990). Furthermore, OWCF gametophytes reach sexual maturity in five weeks and

produce sporophytes from five to 12 weeks, which is highly advantageous in the

seasonal wet and dry cycle of southern Florida (Lott et al. 2003).

OWCF has invaded a variety of mesic and hydric natural and disturbed habitats in

central and southern Florida (Pemberton and Ferriter 1998; Hutchinson and Langeland

2006). Natural habitats invaded by OWCF include cypress swamps, bayhead swamps,

hardwood swamps, pine flatwoods, Everglades tree islands, coastal prairies, and to a

lesser degree marsh, lake edges, and mangrove edges. Environmental factors

correlated with the presence of OWCF are the length and depth of inundation and soil

water content (Volin et al. 2004). The fern is also commonly found in any type of

disturbed habitat with mesic or hydric soils. Three native plants most commonly

associated with OWCF are swamp fern (Blechnum serrulatum), false nettle (Boehmeria

cylindrica), and royal fern (Osmunda regalis) (Volin et al. 2004). These three species

are obligate or facultative indicators of wetland habitat (Tobe et al. 1998), indicating that

OWCF could potentially invade any wetland habitat in southern or central Florida.

OWCF has invaded some of the least disturbed and isolated habitat in Florida such as

Everglades tree islands, coastal prairies, cypress swamps, bay swamps, and pine

flatwoods.









On Everglades tree islands, OWCF has reduced the abundance of native plants

and created more homogeneous tree islands dominated by OWCF (Brandt and Black

2001). Lower native plant cover, richness and diversity were documented in habitats

heavily infested with OWCF compared to similar habitats in which no OWCF was

present (Clark 2002). In addition, OWCF is a severe threat to many threatened and

endangered plants (Woodmansee and Sadie 2005; Hutchinson et al. 2006).

Two of the most common land management activities conducted by natural

resource agencies in Florida are prescribed fire and invasive plant treatment with

herbicides. Management of OWCF has primarily relied on ground or aerial applications

of glyphosate {N-(phosphonomethyl) glycine}, metsulfuron {Methyl 2-[[[[(4-methoxy-6-

methyl-1,3,5-triazin-2-yl)amino]-carbonyl]amino]sulfonyl]benzoate}, and triclopyr {3,5,6-

trichloro-2-pyridinyloxyacetic acid} (Hutchinson and Langeland 2006). However,

successful management of OWCF requires long-term monitoring and multiple follow-up

treatments (Hutchinson et al. 2006). Managers of public lands in Florida have reported

that lack of personnel, limited funding, other duties, site access, and OWCF on adjacent

lands as limiting factors in managing the fern and most are hoping for identification of

successful biocontrol agents (Hutchinson and Langeland 2006). The majority of large

scale herbicide application for control of OWCF on public lands is conducted by private

contractors through funding from the Florida Fish and Wildlife Conservation

Commission's Invasive Plant Management Section.

Manual (i.e., hand-pulling, cutting) and mechanical control (i.e., mowing) can

reduce above ground biomass, but OWCF re-sprouts from rhizomes in 6-7 days.

Manual removal is expensive, labor intensive, and time consuming (Roberts 1998), in









addition to potentially moving the spores to other non-infested sites. Due to the hydric

soils typical of the habitat in which OWCF occurs mechanical treatment is unfeasible in

natural areas due to significant soil disturbance and accessibility.

Prescribed fire does not control OWCF (Roberts 1998), which recovers to its initial

cover in 12-24 months (Roberts 1996). There is cursory evidence that prescribed fire

can result in OWCF invading a site where it was not initially present prior to the fire

(Hutchinson et al. 2006). Lipids, present as food reserves in many fern spores, make

fern spores more ignitable (Moran 2004), which could explain the intense heat

generated as the fire spreads upwards on the rachis mats where most of the fertile

leaflets are present. The pyrogenic nature of this plant can easily spread fire,

prescribed or natural, into wetland habitats that seldom burn. As a result, OWCF has

altered the prescribed burn programs in some natural areas (Roberts et al. 2006).

Following a fire, interspecific competition is decreased and OWCF re-sprouts rapidly

and further infests the habitats (Goolsby et al. 2003). However, Stocker et al. (2008)

found that in small plots, prescribed fire followed by biannual herbicide treatments

resulted in < 1 % cover of OWCF after three years.

Flooding of OWCF under greenhouse conditions resulted in reduced growth rates,

but with a concomitant increase in the number of OWCF fertile leaflets (Gandiaga et al.

2009). Herbicide treatment with glyphosate followed by flooding was successful in

controlling OWCF in hydrologically restored wetlands (Toth 2009).

Multiple biocontrol agents are being evaluated for control and three species

(Austromusotima camptozonale, Floracarus perrepare, and Neomusotima

conspurcatalis) have been released in southeastern Florida (Goolsby et al. 2003;









Pemberton 2006). One species, Neomusotima conspurcatalis, has shown promise by

defoliating 1.4 ha of OWCF among three release sites (Boughton and Pemberton 2009).

Herbicides are currently the only option available to manage and halt the spread of

OWCF infestations in Florida. Herbicide control of OWCF has shown some success but

OWCF has never been eliminated from a heavily infested site and recovers after a

single herbicide application (Thomas and Brandt 2003; Langeland and Link 2006;

Stocker et al. 2008). It is unclear if the three most commonly used herbicides

(glyphosate, metsulfuron, and triclopyr) for OWCF control are translocating into the

rhizomes of OWCF, which in theory, would prevent re-sprouts. No information exists on

the effects or tolerance of OWCF spores and gametophytes to herbicides. There are no

known studies that have examined the efficacy of repeated applications of glyphosate

and metsulfuron for control of OWCF and their effects on native plants on Everglades

tree islands.

The objectives of this research were: 1) examine herbicide translocation in OWCF,

2) assess the impact of herbicides on the survival of OWCF spores and gametophytes,

and 3) evaluate the efficacy of herbicides on OWCF for multiple ground and aerial

herbicide applications. Additional research was conducted to determine the impacts of

herbicides on native plant richness and diversity for objective 3.















30
U)



E 20


15 -

L.
10 -
10






1950's 1960's 1970's 1980's 1990's 2000's
Decade

Figure 1-1. Number of OWCF herbarium records per decade from the University of Florida (FLAS, 2008), Florida Sate
University (FSU, 2008), the University of South Florida (Wunderlin and Hansen 2008) and Fairchild Tropical
Gardens (FTG, 2008). FTG herbariums records include all herbarium records from the University of Miami.









CHAPTER 2
ABSORPTION AND TRANSLOCATION OF GLYPHOSATE, METSULFURON, AND
TRICLOPYR IN OLD WORLD CLIMBING FERN (LYGODIUM MICROPHYLLUM)

Introduction

Ferns (Class Pteridopsida), both non-native and native, can cause problems in

natural and cultivated areas. Non-native ferns have become problematic in natural

areas in many parts of the world (Langeland et al. 2008; Wilson 2002) and some native

ferns are considered weedy in agriculture and forestry sites (De la Cretaz and Kelty

1999; LeDuc et al. 2000). Old World climbing fern [OWCF; Lygodium microphyllum

(Cav.) R. Br.] is a highly invasive plant that can alter the structure and composition of

mesic and hydric natural areas in central and southern Florida (Brandt and Black 2001;

Nauman and Austin 1978). This species is now considered one of the worst threats to

natural areas by land managers in central and southern Florida (Hutchinson and

Langeland 2006).

Chemical control of OWCF in natural areas can be achieved, but consecutive

treatments are required to reduce and maintain populations at low levels (Hutchinson

and Langeland 2006; Langeland and Link 2006). Presently, there are no known cases

where OWCF has been eradicated from a natural area, though suppression has been

achieved with annual treatment (Hutchinson et al. 2006). While herbicide treatments are

effective in eliminating mature plants, new growth will develop from OWCF spores

present in the soil or blown in from nearby plants. In field sites infested with OWCF,

coverage is often >75% over several ha and the fern can grow > 15 m into the canopy

which makes treatment difficult. In field studies using several herbicides and tank

mixes, it took 2 to 3 herbicide applications every six months to eliminate OWCF from the

study plots, but new growth from spores was still present (J. T. Hutchinson and K. A.









Langeland, unpublished data). In some cases, multiple re-treatments may not be

possible due to funding, limited personnel, and inaccessibility. Consequently, the fern

has proven to be extremely difficult to control once it has become established and

begins to reproduce.

Management of OWCF with herbicides is conducted year-round in southern

Florida by multiple local, state, federal and private land management agencies. Except

when occasional freezing temperatures occur, OWCF is an evergreen that grows

vigorously throughout the year in southern Florida, where the average annual

temperature is 230 C with < 2 days/year when temperatures drop below 0 C (Henry et

al. 1994). The most frequently used herbicides for OWCF management in Florida are

glyphosate, metsulfuron, and triclopyr (Hutchinson and Langeland 2006). Application

rates for glyphosate and metsulfuron reported by land managers vary from 7.0 to 17.5 g

ai / I and 0.1 to 0.2 g ai / I, respectively, for spot treatments to individual infestations.

Land managers use triclopyr less frequently to control OWCF in Florida, but reported

rates varying from 22.0 to 55.0 g ai / I. Glyphosate and triclopyr have been used to

manage OWCF for more than 15 years. Metsulfuron has only been used for four years

since a Special Local Need 24c Label was obtained for use on OWCF in wet areas of

Florida (EPA SLN No. FL-030010).

Despite the spread of OWCF across the landscape of central and southern

Florida, relatively little research has been conducted to determine the most effective

treatment method for control. Stocker et al. (1997) reported that treatments of

glyphosate and triclopyr resulted in 100% above-ground mortality of OWCF at rates as

low as 1.5% product, but neither herbicide completely eradicated the plant as re-growth









was observed for both treatments after one year. In the Arthur R. Marshall Loxahatchee

National Wildlife Refuge in Palm Beach County, effective control of OWCF has been

achieved for > 3 years post-treatment in research plots by cutting the vertically

ascending rachis at 1.5 m above ground level and foliar spraying the lower portion with

glyphosate at 0.12 kg ai 3.8 I -1 (Thomas and Brandt 2003). OWCF plots treated with

foliar sprays of metsulfuron had the same above-ground dry weight biomass as control

plots at 2.5 years post-treatment (Langeland and Link 2006). Glyphosate and

metsulfuron have been the most effective herbicides to date to control OWCF, but

results are inconsistent with regard to long-term management (Hutchinson et al. 2006).

Triclopyr is one of the most common herbicides used in natural areas for management

of invasive plants, but it is unknown if this herbicide can be used effectively for control of

OWCF.

The ability of OWCF to resprout from rhizomes following herbicide treatment is an

indication that herbicides have limited translocation to the rhizomes or throughout the

entire rhizome of an individual fern. Some ferns have high root to shoot ratios (LeDuc et

al. 2003), and roughly 50% of the carbon sequestered by OWCF is allocated to

belowground root mass (John Volin, personnel communication). Application techniques

that maximize herbicide translocation to the rhizomes of OWCF should result in greater

or complete control of the fern.

Studies to determine herbicide uptake and translocation in OWCF should provide

a better understanding of management strategies for long-term control. The objective of

this research was to determine herbicide uptake and translocation patterns as a function









of herbicide placement in OWCF for glyphosate, metsulfuron, and triclopyr using 14C-

labeled compounds.

Materials and Methods

Plant Material

Experiments were replicated twice using OWCF plants grown from two different

sources. For the first experiment, OWCF was grown from rhizomes in Okeelanta muck

soil collected form Allapattah Flats, Martin County, Florida. Rhizome sections of 103

cm2 were cut with a machete from mature OWCF, and placed in 0.5-L plastic pots (12.7

cm in depth). Rhizomes were excavated with native soil intact. All fronds (live and dead)

from the rhizomes were cut at soil level, and the rhizomes allowed to re-sprout. Plants

were grown for 90 days to a height of 1 m prior to treatment.

The experiment was repeated with OWCF grown from spores. Soil excavated

below mature OWCF plants was placed in shallow 2.5-L tubs and maintained under

saturated conditions. Soil was collected from the same site as rhizomes in Martin

County, Florida. The spores were allowed to germinate and form gametophytes;

fertilization to sporophytes occurred thereafter. Approximately 6 months after

germination, the sporophytes, 7.6 cm in height, were transferred into 0.5-L pots in

Okeelanta muck soil and grown to 1 m in height prior to treatment.

Plants were grown in a greenhouse at the University of Florida campus in

Gainesville and exposed to a natural photoperiod (10.5 to 14.0 hours sunlight) under a

50% shade cloth with a temperature range of 21 to 370 C. The first experiment was

conducted from March to June 2005, and the second experiment was conducted from

January to October of 2005. No additional nutrient supplements were added to the









ferns over the duration of the study. Pots were placed in 7.5-L rectangular tubs with

water depth maintained at 2 to 3 cm below soil level and watered 3 to 4 times per week.

Absorption and Translocation

OWCF plants were treated with glyphosate1, metsulfuron2, and triclopyr3

herbicides at rates of 1.67 kg ae ha -1, 0.10 kg ai ha -1, and 1.20 kg ae ha -1,

respectively. Sources of material are listed in Table 2-1. Five different application

scenarios were used for each herbicide as follows: 1) plants were cut 6 cm from the

base of the plant and the remaining foliage was sprayed (denoted as cut and spray), 2)

foliage was treated from 0 to 6 cm above the base of the plant (denoted as basal spray),

3) foliage was treated from 0 to 25 cm above the base of the plant (denoted as 25%

foliar), 4) foliage was treated 0 to 50 cm above the base of the plant (denoted as 50%

foliar), and 5) 100% of the plant was treated (denoted as 100% foliar). Herbicides were

applied with a C02 sprayer at 172 kPa with a single TeeJet 11003 flat fan nozzle4. All

treatments included non-ionic surfactant5, 0.5 v/v, at a spray volume of 370 L ha-1.

Uptake and translocation were determined using 14C-labeled glyphosate6 (specific

activity 1998.0 kBq mg-1), metsulfuron7 (specific activity 1845.2 kBq mg-1), and triclopyr8

(specific activity 895.4 kBq mg-1). Immediately after spraying, application of 5 [L

droplets 14C-radiolabeled material was made to the adaxial surface of five separate

pinnae or rachis on each plant. Spotting of radiolabeled material was made at 6 cm (cut

and spray, and basal), 25 cm (25% foliar), 50 cm (50% foliar) and 100 cm (100% foliar).

This resulted in a total of 12.12 kBq mg-1 (0.33 [Ci) 14C glyphosate, 11.84 kBq mg-1

(0.32 [Ci) 14C metsulfuron, and 12.95 kBq mg-1 (0.35 [Ci) 14C triclopyr per plant.









At nine days after treatment (DAT), plants (including rhizomes) were removed from

pots, radiolabeled leaflets excised and the soil washed from roots using detergent and

water. Treated areas on leaflets or rachis were excised and washed in five sequential 1-

ml aliquots of deionized water applied to each radiolabeled spot to determine percent

uptake. A total of 25 ml of rinsate (combined total for all five spots) was obtained for

each treatment. Unabsorbed 14C from the leaf wash solution was quantified with liquid

scintillation spectrometry (LSS). 9

Whole plants were placed on blotter paper in plant presses and oven-dried at 700

C for 72 hours. Dried plants were covered with X-ray film10 and stored 40 days at room

temperature to obtain autoradiographs. Following autoradiography, the plants were

sectioned into rhizomes and stolons, lower 50% rachis and leaves, and upper 50%

rachis and leaves. Radioactivity was determined by LSS. Translocation was determined

by measuring radioactivity in individual plant parts. Plant tissues were finely ground

through a 0.5 mm screen using a Wiley Mill" and 0.15 to 0.20 g of ground tissue was

oxidized to recover 14C using a Harvey OX-500 biological oxidizer12 with a proprietary

14C-trapping liquid scintillation cocktail13. Absorbed radioactivity is defined as the total
14C- kBq obtained from all plant parts and presented as the percent of applied

radioactivity for each herbicide. Percent recovery is the absorbed radioactivity plus the

amount recovered from leaf-rinses. Translocated radioactivity is reported as the

percent of absorbed radioactivity in plant parts (top, bottom, rhizomes) and presented

as a percent of absorbed radioactivity.

Experimental Design and Data Analysis

Treatments were arranged in a completely randomized design. Each treatment

was replicated five times and the experiment was repeated using plants grown from









rhizomes (n = 25) and spores (n = 25). There was no significant (P > 0.05) treatment by

experiment interaction between rhizome and spore grown plants, therefore data

represent the average of both experiments. Total recovery, leaf wash data, absorbed

and translocated radioactivity were combined among treatment scenarios for each

herbicide and subjected to analysis of variance (ANOVA) using Proc GLM (SAS14)

procedure, and means were separated using Fisher's protected LSD test at the 5%

level of probability.

Radioactivity in plant parts was analyzed by herbicide and treatment scenario

separately to determine patterns of translocation for each herbicide. Data were

subjected to ANOVA and means were separated using Fisher's protected LSD test at

the 5% level of probability. Translocation was analyzed for each treatment method for

percentage radioactivity in top 50%, bottom 50% and rhizomes using an ANOVA with

means separated using Fisher's protected LSD test at the 5% level of probability. For

translocation data with the cut and spray method, a paired t-test using SAS was used to

compare bottom 50% and rhizomes at the 5% level of probability.

Results and Discussion

Visual Effects of Herbicides on OWCF

OWCF treated with triclopyr exhibited chlorosis, epinasty, and necrosis nine DAT

(days after treatment) on 30-40% of OWCF leaflets for all treatments scenarios. There

were no observable effects on OWCF treated with glyphosate or metsulfuron at nine

DAT. This may indicate that triclopyr is absorbed or translocated at a faster rate relative

to glyphosate or metsulfuron, or the rate at which symptoms of the different

mechanisms of actions are expressed are different. In field trials with triclopyr at similar

rates, necrosis was observed in < 4 hours in OWCF (J. T. Hutchinson, personal









observation). Rapid burning and tissue necrosis by triclopyr in OWCF may increase

absorption, but could inhibit translocation by limiting entry of the herbicide into vascular

tissue. It was discovered that rapid wilting and yellowing in purple loosestrife treated

with triclopyr stimulated new growth (Gardner and Grue 1996). Hofstra et al. (2006)

suggested that lower concentrations of triclopyr may result in more effective

translocation in parrotsfeather compared to higher rates and prevent basal stem re-

sprouts.

Autoradiography

The general pattern observed from autoradiographs indicated most of the 14C

material remained in the treated leaflets or rachis. For all three herbicides, some

basipetal movement was observed for cut and spray and basal spray (Figure 2-1).

However, basipetal movement with 100% foliar spray for any herbicide was minimal

with almost no movement of 14C material into the rhizomes. With 25% and 50% foliar

spray, herbicide movement was primarily acropetal with some basipetal movement for

all three herbicides.

Basipetal movement into the rhizomes was observed for cut and spray and basal

spray treatments of triclopyr and glyphosate. However, there was no evidence of

horizontal movement for triclopyr or glyphosate along rhizomes in autoradiographs.

Basal treatment with triclopyr also indicated some acropetal translocation (Figure 1c).

Autoradiographs of OWCF treated with metsulfuron indicated minimal translocation of

the herbicide into the rhizomes. These results corroborate field observations of all three

herbicides in which OWCF is highly susceptible to herbicide treatment, but re-sprouting

occurs within several weeks to months post-treatment. Re-sprouts from OWCF treated









with herbicide is likely due to limited translocation into the rhizomes and no horizontal

translocation within the rhizomes.

Absorption

There was no significant difference in total recovery between treatments or plant

part other than the treated leaf among herbicides (Table 2-2). Total recovery was 67%

for metsulfuron, 65% for glyphosate, and 69% for triclopyr. However, differences were

observed among herbicides in leaf wash with only the lowest percentage, 9%, washed

from leaves treated with triclopyr followed by 34% for glyphosate and 47% for

metsulfuron. Accordingly, differences in absorption were observed among herbicides

(Table 2-2). The highest level of absorption was observed for triclopyr (60%) followed by

glyphosate (31%) and metsulfuron (20%). Total translocation was different for triclopyr

(14%) compared to glyphosate (6%) and metsulfuron (5%).

Overall, more of the applied triclopyr (87%) was absorbed compared to only 48%

of glyphosate and 30% of metsulfuron. However, 23% of the applied metsulfuron was

translocated out of the treated leaflets compared to 23% of triclopyr and 20% of

glyphosate. This indicates that translocation out of the treated leaflet is a better indicator

of the effectiveness of a herbicide in reaching its target site and blocking specific

physiological processes inside cells. There was no difference in treatment scenarios for

absorption of metsulfuron and triclopyr, but there were differences between the cut and

spray (38%) and 100% foliar spray (27%) with glyphosate (Tables 2-3 to 2-5).

Reddy (2000) reported 22% absorption of glyphosate in redvine [Brunnichia ovata

(Walt.) Shinners] and Chachalis and Reddy (2004) reported 20% absorption of

glyphosate in trumpetcreeper [Campsis radicans (L.) Seem. ex Bureau]. These values

are similar to the 31% absorption of glyphosate observed in OWCF. While the data









indicates that triclopyr and glyphosate are absorbed in greater amounts than

metsulfuron by OWCF, increased rates of herbicide absorption do not necessarily mean

better control, but the relationship is species-specific (Norsworthy et al. 2001; Chachalis

and Reddy 2004). Norsworthy et al. (2001) found that some species have more

tolerance for glyphosate even at higher absorption rates.

Translocation

Radioactivity was detected in all portions of treated plants for all herbicide

treatments indicating that all herbicides were translocated to some extent (Table 2-3 to

2-5). The majority of radioactivity remained in treated leaves for all herbicides with only

small percentages of the absorbed radioactivity being detected in other plant parts (top

50%, bottom 50%, and rhizomes). Averaged over all treatment methods for each

herbicide, more radioactivity remained in the triclopyr (47%) treated leaves compared to

metsulfuron (15%) and glyphosate (25%) (Table 2-2). Necrosis at 9 DAT and the large

amount of radioactivity remaining in treated leaves indicates that translocation may

have been limited with triclopyr. In a review of phytotoxic action on herbicide

translocation, it was suggested that herbicides may inhibit phloem translocation and

herbicide distribution (Geiger and Bestman 1990). Our results indicate that this may

have been the case in this study where inhibition of normal carbon metabolism and

phloem transport occurred in OWCF.

Radioactivity was detected in the upper plant portions for all three herbicides (top

50%) when lower portions of plants (basal, 25% foliar, 50% foliar) were treated

indicating that all three herbicides moved acropetally (Table 2-3 to 2-4). Radioactivity

was also detected in rhizomes in all treatment scenarios for all herbicides indicating that

all three herbicides moved basipitally. While some significant differences in radioactivity









in plant parts, other than treated leaves, were observed, these differences were small

and radioactivity, for the most part, was evenly distributed among the top 50%, bottom

50%, and rhizomes. The greatest amount of radioactivity movement into the rhizomes

was from the cut and spray and basal applications for both triclopyr and glyphosate.

This might be explained because the herbicide is applied to older leaves at the base. In

many plant species, photoassimilates are translocated from older leaves into sinks

(rhizomes, tubers, bulbs, etc.) (Williams 1964b; Khan 1981; Wolf 1993). Veerasekaran

et al. (1977) found that asulam applied to mature fronds of bracken fern resulted in

greater translocation into the rhizomes. In contrast, translocation of 14C glyphosate in

pitted morningglory (Ipomoea lacunosa L.) indicated that herbicide movement did not

differ whether the herbicide was applied to the top, middle, or bottom portion of the plant

(Koger and Reddy 2005). In addition, when using the cut and spray method to treat

OWCF, herbicides can diffuse basipitally in the rhizomes through cuts in the rachis.

Translocation of a given herbicide varies among species. Basipetal movement

was reported to be 49% for 14C-glyphosate in trumpetcreeper (Chachalis and Reddy

2004), which is substantially greater than the 4% that we observed with OWCF. Similar

to observations for OWCF, limited basipetal movement was reported for stoloniferous

ground ivy (Glechoma hederacea L.) treated with 2,4-D (Kohler et al. 2004) and ivyleaf

morningglory [Ipomoea hederacea (L.) Jacq.] treated with glyphosate, imazethapyr, or

glufosinate (Hoss et al. 2003).

These results indicate that translocation into the rhizomes is greater with the cut

and spray and basal treatment methods using triclopyr and glyphosate where herbicide

is concentrated on the older leaves at the base of the plant. The cut/spray method is









the most commonly used method for treating OWCF growing high up into trees

(Hutchinson and Langeland 2006).

OWCF has extensive rhizomes, especially in long established populations. It can

be assumed that translocation of herbicide through the rhizomes to prevent re-sprouting

is needed for long-term control. Veerasekaran et al. (1977) reported that asulam (4.4 to

8.8 kg a.i. ha -1) eliminated 90% of the shallow active frond buds in bracken fern

(Pteridium aquilinum (L.) Kuhn), but only eliminated 52% of the dormant buds on deeper

rhizome branches. Translocation of herbicide from the treated leaves may not be an

indication of the efficacy of the herbicide (Kalnay and Glenn 2000). This is especially

true with low volume, high potency acetolactate synthase inhibiting herbicides such as

metsulfuron in which small amounts of herbicide are toxic to plants (Fairbrother and

Kapustka 2001). Based on autoradiographs of OWCF, once herbicide translocates to

the rhizomes, there is little or no horizontal translocation along the rhizome. Koger and

Reddy (2005) suggested that control of pitted morningglory is more likely affected by

rate than spray coverage. Reddy (2000) suggested that higher rates of glyphosate will

be required for effective control of redvine due to its deep rooted rootstocks where new

sprouts emerge following treatment. Likewise, the use of higher herbicide rates for cut

and spray and basal treatments may be required for greater translocation into the

rhizomes of OWCF.

Conversely, it is also possible that the herbicides tested may interfere with phloem

transport at the junction of the rachis and rhizome. This could limit horizontal movement

of herbicides through the rhizome and increased rates may not result in greater

translocation in to rhizomes. Pakeman et al. (2000) reported variable control of bracken









fern with aerial application of asulam in which ca. 75% of the sites sprayed exhibited

bracken recovery. This corresponds with reports from land managers in Florida who

have reported inconsistent results when treating OWCF with glyphosate and

metsulfuron (Hutchinson et al. 2006). Research has shown that glyphosate inhibits

phloem transport in plants 24 to 48 h after treatment (Geiger and Bestman 1990;

Kirkwood et al. 2000; Walker and Oliver 2008). These results, which show limited

translocation in OWCF may indicate that triclopyr and metsulfuron also interfere with

physiological processes that block phloem transport in OWCF.

There are no studies of herbicide translocation in OWCF. Other authors have

reported limited basipetal movement of herbicides in other vines (Unland et al. 1999;

Hoss et al. 2003; Kohler et al. 2004). Unlike in bracken fern, where radiolabeled asulam

accumulated in rhizomes (Veerasekaran et al. 1977), I found no observable

accumulation of radiolabeled material in the rhizomes with the three herbicides tested in

OWCF. Based on autoradiographs, 2,4-D movement into the rhizomes of bracken fern

was greatest when the herbicide was applied to immature fronds (McIntyre 1962).

Treatment of immature fronds of OWCF could only be accomplished where the fern has

first invaded or on new sprouts following herbicide treatment, cutting, or fire. Main et al.

(2006) found that triclopyr displayed excellent basipetal translocation in Chinese yam

(Dioscorea oppositifolia L.) resulting in the elimination of tubers. In pitted morning glory,

movement of 14C-glyphosate to the roots was highest for 100% foliar treatment (Koger

and Reddy 2005). Control of OWCF is dependent on controlling the rhizomes, and all

three herbicides used in this study exhibited some basipetal movement into the









rhizomes. However, autoradiographs revealed limited movement through the rhizomes

indicating that all rachises must be treated to prevent re-sprouts.

Limited field research on OWCF is available for the three herbicides tested in this

study. The greater effectiveness of the cut and spray treatment for basipetal movement

into the rhizomes I found are consistent with the results of Thomas and Brandt for field

treatment of OWCF (2003). Metsulfuron at rates of 0.04 and 0.08 kg ai ha -1 has been

shown to provide control of OWCF for 12 months using 100% foliar treatment

(Langeland and Link 2006). Based on these results, metsulfuron may be effective for

ground treatments using the cut and spray, basal, or 100% foliar treatments. Aerial

applications of metsulfuron at rates of 0.08 and 0.16 kg ai ha -1 over OWCF resulted in >

90% mortality at 1 year post treatment (J. T. Hutchinson and K. A. Langeland,

unpublished data). Stocker et al. (1997) reported that spot treatments with triclopyr

(0.13 kg ai 3.8 I-1) foliar sprayed and band-sprayed from the base of OWCF to 1.2 m

above ground resulted in 100% browning of OWCF, but re-growth was observed and

they suggested limited translocation. Using prescribed fire and/or triclopyr, it was found

that bimonthly and biannual treatments of OWCF were successful in reducing OWCF

cover to < 1 % but re-sprouts were still documented after 3 years (Stocker et al. 2008).

The results of this study suggest that cut and spray and basal treatments with triclopyr

should result in greater basipetal movement into the rhizomes.

These results revealed limited movement of the herbicides tested through the

rhizomes of OWCF, which explains why re-sprouts from treated OWCF are commonly

observed in the field < 12 mo post-treatment. In a greenhouse study on OWCF using

twice the labeled rates of metsulfuron (0.16 kg ai) and glyphosate (11.20 kg ai), no re-









sprouts were observed 52 weeks post-treatment, indicating that herbicide rates of

metsulfuron and glyphosate may need to be increased for initial treatments to effectively

control the fern (Hutchinson and Langeland 2008). The current treatment methods (cut

and spray and 100% foliar spray) used by applicators in Florida should continue to be

used to manage OWCF, but re-treatment will be required. Additional research is

needed to determine if basal treatments (i.e., band spraying) are effective under field

conditions since our results revealed some herbicide movement to the rhizomes using

this method. Glyphosate and metsulfuron are the most commonly used herbicides to

treat OWCF in Florida, but these results indicate that triclopyr exhibits greater

translocation to the rhizomes. Field research is needed to determine the efficacy of

triclopyr and herbicide combinations on OWCF for both ground and aerial treatments.

Research on herbicide rates above those used in this project are also needed to

determine their efficacy on controlling OWCF.









Table 2-1. Sources of material used in OWCF 14C translocation study.
1 Glyphosate herbicide, Monsanto Agricultural Company, 800 N. Lindbergh Blvd.,
St. Louis, MO 63167 (Roundup Ultramax, 50.2 a.i.)
2 Metsulfuron herbicide, DuPont Corporation, 1007 Market St., Wilmington, DE
19898 (Escort XP, 60% a.i.)
3 Triclopyr herbicide, SePRO Corporation, 11550 N. Meridian St., Suite 600, Camel,
IN 46032 (Renovate 3, 44.4% a.i.)
4 TeeJet MidTech Southeast, P.O. Box 832, Tifton, GA 31793
5 Entry non-ionic surfactant, Monsanto Corporation, 800 N. Lindbergh Blvd., St.
Louis, MO 63167
6 14C-glyphosate, Amersham Life Sciences Inc., 3350 N. Ridge Ave., Arlington
Heights, IL 6004
7 14C-metsulfuron, gift from DuPont Corporation, 1007 Market St., Wilmington, DE
19898
8 14C-triclopyr, gift from Dow AgroSciences LLC, 9330 Zionsville Rd., Indianapolis,
IN 46268
9 Packard Tricarb 1600CA Liquid Scintillation Analyzer, Packard Instrument Co.,
800 Research Parkway, Meriden, CT 06450
10 Kodak X-OMAT XAR-5 film, Sigma-Aldrich, 3050 Spruce St., St. Louis, MO 63103
11 Wiley Mill. Arthur W. Thomas Company, Philadelphia, PA (No model number)
12 R. J. Harvey Biological Oxidizer, Model OX-500, R. J. Harvey Instrument Co., 123
Patterson Street, Hillsdale, NJ 07642
13 R. J. Harvey 14-Carbon Cocktail (proprietary blend of biscumene and PPO
(diphenyloxazole) in mixed xylenes), R. J. Harvey Instrument Co., 123 Patterson
Street, Hillsdale, NJ 07642
14 SAS Institute, 2002-2003, SAS software, Version 9.1, SAS Institute, Cary, NC
57513









Table 2-2. Radioactivity (percent of applied) and standard error for total recovery, leaf wash, absorbed, treated leaf, and
translocation following application of 14C-labeled metsulfuron, glyphosate, and triclopyr to OWCF (all
treatments combined).
Total Leaf Treated
Herbicide Recovery Wash Absorbed Leaf Translocated


Metsulfuron 66.7 (4.9)1 46.9 (4.7)
Glyphosate 65.2 (4.7) 34.0 (5.5)
Triclopyr 69.0 (4.3) 8.7 (1.4)
LSD (0.05%) NS 6.2
1 Means of 10 replications followed by standard error.


% Recovered
19.8(0.6)
31.2 (1.8)
60.3 (3.4)


15.2(0.9)
25.0(1.3)
46.7 (2.0)
4.1


4.6 (0.8)
6.2 (0.4)
13.6(1.1)
2.9


Table 2-3. Percent 14C-metsulfuron absorption, translocation, and distribution in OWCF for five treatment methods.
Values represent the means of 10 replications.
% 14C-metsulfuron recovered
Movement out of Treated Leaves
Leaf Treated Total Top Bottom LSD 1
Treatment Wash Absorbed Leaves Translocated 50% 50% Rhizomes (0.05%)
Cut/Spray 32.2 19.1 15.8 3.3 N/A 0.9 2.4 P = 0.03
Basal 41.4 20.5 16.4 4.1 0.6 1.3 2.2 1.5
25% Foliar 49.4 21.8 16.1 4.7 0.7 2.5 1.5 1.5
50% Foliar 51.6 17.9 10.3 7.6 3.2 2.4 2.0 NS
100% Foliar 60.1 20.7 17.3 3.4 0.6 1.3 1.5 NS
LSD (0.05%) 14.4 NS 5.8 3.1 1.7 NS NS
1 Significance test comparing % DPM in top 50%, bottom 50% and rhizomes within rows. For cut/spray treatment,
significance test was a paired t-test (P < 0.05).









Table 2-4. Percent 14C-glyphosate absorption, translocation, and distribution in OWCF for five treatment methods. Values
represent the means of 10 replications.


Treatment
Cut/Spray
Basal
25% Foliar
50% Foliar
100% Foliar
LSD (0.05%)
1 Significance


Leaf
Wash
15.4
30.4
48.4
38.2
37.4
19.1
test comparing


% 'C-,qlyphosate recovered
Movement out of Treated Leaves


Treated Top Bottom L
Absorbed Leaves Translocated 50% 50% Rhizomes ((
37.5 31.3 6.2 N/A 1.8 4.4 P
28.8 22.5 6.3 1.4 1.8 3.1 I
32.2 25.0 7.2 1.7 3.3 2.2 I
30.6 24.5 6.1 2.8 1.1 2.2 I
26.7 21.7 5.0 0.4 1.3 3.3 1
10.1 8.0 NS 1.8 NS 1.8
% DPM in top 50%, bottom 50% and rhizomes within rows. For cut/spray treatment,


significance test was a paired t-test (P < 0.05).


SD1
0.05%)
= 0.06
IS
IS
IS
.4


Table 2-5. Percent 14C-triclopyr absorption, translocation, and distribution in Old World climbing fern for five treatment
methods. Values represent the means of 10 replications.
% 14C-triclopyr recovered
Movement out of Treated Leaves
Leaf Treated Top Bottom LSD
Treatment Wash Absorbed Leaves Translocated 50% 50% Rhizomes (0.05%)
Cut/Spray 5.8 51.1 38.4 12.7 N/A 4.6 8.1 P = 0.003
Basal 6.9 53.3 36.2 17.1 2.8 7.0 7.3 4.1
25% Foliar 11.1 64.6 52.0 12.6 4.1 5.9 1.9 NS
50% Foliar 12.8 64.3 49.4 14.9 5.5 6.3 1.5 NS
100% Foliar 6.8 68.3 57.7 10.6 4.2 9.7 0.7 NS
LSD (0.05%) 5.9 NS 11.1 NS NS NS 3.3
1 Significance test comparing % DPM in top 50%, bottom 50% and rhizomes within rows. For cut/spray treatment,
significance test was a paired t-test (P < 0.05).

















a)










b)













Figure 2-1. Dried plants (left) and autoradiographs (right) of OWCF treated basally with
a) metsulfuron, b) glyphosate, and c) triclopyr. Arrows on left indicate site of
treatment and arrows on right indicate spotted pinnae.









CHAPTER 3
TOLERANCE OF OLD WORLD CLIMBING FERN SPORES AND GAMETOPHYTES
TO HERBICIDE EXPOSURE

Introduction

Few studies have used spores and gametophytes of ferns in herbicide screening

trials. Spores are single-celled, without an embryo, and have limited food reserves

compare to seeds which are comprised of 1000's of cells, contain an embryo and

greater food reserves (Moran 2004). In addition, spores germinate into gametophytes

which require moisture for fertilization prior to growing into sporophytes while seeds

germinate directly into a new sporophyte generation.

No previous studies have examined the effects of herbicides on Old World

climbing fern (Lygodium microphyllum; OWCF) gametophytes and spores.

Gametophytes of Lygodium japonicum exposed to 2,4-D at concentrations > 10.0 mg / I

exhibited decreased growth rates and remained as filaments (Swami and Raghavan

1980). Gametophytes have been used to study the physiological basis of tolerance in

the paraquat-tolerant fern, Ceratopteris richardii Brongn. (Carroll et al. 1988). 2,4-D

was shown to affect gametophyte growth, cell size, antheridia number, and rhizoid

development in Ceratopteris thalictroides (L.) Brongn. (Hickok and Kiriluk 1984).

Gametophytes of Pteridium aquilinum [L. (Kuhn)] exposed to asulam at 100 g / I

suffered 100% mortality, and exhibited lower levels of mortality with decreasing

concentrations (Keary et al. 2000). Spores of Onoclea sensibilis treated with colchicine,

a microtubule inhibitor, resulted in single cell division without further growth (Vogelmann

et al., 1981).

High propagule production combined with a high germination rate may accelerate

herbicide resistance in ferns because susceptible plants are selected against, leaving









resistant genotypes in the population (Prather et al. 2000). Volin et al. (2004) estimated

that single fertile pinnae of OWCF could produce 28,600 spores. Germination rates of

OWCF spores on natural soils have been reported to range 84 97 % (Call et al. 2007).

Spore dispersal in OWCF has been recorded to be 724 spores / m3 / hour which explain

how the fern has rapidly invaded natural areas of south and central Florida (Pemberton

and Ferriter 1998).

Spores and gametophytes are easily cultured in the laboratory and are genetically

identical to sporophytes (Voeller 1971; Keary et al. 2000; Fernandez and Revilla 2003).

The use of gametophytes and spores could provide insight into the tolerance level of

OWCF to various herbicides because multiple populations and different ages can be

tested under controlled conditions. Detection of tolerance levels in which OWCF spores

and gametophytes survive treatment could be used to set a minimum rate for field

herbicide application.

In 2003, metsulfuron received a special local need 24 (c) registration for

application to treat OWCF (Dupont Corporation 2003). Since 2003, metsulfuron has

been used to treat OWCF with aerial and ground-based applications (Hutchinson et al.

2006). The acetolactate synthase (ALS) inhibiting herbicide class, sulfonylurea, which

includes metsulfuron, inhibits branched chain amino acids and was introduced

commercially in 1982. Since 1982, more species have become resistant to this class of

herbicides than any other class of herbicides, with some weedy plants developing

resistance in < 5 years (Tranel and Wright 2002). Resistance to herbicides with an ALS

mode of action occurs through the selection of a biotype with an amino acid substitution

of the ALS enzyme rendering the herbicide ineffective. The ALS gene is nuclear









inherited being transmitted in pollen, spores, and seeds (Tranel and Wright 2002), so

OWCF spores and gametophytes could be used for evaluation of herbicide tolerance.

With the heavy reliance on metsulfuron, coupled with the known problems with

resistance and the tremendous reproductive capacity of OWCF, the objectives of this

research were to assess the susceptibility and/or tolerance level of OWCF

gametophytes to metsulfuron, and OWCF spores to metsulfuron and other herbicides

with different modes of actions.

Materials and Methods

Source of Spores

Mature fertile fronds of OWCF were collected from 12 locations in natural areas of

central and southern Florida from populations that had never been treated with

herbicide (Table 3-1; Figure 3-1). Fronds were collected from > 6 OWCF plants at each

site and combined for one composite sample per site. Fronds were placed in open

plastic bags and allowed to air dry at ambient temperature. Age of the spores ranged

from 1 to 380 weeks for experiment 1, 20 to 29 weeks for experiment 2, and 4 to 432

weeks for experiment 3 (Table 3-2). Spore age, in weeks, is the time elapsed between

collection of the fertile leaflets from live plants and the initiation of each experiment and

varies among experiments and the collection time of spores. Ranker and Houston

(2002) suggested using spores of different age classes would be more reflective of a

natural age structure. The use of OWCF spores from different populations and various

ages should provide a more comparative view of the response of OWCF spores to

metsulfuron under natural conditions.









Spore Preparation

Spores from each location were isolated from leaf material by passing through a

63 pm sieve (Hubbard Scientific Co., Northbrook, Ill.; Part # 3070-6). After sieving, 0.02

(SE=0.001) g portions of spores were weighed into pre-weighed 1.5 ml Thumbs-Up

Microtubes (Diversified Biotech, Boston, MA) on a Mettler analytical balance (Mettler-

Toledo, Inc., Columbus, OH; Part # AE 100). One ml of 1% bleach solution in distilled

water was pipetted into each microtube containing the spores. The microtubes and

spores were vortexed for 2 minutes and then centrifuged for 2 minutes at 1000 rpm. The

liquid was decanted and the process repeated once. After spores were surfaced

sterilized, the liquid was decanted and 1 ml of distilled water was added to the

microtubes.

Spore Counts

Spores were counted by vortexing the microtube containing the spores in 1 ml of

distilled water for 3 minutes to fully suspend the spores in the solution. Immediately

after the spore suspension was vortexed, a 0.1 pl solution was pipetted into each

chamber of a hemacytomer (Hausser Scientific, Horsham, PA.; Bright Line Counting

Chamber, Part # 1490). Spores were counted in 10 1 mm2 grids of the hemacytomer

under a microscope at 4x power and averaged to estimate spore density. The mean

number of spores was multiplied by 10,000 to estimate the number of spores per 0.02

grams, which equated to approximately 175,000 (SE = 10,200) spores.

Germination Medium

A stock solution of modified Parker / Thompson's basal nutrient medium

containing macro- and micronutrients, and a modified iron solution (Klekowski, 1969;

Hickok et al., 1995) was prepared in distilled water based on the methods of Lott et al.









(2003). The nutrient solution was titrated to a pH of 6.0 using concentrated hydrochloric

acid and placed on a heated stir plate. For experiment 3, metsulfuron was added to the

solution for a concentration of 0.4 g a.i. / I (equivalent to a field application rate of rate of

0.08 kg a.i. / 187 I). Ten grams (1% w/v) of agar were added to the nutrient solution and

heated at 125 oC on a stir plate until the agar and nutrients were clear and in solution.

The agar / nutrient solution was then autoclaved for 30 minutes.

Within 15 minutes of autoclaving, 100 Petri dishes containing the agar / nutrient

solution were prepared under sterile conditions using a Laminar flow hood by pipetting

10 ml of agar / nutrient solution into each dish. The dishes were allowed to solidify over

3 hours and subjected to 1 hour of ultraviolet light for further sterilization. Petri dishes

were sealed in sterilized plastic bags and stored at 4 oC until inoculation with spores.

Spore Inoculation

Under a laminar flow hood, 1 ml microtubes containing distilled water and 0.02

grams of spores were suspended in solution for two minutes using a vortex mixer.

Immediately after mixing, a 0.2 ml aliquot suspension of spores was pipetted into a Petri

dish. Each dish was slowly swirled to distribute the spores on the agar medium. The

remainder of the microtube was vortexed again and pipetted onto 4 additional dishes.

Approximately 35,000 spores were pipetted into each Petri dish with the number of

replications varying among experiments as described below. Each Petri dish was

sealed in parafilm. The Petri dishes were placed under cool-white fluorescent light (80

pmol/m2/s1) for a photoperiod of 16/8 hours with a mean temperature of 25.0 (SE =

0.02) oC. Temperature was recorded at 30 minute intervals using a Hobo-Temp data-

logger (Onset Computer Corporation, Bourne, MA).









Germination Rates

Germinated spores were counted 15, 30 and/or 45 days after inoculation under a

microscope (Fisher Scientific Micromaster II Video Microscope) at 4x magnification

depending on the experiment. Germination rates were calculated by dividing the

number of germinated spores by the total number of spores within 20 5 mm2 reticule

grids (Fisher Scientific, Cat. # 12-561-RG1) and converting to percent. Spores were

counted as germinated if the primary prothallia cell had emerged from the spore. The

sexuality of gametophytes was not quantified, but the majority of all gametophytes were

male filamentouss, plate-like, and amerisitc) while <5% were females containing a

meristem notch.

Sporophyte Development

Sixty days post-treatment and a total of 90 days, all sporophytes that developed

from treated gametophytes were transferred into 0.15 liter pots containing commercially

purchased humus. Sporophytes were scraped from the agar with a spatula and placed

in a small depression (2 cm wide x 0.05 cm deep) in the humus. A small amount of

loose rock-wool was incorporated into the humus to raise sporophytes 0.05 cm above

humus level. At 60 days following the potting of sporophytes in humus, all sporophytes

were counted and then cut at soil level. Percent survival of sporophytes for each

treatment was determined by dividing the number of live sporophytes per treatment by

the number of gametophytes that survived each treatment. All cut OWCF plant material

was placed in paper bags and dried at 90 C for seven days to determine dry weight

biomass.









Experiment 1 Tolerance of Gametophytes to Metsulfuron

Thirty days after being placed on agar, gametophytes were exposed to

metsulfuron at concentrations of 27, 54, 108, 216, 432, and 864 mg a.i. / I (equivalent to

field application concentrations of 0.005, 0.10, 0.20, 0.40, 0.80, and 0.16 kg a.i. / 187 1)

or 0.5% v/v surfactant (DyneAmic, Helena Chemical Co.). Surfactant was not added to

metsulfuron treatments. Untreated checks contained no metsulfuron or surfactant.

Metsulfuron was added to each Petri dish under sterile conditions by pipetting 1.0 ml

aliquots of herbicide solutions into the Petri dishes. This volume ensured that all

gametophytes were completely exposed to the herbicides. Germination rates were

determined by dividing the number of gametophytes by the number of spores and

expressing as percent. Each treatment was replicated eight times and the experiment

was repeated four times for 32 samples per treatment.

Germination rates were determined separately for age group (<10, 11-20, 21-40,

70-100, 101-200, and >300 weeks) and location 30 days prior to treatment.

Germination rate was determined for each treatment (including surfactant and untreated

checks) prior to treatment (30 days) and 30 days post-treatment. The site with spores

>300 weeks old (DuPuis Reserve) was eliminated from further analysis due to low mean

germination rate (1.5%). Sporophyte survival was determined by counting the number

of individual sporophytes that developed in potted humus at 120 days post-treatment.

Individual sporophytes were cut at soil level 120 days post-treatment to determine dry-

weight.

All data were subjected to analysis of variance (ANOVA) using Proc GLM

procedure, and means were separated using Fisher's protected LSD test at the 5%

level of probability (SAS, 2002). Data from all four experiments were combined and









separated by herbicide because there was no treatment by experiment interaction (P >

0.05).

Non-linear regression (Proc NLIN; SAS 2002) was used to calculate a dose-

response curve for each treatment rate based on gametophyte survival 30 days post-

treatment and sporophyte survival and dry weight 120 days post-treatment. The

exponential decay equation [y = ae -(bx)] was used for non-linear regression analysis.

Inhibition concentrations (150 and 195) were calculated for gametophyte survival at 30

days post-treatment and sporophyte survival at 120 days post-treatment. Growth

reduction concentrations (GRso and GR95) were calculated for sporophyte dry weight

biomass 120 days post-treatment.

Experiment 2 Tolerance of Spores to Selected Herbicides

OWCF spores were placed for 1, 3, 6, 12 and 24 hours in 1 ml aliquots of solutions

containing one of five concentrations of six herbicides (Table 3-3) or 0.5% v/v surfactant

(DyneAmic, Helena Chemical Co.). Untreated checks contained no herbicide or

surfactant. Each treatment was replicated two times and the experiment was repeated

three times for six samples per treatment. Following exposure, the herbicide solution

was decanted and the spores washed with distilled water three times. Treated spores

were placed on an agar/nutrient media and monitored for germination after 30 days.

Germination rates were determined after 30 days and means, standard error, and 95%

confidence intervals were calculated. Germination rates were determined by dividing

the number of gametophytes by the number of spores and expressed as percent.

Data for exposure time from each experiment was combined because there were no

differences among exposure times or treatment by time interaction for each herbicide (P

> 0.05).









Non-linear regression (Proc NLIN; SAS 2002) was used to calculate a dose-

response curve for each herbicide based on percent germination at 30 days post-

treatment (SAS, 2002). The exponential decay equation [y = ae -(bx)] was used for non-

linear regression analysis. Inhibition concentrations for 150 and 195 were calculated for

percent spore germination at 30 days post-treatment for each herbicide.

Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium

Suspensions of OWCF spores were pipetted onto an agar/nutrient media

containing 0.4 g a.i. / I metsulfuron (equivalent to a field application rate of 0.08 kg a.i. /

187 I / ha). In the first experiment, spores from 12 locations that varied in age were

combined and tested (Table 3-2). Each treatment was replicated eight times with an

equal number of untreated checks and the experiment was conducted twice for a total

of 16 samples. In the second experiment, spores from two locations (IRCC and JDSP)

were placed on agar/nutrient media (untreated check) or agar/nutrient media in which

metsulfuron was incorporated to test individual populations of the same age. The ages

of spores were 46 and 36 weeks for the IRCC and JDSP sites, respectively. Each

treatment was replicated eight times with an equal number of untreated checks and the

experiment was conducted twice for a total of 16 samples per location.

Percent spore germination was counted after 15 and 45 days for the first

experiment, and at 45 days for the second and third experiments. Differences between

untreated checks and agar/metsulfuron samples were analyzed with a paired t-test at P

< 0.05 (SAS 2002).









Results

Experiment 1 Tolerance of Gametophytes to Metsulfuron

Germination of OWCF spores 300 weeks of age was lower than all other age

groups (P = 0.0005) but there was no difference in germination rates among other age

groups (Figure 3-2). Germination rates were not different among the more southerly

locations with a trend of lower germination in more northerly collection sites (Figure 3-

3).

There was no difference in germination rates of OWCF spores 30 days pre-

treatment (Table 3-4). Thirty days post-treatment (60 total days on agar), survival was

reduced to 5 1.4% by all metsulfuron concentrations or surfactant compared to 36.6%

survival of untreated checks (Table 3-4). Exposure to metsulfuron greatly reduced

gametophyte survival (Figure 3-4). Inhibition concentrations were 6.1 and 26.4 mg a.i. /

I for the 150 and 195 values, respectively. The survival rate of OWCF gametophytes after

30 days exposure to metsulfuron at 27 mg a.i. / I indicates that 14,000 out of 1 million

gametophytes survive treatment, while the survival rate of OWCF gametophytes treated

with 432 mg a.i. / I of metsulfuron drops to 4,000 out of 1 million.

Survival of sporophytes treated with metsulfuron or surfactant only 120 days post

treatment was lower compared to untreated sporophytes (Table 3-4). No OWCF

sporophytes developed following any treatment with metsulfuron at concentrations >

432 mg a.i. / I metsulfuron. Sporophyte development in surfactant treatments exhibited

higher survival rates than all metsulfuron treatments suggesting that the herbicide has

an added effect. Sporophyte survival declined with increasing concentrations of

metsulfuron (Figure 3-5). Inhibition concentrations for OWCF sporophytes treated with

metsulfuron were 5.6 and 24.1 mg a.i. / I for the 150 and 195 values, respectively.









Dry weight biomass of OWCF sporophytes treated with herbicide or surfactant

only 120 days post treatment was lower compared to untreated checks (Table 3-4).

Sporophytes treated with surfactant exhibited higher dry weight biomass compared to

all concentrations of metsulfuron. Similar to sporophyte survival, sporophyte weight

declined at increasing concentrations of metsulfuron (Figure 3-6). Growth reduction

concentrations of OWCF dry weight biomass treated with metsulfuron were 7.7 and

33.1 mg a.i. / I for the 150 and 195 values, respectively. Based on the total number of

gametophytes treated, the potential for a treated gametophyte to develop into a

sporophyte would be 140 out of 1 million and 20 out of 1 million for metsulfuron at

concentrations of 27 and 216 mg a.i. / I, respectively.

Experiment 2 Tolerance of Spores to Selected Herbicides

OWCF spores exposed to all concentrations of metsulfuron had lower germination

rates compared spores exposed to the other five herbicides tested in inhibiting OWCF

spore germination at 30 days (Figures 3-7 to 3-12). Germination was 0.4% for spores

exposed to 0.1 g a.i. / I metsulfuron and germination was completely inhibited at higher

concentrations. There was no difference between the germination rates of spores

exposed to surfactant, the lowest two concentrations of asulam and glyphosate, and the

lowest rate of fluroxypyr compared to untreated checks. The average germination rate

in untreated checks and surfactant was 47.9% (SE = 4.6) and 44.7% (SE = 5.8),

respectively. Reduction in spore germination was observed with all other

concentrations tested, ranging from 10.4% with triclopyr (40.0 g a.i. / 1) to 42.6% with

asulam (4.2 g a.i. / 1). Spores treated with herbicides other than metsulfuron

transitioned into two-dimensional gametophytes. Gametophytes that developed from

spores treated with 0.1 g a.i. / I metsulfuron remained filamentous after 30 days.









Spores were highly sensitive to metsulfuron with 150 and 195 concentrations of 0.014

and 0.063 g a.i. / I, respectively, and > 1000-fold more sensitive to metsulfuron

compared to 195 concentrations of any other herbicide tested (Table 3-5). Germination

declined exponentially with increasing concentrations of metsulfuron (Figure 3-7). The

five other herbicides tested resulted in a gradual decrease in spore germination with

increasing concentration (Figures 3-8 to 3-12), but no herbicide completely inhibited

germination. The 195 concentration (69.7 g a.i. / 1) of imazapyr, which is another ALS

inhibitor, was > 1100-fold greater than metsulfuron. The 195 concentrations for

glyphosate, imazapyr, triclopyr, fluroxypyr, and asulam are higher than the maximum

label rate of these herbicides.

Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium

Germination of OWCF spores from 12 locations of varying age exposed for 15

days on agar with metsulfuron (0.4 g a.i. / 1) was not different than those on untreated

agar (P=0.85, Figure 3-13). After 45 days of exposure, survival of gametophytes was

76% lower on agar with metsulfuron compared to those on untreated agar (P=0.0001).

The length and width of OWCF gametophytes grown on untreated checks were 3.5 x

and 4.0 x greater in length and width, respectively, than germinated OWCF spores on

agar + metsulfuron at 45 days post-exposure (data not shown). Gametophyte

development on untreated checks exhibited two-dimensional shapes, while those grown

on agar + metsulfuron remained filamentous or plate-like.

Germination rates of OWCF spores collected from IRCC of the same age were

18% higher for spores germinating on untreated agar compared to spores growing on

agar + metsulfuron medium (P=0.001, Figure 3-14). Germination rates of spores

collected from JDSP of the same age were 43% higher for spores germinating on









untreated checks relative to spores growing on agar + metsulfuron medium (P=0.0002,

Figure 3-14).

Discussion

Experiment 1 Tolerance of Gametophytes to Metsulfuron

Findings of this study indicate that 30 day old OWCF gametophytes are highly

susceptible to metsulfuron at concentrations > 27 mg a.i. / I. Similar results of high

mortality have been reported for fern gametophytes exposed to asulam and 2,4-D

(Hickok and Kiriluk 1984; Keary et al. 2000). Thirty days post-treatment, gametophytes

exhibited some tolerance, 0.3 to 1.2 % survival, with all concentrations of metsulfuron.

The surfactant used in this study acted as a contact herbicide with high initial mortality

to gametophytes, but did not affect subsequent spore germination. No new spore

germination was observed at 30 days post-treatment for all concentrations of

metsulfuron. Sporophyte development 120 days following treatment of gametophytes

was 5 0.014% for all concentrations of metsulfuron. No sporophytes developed from

gametophytes treated with metsulfuron at concentrations > 432 mg a.i. / I, indicating that

these concentrations may be the most effective in inhibiting the ALS enzyme in mature

OWCF. This concentration (432 mg a.i. / 1) is lower than the most common

concentration (450 mg a.i./I) of metsulfuron used for aerial treatment of OWCF in

Florida, but higher than the most common concentration (90 mg a.i./I) of metsulfuron

used for ground treatments (Hutchinson et al. 2006).

Volin et al. (2004) estimated that approximately 3 x 1010 OWCF spores could be

produced in heavily infested sites 5 1 ha if 1% of OWCF had fertile leaflets. Based on

aerial estimates, approximately 43,000 ha in southern Florida are infested with OWCF

(Ferriter and Pernas 2006). Theoretically, if 10% of 43,000 ha are heavily infested with









OWCF and the entire area is treated with 216 mg a.i. / I metsulfuron, the number of

tolerant sporophytes that would develop would be approximately 4 x 108 based on an

average germination rate of 37% with 0.009 % of these gametophytes developing into

sporophytes. This indicates high potential for OWCF to develop tolerance to

metsulfuron at concentrations 5 216 mg a.i. / I. In this study, the tolerance rate of

metsulfuron treated OWCF gametophytes that developed into sporophytes was 2.0 x

10-5 treated with 216 mg a.i. / I. This tolerance is greater than the estimated resistance

rate of Arabidopsis thaliana at 1 x 10-9 (Haughn and Somerville 1987) and Nicotiana

tabacum at 2.7 x 10-8 (Harms and DiMaio 1991), but about equal to Lolium rigidum at

2.2 x 10-5 to 1.2 x 10-4 (Preston and Powles 2002) for other ALS herbicides.

Sulfonylurea herbicides, such as metsulfuron, act at the cellular level by inhibiting

acetolactate synthase, the first enzyme in the biosynthesis of the branched chain amino

acids valine, leucine, and isoleucine (Ray 1986). The effectiveness of metsulfuron on

OWCF gametophytes would be due to the concentration within the cells regardless of

the amount of active ingredient used. In bioassays of OWCF sporophytes, 0.06 mg / ml

of the ALS enzyme was extracted from leaf tissue and 5 100 pg / I of metsulfuron was

needed to inhibit the enzyme in OWCF (Atul Puri and Jeffrey Hutchinson, unpubl. data).

The amount of ALS enzyme in OWCF appears to be low; thus, only small amounts of

metsulfuron would be needed to inhibit plant growth.

Herbicide resistance for any plant can only be validated by treating second

generation plants in pot bioassays (Beckie et al. 2000). None of the surviving OWCF

sporophytes treated with metsulfuron developed sporangium 1.5 years post-treatment

from re-sprouts following cutting (for dry weight biomass determinations) 120 days









following the initial treatment. This may indicate that metsulfuron inhibits normal

development and sporangium formation in OWCF sporophytes after gametophytes

have been exposed to the herbicide. Herbicides have been reported to alter phenology

and reproduction traits in plants without effecting leaf area or biomass (Crone et al.

2009; Rice et al. 1997).

The results of this study indicate that a small percentage of OWCF may possess

an ALS gene that imparts tolerance to metsulfuron at concentrations under 432 mg

a.i./l. Under field conditions, in which the same population of OWCF is treated multiple

times with metsulfuron, this could result in resistant populations. The use of sub-lethal

concentrations of herbicides can provide inconsistent control and select for metabolic

resistance in plants (Doyle and Stypa 2004). Minimum treatment concentrations of

OWCF with metsulfuron should be > 432 mg a.i. / I and herbicides with other modes of

action should be used in rotation with metsulfuron.

Experiment 2 Tolerance of Spores to Selected Herbicides

No herbicide except metsulfuron completely inhibited OWCF spore germination

suggesting that metsulfuron translocated through the spore wall. It is unclear how

metsulfuron would affect spore germination during field application. The herbicide

concentrations used in this study are considerably greater than would be expected to

contact OWCF spores in a natural setting assuming that only a small percentage of the

applied herbicide translocates into the spores. However, if metsulfuron is translocated

into the tapetum of the sporangium of fertile leaflets, the herbicide could inhibit spore

development and result in non-viable spores. Concentrations of metsulfuron > 63 mg

a.i. / I would have to translocate into the fertile leaflets to inhibit 95% of OWCF spores.









Herbicide translocation was limited in OWCF with the majority of the herbicide

remaining at the spot of application (Chapter 2). Treatment of OWCF with radiolabeled

metsulfuron at a concentration of 270 mg a.i. /I (0.10 kg a.i. / 370 I) indicated that

17.3%, or approximately 46.7 mg a.i. / I, of the applied herbicide remains at the spot of

application. This concentration is less than the concentration of 63 mg a.i. / I required

for 95% spore inhibition. At the standard aerial treatment rate of 450 mg a.i. / 1(0.084

kg a.i. / 76 I) for metsulfuron on OWCF in natural areas, approximately 78 mg a.i. / I of

metsulfuron would remain in the upper foliage where the majority of fertile leaflets are

present. For standard ground treatments of OWCF using 90 mg a.i. / 1(0.084 kg a.i. /

379 I) of metsulfuron, approximately 15.5 mg a.i. would remain at the spot of application

which is less than the 63 mg a.i. / I required for 95% spore inhibition. Theoretically, the

standard amount of metsulfuron foliar applied to OWCF during aerial treatments in

natural areas could inhibit 95% of OWCF spores, while the standard ground treatments

are unlikely to affect OWCF spore germination. It is unlikely that standard herbicide

application of the other herbicides tested in this study will inhibit germination of OWCF

spores. Growth and development of seedlings following germination of seeds exposed

to herbicides was inhibited for some pasture and turf weeds (Egley and Williams 1978;

Young et al. 1984). These results may explain was why lower OWCF cover was

observed on tree islands treated with metsulfuron compared to glyphosate (Chapter 4).

The differential susceptibility of OWCF spores to different metsulfuron

concentrations was not seen in the other five herbicides. The spore walls of ferns are

composed of fatty acids and cellulose (Banks 1999; Pettitt 1979), and fern spores have

a low affinity to water and hydrate slowly (Ballesteros and Walters 2007). It is possible









that metsulfuron is capable of penetrating the spore wall of OWCF spores and

preventing germination, while the other herbicides tested do not penetrate the spore

wall, even after 24 hours exposure. Germination rates of OWCF spores with

metsulfuron were < 0.5% at 0.1 g a.i. / I and 0.0% at higher concentrations, while

germination rates of OWCF spores exposed to imazapyr, also an ALS inhibitor, were

19% with highest rate (24 g a.i. / 1) tested. Different responses of tall morningglory

germination rates with herbicides of similar chemical structures have been reported

(Cole and Coats 1973). These differences between the germination rates of OWCF

with two ALS inhibiting herbicides may also be due to differences in the inert ingredients

associated with the metsulfuron-containing product used (Escort XP) that allow

metsulfuron to permeate the cell wall.

Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium

OWCF spores grown on agar medium containing metsulfuron had lower

germination and growth rates than spores grown on untreated checks. These results

suggest that the residual activity of metsulfuron impacts gametophyte development,

growth and survival. Under greenhouse conditions, the growth rate of new OWCF

sporophytes that developed under OWCF foliar treated with metsulfuron at 0.8 g a.i. / I

exhibited slower growth rates from 15 to 35 weeks post-treatment compared to

sporophyte growth rates under OWCF climbing fern treated with 50 g a.i. / I of

glyphosate (Hutchinson and Langeland 2008). Sulfometuron, an ALS inhibiting

herbicide, was reported to inhibit spore germination of Dennstaedtia punctilobula when

applied to soils inoculated with spores at concentrations ranging from 30 to 210 g a.i.

(Groninger and McCormick 1991).









The average half-life of metsulfuron is 30 days in soil (Vencill, 2002). Based on

the concentration of 0.4 g a.i. / I metsulfuron incorporated into the agar medium, the

amount of herbicide present in the agar would be 0.025 g a.i. / I at 120 days. This

concentration is slightly lower than the 195 value for gametophyte and sporophyte

survival in experiment 1, but 2.5-fold lower than the 195 value for spore germination in

experiment 2. Control of new OWCF growth from spore germination could be achieved

for 120 days under ideal conditions in which the half-life of metsulfuron is 30 days.

At germination, a single cell rhizoid is formed which acts to anchor and supply

nutrients to developing gametophytes (Banks 1999). The initial development of the

single cell rhizoid could absorb small amounts of metsulfuron and result in reduced

growth or eventual mortality to gametophytes. Metsulfuron at concentrations of 26.4 mg

a.i. / I inhibited OWCF gametophyte survival by 95%. OWCF spores present in the soil

are unlikely to be inhibited from germination by metsulfuron, but recently germinated

spores and young sporophytes could be inhibited by the presence of small amounts of

herbicide present in the soil. Metsulfuron could also be absorbed by expanding

rhizomes and roots as OWCF sporophytes mature, further affecting growth.

Invasion of OWCF occurs primarily on hydric soils in Florida with environmental

conditions of high temperature, high humidity, and soil moisture during the summer

months. The persistence of metsulfuron based on its half-life in the soil decreases

under these conditions (Vencill 2002). Based on a hypothesized half-life of 15 days for

metsulfuron under high temperatures and moist soils, the half-life of the herbicide would

be 25 mg a.i. / I 60 days post-treatment, an amount capable inhibiting gametophyte

survival up to 95%. The persistence of metsulfuron in the soil in Florida may be longer









during the winter months under conditions of lower soil moisture, temperatures and

humidity. This may explain why OWCF treated with metsulfuron results in less new

growth for several months following treatment compared to glyphosate. Glyphosate is

rapidly adsorbed to soil particles and has no soil residual activity (Vencill 2002).

It is unclear how OWCF spores in the mesic and hydric soils of Florida are

affected by aging, climate, water level, and microbes. In experiment 1, OWCF spores

maintained under ambient temperatures in open plastic bags in the lab exhibited

reduced germination for spores > 200 weeks of age. Conway (1953) found that spores

of Pteridium aquilinum buried under soil for soil for 14 months and then sown on agar

had germination rates of 19.7% while spores kept dry for 14 months and then sown on

agar had germination rates of 65.5%. OWCF spores exposed to environmental

conditions may exhibit a decrease in viability over time, and possibly become more

susceptible to lower concentrations of residual metsulfuron in the soil as they germinate.

Thus, the concentration of metsulfuron used should influence the impact of the

treatment, and possibly pre-emergent and post-emergent effects on OWCF spores,

gametophytes and sporophytes. The results of this study suggest that the most effective

control will be achieved using metsulfuron at > 432 mg a.i./ I and may prolong

resistance of OWCF to metsulfuron.









Table 3-1. Site, county, managing agency and geographical pre-positioning system (GPS) coordinates of locations where
OWCF spores were collected.
Managing GPS
Site County Agency 1 Coordinates
Allapattah Flats Martin SFWMD N270 08'47.6", W0800 32'17.0"
Balm Scrub Hillsborough HCPRCD N270 44'26.8", W0820 17'43.0"
Blue Cypress Conservation Area Indian River SJRWMD N270 40'29.9", W0800 44'41.2"
DuPuis Reserve Martin SFWMD N260 53'01.5", W0800 29'57.8"
East Orlando Water Facility Orange Orange County N280 30'20.0", W0810 11'22.2"
Fisheating Creek W.M.A. Glades FFWCC N270 01'54.7", W0810 26'17.1"
Highlands Hammock State Park Highlands Florida Park Service N270 26'11.4", W0810 31'46.5"
Immokalee (Pacific Tomato Growers) Collier Private Business N260 17'20.5", W0800 25'10.3"
Indian River Community College St. Lucie State of Florida N270 19'38.8", W0800 24'04.8"
Jonathan Dickinson State Park Martin Florida Park Service N27 00'33.3", W0800 07'41.6"
Lakeland Waste Water Management Facility Polk Polk County N270 55'41.2", W0810 57'04.9"
Loxahatchee National Wildlife Refuge Palm Beach USFWS N260 31'24.0", W0800 16'51.7"
1 SFWMD (South Florida Water Management District), HCPRCD (Hillsborough County Parks, Recreation, and
Conservation Department), SJRWMD (St. John's River Water Management District, FFWCC (Florida Fish and Wildlife
Conservation Commission), USFWS (United States Fish and Wildlife Service).









Table 3-2. Collection site and age of OWCF spores in weeks for each experiment.
Age of OWCF Spores (weeks)
Collection site Experiment 1 Experiment 2
Allapattah Flats 1 & 72 N/A 2
Balm Scrub 37 N/A 1
Blue Cypress Conservation Area N/A N/A 1
DuPuis Reserve 169 & 380 N/A 2
East Orlando Water Facility N/A N/A 4
Fisheating Creek W.M.A. N/A N/A 7
Highlands Hammock State Park N/A N/A 1
Immokalee (Pacific Tomato Growers) 6, 16 & 31 N/A 1
Indian River Community College 11 & 36 29 2
Jonathan Dickinson State Park 4, 10, 12, 141 & 178 20 3
Lakeland Waste Water Management Facility 7 & 35 N/A 1
Loxahatchee National Wildlife Refuge 9, 12, 19 & 83 27 4


Experiment 3
5
2
70
?6 & 432
10
8
3 & 74
6 & 40
?1 &46
6, 45 & 194
7
& 20


Table 3-3. Herbicide concentrations used to evaluate effects on Old World climbing fern spore germination.
Herbicide Concentrations (g a.i. / 1)
Imazapyr 1.5, 3.0, 6.0, 12.0, and 24.0
Glyphosate 3.2, 6.4, 12.8, 25.6, and 51.2
Metsulfuron 0.1, 0.2, 0.4, 0.8, and 1.6
Fluroxypyr 1.3, 2.6, 5.2, 10.4, and 20.8
Asulam 2.1, 4.2, 8.4, 16.8, and 33.6
Triclopyr 2.5, 5.0, 10.0, 20.0, and 40.0
Surfactant 0.05% (v/v)
Untreated checks Distilled water








Table 3-4. Effects of various concentrations of metsulfuron on germination of Old World climbing fern spores and
sporophyte survival.
Metsulfuron Gametophyte Sporophyte
Concentration % Germination % Survival % Survival Dry Weight (g)
(mg a.i. / 1) Pre-treatment 30 days post-trt 120 days post-trt 120 days post-trt
27 34.9 (4.9)1 1.4 (0.3) 0.014 (0.007) 0.061 (0.019)
54 39.7 (4.3) 1.2 (0.4) 0.009 (0.002) 0.003 (0.002)
108 39.8 (5.0) 0.8 (0.3) 0.010 (0.005) 0.002 (0.001)
216 37.6 (4.5) 0.9 (0.3) 0.002 (0.001) 0.001 (< 0.001)
432 38.9 (4.7) 0.4 (0.2) 0.000 (0.000) 0.000 (0.000)
864 34.0 (4.9) 0.3 (0.1) 0.000 (0.000) 0.000 (0.000)
Surfactant 40.1 (4.5) 1.0 (0.3) 0.048 (0.019) 0.141 (0.034)
Untreated checks 32.0 (5.1) 36.6 (5.4) 0.421 (0.187) 0.697 (0.094)
LSD (P < 0.05) NS 4.8 0.185 0.106
1 Means of 32 replications followed by standard error.









Table 3-5. Inhibition rates (15o and 195) for germination of Old World climbing fern spores
following exposure to six different herbicides based on non-linear regression.
Old World Climbing Fern
Spore Inhibition
Herbicide Rate (g a.i. /1I) 95% Confidence Interval
Imazapyr
150 16.12 13.22 21.12
195 69.67 52.25 103.81
Glyphosate
15o 56.35 47.89 70.43
195 243.56 165.62 435.97
Metsulfuron
15o 0.014 0.011 0.017
195 0.063 0.057 0.069
Fluroxypyr
15o 22.88 18.76 29.06
195 98.86 71.18 158.18
Asulam
Iso 66.01 52.15 89.11
195 285.31 228.25 370.90
Triclopyr
Iso 17.96 15.80 21.01
195 77.61 61.31 104.77



















Collection Sites:

1) East Orlando Water Facility
2) Lakeland Waste Water Mngt. Facility
3) Blue Cypress Conservation Area
4) Balm Scrub 2
5) Indian River Community College 3
6) Highlands Hammock State Park 6 5
7) Allapattah Flats
8) Jonathan Dickinson State Park 97 8
9) DuPuis Reserve 11
10) Loxahatchee N.W.R. 10
11) Fisheating Creek W.M.A. 12
12) Immokalee (PTG)








Figure 3-1. Location of collection sites of Old World Climbing Fern spores in Florida.













45 -
a

40 a


35
a
c 30-

0
E 25 -


S20


15


10


5
b

0 ---10 -- 11-2 0---
< 10 11-20 21-40 70-100 101-200 >300
Age (weeks)

Figure 3-2. Germination rates of Old World climbing fern spores of different age groups (LSD = 15.0). Lines represent
standard error.











60



a
50

a a
a
a ab
40 -


.0

E 30
( bc


0c
-I----




20 -




10
0 -- ---- ---- ---- --------------








Immokalee Lox NWR JDSP Allapattah IRCC Lakeland Balm
Location (South to North)

Figure 3-3. Germination rates (%) of Old World climbing fern spores collected from different location in Florida (LSD =
14.1). Means of 28 replications with standard error bars.










40

y = 31.99 e -(0.1135x)
35 R2 = 0.98


30


> 25




15
a 20 -

C.
,w 15 -
E

10


5


0 i
0 27 54 108 216 432 864
Metsulfuron Concentrations (mg a.i. / I)
Figure 3-4. Survival of Old World climbing fern gametophytes 30 days after (60 total days in agar + nutrient medium)
exposure to metsulfuron in agar medium. Means of 32 replications with standard error bars.










0.6


y = 0.42 e -(0.1245x)
0.5 R2 = 0.98



0.4



n 0.3


0.
C 0.2 -



0.1





0 27 54 108 216 432 864
Metsulfuron Concentration (mg a.i. / I)
Figure 3-5. Survival of Old World climbing fern sporophytes 120 days (60 days in agar + nutrient medium and 60 days in
potted soil) after exposure to metsulfuron. Means of 32 replications with standard error bars.













0.7 I
0.7 R2 = 0.99

0.6


.M 0.5






0.4

0.2


0.1



0 27 54 108 216 432 864
Metsulfuron Concentrations (g a.i. I1)
Figure 3-6. Old World climbing fern sporophyte dry weight (grams) at 120 days (60 days in agar + nutrient medium and
60 days in potted soil) after exposure to metsulfuron on agar medium and transplanted to potting soil. Means of
32 replications with standard error bars.










9


8-

S7y = sqrt 47.9 e -(47.85x)
7-7
= R 0.99
'6-



5


4
r,-
0

E- 3
E


0



0 7

0.0 0.1 0.2 0.4 0.8 1.6
Metsulfuron Concentration (g a.i. I)
Figure 3-7. Effects of metsulfuron on Old World climbing fern spore germination 30 days post exposure. Data are shown
as percent germination with all exposure times combined. Means of six replications per concentration with
standard error bars.










60


y = 47.51 e -(0.0123x)
R50 IR2 = 0.94






30



( 20
o
Q.

10



0
0.0 3.2 6.4 12.8 25.6 51.2
Glyphosate Concentration (g a.i. I1)
Figure 3-8. Effects of glyphosate on Old World climbing fern spore germination 30 days post exposure. Data are shown
as percent germination with all exposure times combined. Means of six replications per concentration with
standard error bars.










60


y = 35.06 e -(0.0384x)
50 R2 = 0.47



40



0 30



20
0
0.
Cn






0.0 1.5 3.0 6.0 12.0 24.0
Imazapyr Concentration (g a.i. I)
Figure 3-9. Effects of imazapyr on Old World climbing fern spore germination 30 days post exposure. Data are shown as
percent germination with all exposure times combined. Means of six replications per concentration with
standard error bars.










60


y = 44.02 e -(0.0385x)
50 R2 = 0.95



40



0 30 -



( 20

0
10 M



0
0.0 2.5 5.0 10.0 20.0 40.0
Triclopyr Comcentration (g a.i. I)
Figure 3-10. Effects of triclopyr on Old World climbing fern spore germination 30 days post exposure. Data are shown as
percent germination with all exposure times combined. Means of six replications per concentration with
standard error bars.










60


y = 43.75 e -(0.0301x)
50 R= 0.81



40



0 30 T



S20
0

10



0
0.0 1.3 2.6 5.2 10.4 20.8
Fluroxypyr Concentration (g a.i. I1)
Figure 3-11. Effects of fluroxypyr on Old World climbing fern spore germination 30 days post exposure. Data are shown
as percent germination with all exposure times combined. Means of six replications per concentration with
standard error bars.










60


y = 46.44 e -(o.0105x)
50 R2 =0.87



40



30



( 20
0
C.

10



0
0.0 2.1 4.2 8.4 16.8 33.6
Asulam Concentration (g a.i. I1)
Figure 3-12. Effects of asulam on Old World climbing fern spore germination 30 days post exposure. Data are shown as
percent germination with all exposure times combined. Means of six replications per concentration with
standard error bars.












El Metsulfuron
0 Untreated check


45

40

35

" 30

2 25

' 20

o 15


a
I+


Initial germination (15 days)


Gametophyte survival (45 days)


Figure 3-13. Germination rates of Old World climbing fern spores placed on an agar mixed with metsulfuron (0.4 g a.i. / 1)
or agar alone (untreated checks) 15 and 45 days post-treatment from spores collected at 12 locations. Different
letters indicate significantly different (P < 0.05) treatment means within times (days) based on paired t-tests.
Means of 16 replications with standard error bars.


a

I+


10

5

0


I


- .I.












E Metsulfuron
O Untreated check


50

45

40

S35

S30

C,
-25
0
S20

-15


IRCC (46 weeks)


JDSP (36 weeks)


Figure 3-14. Germination of Old World climbing fern spores collected Indian River Community College (IRCC) and
Jonathan Dickinson State Park (JDSP) placed on an agar media mixed with metsulfuron (0.4 g a.i. / 1) or agar
alone (untreated checks) 45 days post-treatment. Different letters represent significantly different (P < 0.05)
treatment means based on paired t-tests. Means of 16 replications with standard error bars.


a
- I -









CHAPTER 4
REPEATED HERBICIDE APPLICATION FOR CONTROL OF OLD WORLD CLIMBING
FERN AND THE EFFECTS ON NON-TARGET VEGETATION ON EVERGLADE'S
TREE ISLANDS

Introduction

Lygodium microphyllum (Cav.) R. Br. (Old World climbing fern; OWCF) is a non-

native vine that transforms tree islands in A.R.M. Loxahatchee National Wildlife Refuge

(LNWR) from an open community of herbaceous ground cover of native ferns and a

shrub layer to one with a nearly impenetrable thicket of broken limbs and a closed

canopy of OWCF rachis. OWCF is known to decrease native vegetation cover,

evenness and richness on tree islands (Brandt and Black 2001). Ugarte et al. (2006)

suggested that OWCF, because of its twining growth into the upper portions of the

canopy, acts in synergy with wind events such as hurricanes to alter the structure and

composition of tree islands often causing total collapse of trees. Recruitment of OWCF

is correlated with these disturbances (Lynch et al. 2009).

In the northern Everglades wetland ecosystem of LNWR, OWCF invasion occurs

in woody vegetation (i.e., tree islands) 89% of the time compared to sawgrass (2%) and

wet prairies (9%) (Wu et al. 2006). The northern section of the refuge contains higher

densities of OWCF due to the presence of more tree islands and prevailing southwest

winds that distribute more spores into the northern section (Wu et al. 2006). Based on

models, OWCF is estimated to exceed the cover of the five most invasive non-native

species (Melaleuca quinquenervia, Schinus terebinthifolius, Casuarina equisetifolia,

Colubrina asiatica, and Ardisia elliptica) combined in south Florida by 2014 if left

unmanaged (Volin et al. 2004).









Models indicate that OWCF may infest 38% (1910 ha) of all tree islands within

LNWR by 2012 if left untreated (Wu et al. 2006). Woodmansee et al. (2005) similarly

estimated that the cover of OWCF could impact 25,200 ha in LNWR, which represents

42% of the total area within the refuge. Attempts to track the spread of OWCF in

southern Florida with aerial flights have proven difficult. Based on aerial surveys from

fixed-wing aircraft from 1993-2003, changes in OWCF in southern Florida cover ranged

from a 14% increase in 1997, a 255 % increase in 1999, a 64% decrease in 2001, and a

181% increase in 2003 (Ferriter and Pernas 2006). These large differences in changes

in area coverage of OWCF from aerial surveys can be attributed to surveys flown

following periods of freezing weather that top-killed OWCF (Ferriter and Pernas 2006).

Regardless, it is clear that OWCF has become a major invader of forested wetlands in

Florida's natural areas.

Infestations of OWCF often occur in natural areas that are remote and

inaccessible to vehicles and personnel, making treatment difficult. This is especially true

in the Florida Everglades. Spore production can occur throughout the year in south

Florida and each fertile leaflet is capable of producing 28,600 spores (Volin et al. 2004),

indicating that annual treatments will be required to control existing populations as well

as re-sprouts and emergent sporophytes. The rapid spread of OWCF across the

landscape of southern Florida can be attributed to its small (<65 pm) spores which

move with normal wind events and are moved further away from wind events such as

tropical storms and hurricanes. In addition, the four life stages (spore, gametophyte,

sporeling and sporophyte) of OWCF make management of the fern extremely difficult

without continual monitoring and re-treatment.









The greatest area of OWCF infestations occur in Martin and Palm Beach Counties

of southern Florida, the general location of the first documented record from a

naturalized specimen in Florida (Beckner 1968; Pemberton and Ferriter 1998). Within

LNWR, OWCF was first noted in the interior in 1989, and was documented from a few

tree islands in 1992 and 1993 (Thomas 2006). During 1995, OWCF covered an

estimated 7,284 ha of LNWR. The cover increased to 19,433 ha in 2003 based on aerial

transects conducted by the SFWMD (Thomas 2006). As stated above, the most recent

estimate of the coverage of OWCF in LNWR was 25,200 ha (Woodmansee et al. 2005).

In 2007, extensive aerial and ground herbicide treatments were initiated throughout the

refuge with follow-up treatments scheduled in 2009-2010.

Management efforts throughout southern Florida using glyphosate and

metsulfuron have not been successful for long-term control of OWCF; vigilant annual re-

treatments are required to keep the fern in check (Hutchinson et al. 2006). Brandt

(2004) found that glyphosate was more successful than metsulfuron in controlling

OWCF, while metsulfuron resulted in less non-target damage up to 1 year post

treatment in LNWR. Glyphosate is a non-selective herbicide that damages or kills many

types of plants (i.e., grasses, herbs, ferns, shrubs, trees). Graminoids such as Panicum

hemitomon and Cladiumjamaicense have shown tolerance to metsulfuron applied to

control OWCF (Langeland and Link 2006).

Natural areas present a complicated situation for invasive plant treatment while

protecting native, non-target plant species. Selective herbicides are traditionally

developed to control multiple weeds in agricultural areas with no effect on the crop

(Hobbs and Humphries 1995). However, in natural areas, the goal of management is to









control a single weed while protecting or limiting damage to many non-target species.

Evidence suggests that aerial herbicide treatment of OWCF can be selective with

minimal impacts to non-target deciduous trees during winter application in Florida

(Hutchinson et al. 2006). However, evergreen non-target ground vegetation is often

greatly reduced (Hutchinson et al. 2006). This is also true during re-treatment of OWCF

because of its twining and climbing nature as it grows over and around non-target

plants. The mission of the United States Fish and Wildlife Service (USFWS) refuge

system is to manage lands and water for the conservation of fish, wildlife, and plant

resources and their habitats (NWRS 2010); thus, finding herbicides that can effectively

control OWCF while limiting non-target damage would be in compliance with the

mission of USFWS.

Tree islands in this study were in the late stages of invasion by OWCF. However,

in 2004, Hurricanes Frances and Jeanne (National Hurricane Center 2009) defoliated

OWCF on most tree islands (Ugarte et al. 2006). Prior to the hurricanes, tree islands

selected for this study were visually observed from a helicopter to have OWCF

coverage >90%. At the start of the project in December 2005, OWCF cover had

returned to approximately 60% cover indicating recovery can be rapid from established

populations of OWCF following hurricane damage.

Due to the isolation of tree islands (some in LNWR take >1 hour to reach by

airboat) and the large number of islands (several thousand in LNWR), aerial application

of herbicide is the most effective and cost efficient method for initial control of OWCF.

The objectives of this study were to evaluate the effectiveness of consecutive annual

herbicide applications of glyphosate and metsulfuron using aerial and ground









treatments for control of OWCF and to examine the unintended impacts to native

ground and canopy vegetation on northern Everglade's tree islands in LNWR.

Materials and Methods

Study Site

Studies were conducted on tree islands at Arthur R. Marshall Loxahatchee

National Wildlife Refuge (260N, 80W; LNWR) in western Palm Beach County, Florida

from 2006-2009. LNWR encompasses 59,464 ha and represents the last remnants of

the northern Everglades. The Everglades ecosystem is listed as an International

Biosphere Reserve, World Heritage Site, and a Wetland of International Importance

(NPS 2010). It is a peat-based wetland that is comprised of sawgrass strands, wet

prairies, sloughs, and tree islands (Loveless 1959). Tree islands number in the

thousands and are a unique upland feature of the Everglades wetland ecosystem that

provide habitat for wildlife (Brandt et al. 2000, 2003). The largest tree islands are <125

ha but 85% of the tree islands are <0.13 ha (Brandt et al. 2000). Elevation on tree

islands range from 0.2-1.0 m above the surrounding marsh (Wetzel et al. 2005). Tree

islands selected for this study were located in the north portion of LNWR (26035'21" N,

80020'33" W) and <0.13 ha in size. With the exception of one native neotropical tree

species, Rapanea punctata, all understory and canopy plants in the northern end of

LNWR were temperate species. Neotropical plant species are more common in the

southern end of LNWR and the southern Everglades. The three most common invasive

plants in the LNWR are OWCF, Melaleuca quinquenervia, and Schinus terebinthifolius

which are tropical in origin and occur in the northern part of LNWR.

The locations of tree islands heavily infested with OWCF were mapped from a

helicopter with a Garmin 76S global positioning system (GPS). Tree islands with >50%









cover of OWCF based on visual observation and the presence of OWCF in the canopy

were then selected from an airboat. Selected tree islands were marked at the south

end with a 3 m PVC pipe. All PVC markers were painted with a specific color per

treatment for visual reference and numbered with a permanent tag. GPS coordinates

were taken at each tree island using a Garmin 76S. The use of GPS coordinates and

color coded PVC markers allowed for the specific herbicide treatments to be applied by

helicopter to the exact island designated for herbicide treatment with glyphosate or

metsulfuron. A researcher involved with the project flew with the pilot during herbicide

application to verify the designated herbicide treatment was applied to correct tree

islands.

Evaluation Periods

Pre-treatment evaluation periods were conducted December 2005 January 2006

(year 1). Post-treatment evaluation periods were conducted in December 2006 -

January 2007 (year 2), December 2007 January 2008 (year 3), and December 2008 -

January 2009 (year 4). For each evaluation period, cover of the vegetation in two strata

(soil level to 1.5 m and >1.5 m above the soil), and status (live or dead) of dominant

shrubs and trees (l/ex cassine, Persea palustris, Myrica cerifera, and Rapanea

punctata) were recorded.

Herbicide Treatment

One of the eight treatment schedules listed in Table 4-1 were randomly applied to

tree islands (all with >50% OWCF cover). Each treatment was replicated on five

individual islands. An additional five tree islands were randomly selected as untreated

checks (total n = 45 islands). For each treatment, five tree islands were clustered within

400-500 m of each other to facilitate aerial treatment. Aerial applications were applied









by Helicopter Applications, Inc. (Gettysburg, PA) from a Bell 206 L4 helicopter with a

Thru Valve Boom (TVB, Waldrum Specialties Inc., Southampton PA). Ground

treatments were applied on a spray to wet basis using 1.5 I hand-held pressurized spray

bottles. The first ground treatment was applied by LNWR staff, and the second ground

application was applied by contractors (Applied Aquatics Management, Inc., Bartow,

Florida). Aerial applications contained 0.5% v/v DLZ adjuvant and ground applications

contained 0.5% v/v Sunwet adjuvant. Ground applications were directed toward OWCF

only to minimize contact with native vegetation.

Ground Cover

Cover of all ground vegetation <1.5 m was estimated along three randomly placed

transect lines the entire width of each tree island (Canfield 1941) (Figure 4-1). The

beginning and end of each line were marked with PVC pipe. During each evaluation

period, a meter tape was extended from the beginning to end of each transect line. All

ground cover estimates were recorded along the transect line individually by plant

species. Ground cover was calculated for all plants by species along the transect line by

dividing the total distance of each species overlapping the line by the length of the

transect line. The three lines on each tree island were averaged to obtain the mean

ground cover of each species per tree island. Changes in plant composition and

structure on each tree island were evaluated pre-treatment and post-treatment three

times.

Canopy Cover

Percent canopy cover was estimated at the beginning, middle, and end of every

transect using a concave densiometer (Forest Densiometer, Bartlesville, OK) for all

transect lines <10m in length. For transect lines >10m in length, canopy cover points









were taken at the beginning, end, and at 25% and 75% of the length of the line.

Densitometer readings were taken in each cardinal direction at each point. Each

reading was multiplied by 1.04, and the four reading per point were averaged for canopy

cover (Lemmon 1957). Percentage canopy cover was taken for live and dead

vegetation >1.5 m above the ground. Canopy cover along all three lines was averaged

to obtain the mean canopy cover per tree island.

Plant Species

Species lists for both native and exotic plants were compiled at heights <1.5 m

along each line transect to determine compositional changes over the evaluation

periods. Botanical nomenclature followed that of Wunderlin and Hansen (2003). Plants

that could not be identified in the field were submitted to the University of Florida

herbarium for identification. This information was used as a site condition assessment to

determine the effects on compositional changes to plants from repeated herbicide

application. No tree islands used in this study had been previously treated with

herbicide. Plants were classified as late succession (species typical of tree islands with

canopy intact), ruderal (species not typically present on tree islands with canopy intact)

and exotic (non-native species not indigenous to the area). Late succession species

were based on the 16 most common native plant species documented during pre-

treatment evaluation and those recorded by Brandt et al. (2003).

Species richness patterns were determined by counting the number of plant

species that intersected each transect line. The three lines on each tree island were

averaged to obtain the mean species richness per tree island. Species evenness

patterns were determined using the methods of William (1964a). The three lines on

each tree island were averaged to obtain the mean evenness per tree island. Evenness









patterns were calculated as E = 1 / DS, where D is Simpson's Index [Z (n/N)2] and S is

species richness (Williams 1964a).

Native Shrubs and Trees

For each treatment, Myrica cerifera (n = 60), Ilex cassine (n = 60), Persea palustris

(n = 60) and Rapanea punctata (n = 20) individuals were randomly selected and tagged

within 5-10 m on either side of the transect line among the five tree islands per

treatment. Randomly selected shrubs and trees were marked with a numbered

aluminum tag to examine the effects of aerial herbicide application on non-target native

trees over the duration of the project. The location of each tree along the line was

recorded. All tagged shrubs and trees were visually evaluated by 1) height (53m or

>3m), 2) d.b.h. (58 cm or >8 cm) and 3) percent live foliage (575% or >75%). These

values were subjectively categorized, but allowed for rapid assessment of the size

(small or large) and condition (partial or full foliage) of the plants in the field. All shrubs

and trees were documented as alive if any live foliage was present or there were basal

re-sprouts during the final evaluation at year 4. Salix caroliniana (Carolina willow) and

Cephalanthus occidentalis (buttonbush) were omitted because they had entered

dormancy. Both species are common along the edges of tree islands but were not

included in analysis because no or only partial foliage was present during the evaluation

periods.

Statistical Analysis

OWCF

Data were analyzed to determine the most effective aerial treatment (year 2),

follow-up ground treatment (year 3), and consecutive annual treatments (year 2-4) for

control of OWCF.









A linear mixed-effect analysis of variance was used to assess differences in

OWCF cover following aerial herbicide treatments (year 2). Proc Mixed procedure was

used with the initial cover at year 1 as a covariate of year 2 and random effects of tree

islands (SAS 9.2). Means were separated using Tukey's multiple comparison test

(P<0.05). Data were transformed using arcsine square root transformation to improve

homogeneity of variance assumptions (Zar 1999) by visual inspection of the plotted

residuals.

A linear mixed-effect model was used to compare a follow up ground

treatment (year 3) following aerial treatment (year 2) with factors of treatment and time

in Proc Mixed (SAS 9.2). Year two data were used as a covariate of year 3 with random

effects of tree islands. Means were separated using Tukey's multiple comparison test

(P<0.05). Data were transformed using arcsine square root transformation to improve

homogeneity of variance assumptions (Zar 1999) by visual inspection of the plotted

residuals.

Repeated measures analysis of covariance (ANCOVA) was used with the pre-

treatment conditions at year 1 as covariates for three consecutive annual treatments

(year 2-4) for control of OWCF. Data were analyzed using Proc Mixed repeated

measures ANCOVA (SAS 9.2) with time as the repeated measure, treatment as a fixed

factor, and transect lines and tree islands as random factors. Variables were analyzed

by treatment, time, and the treatment by time interaction with pre-treatment conditions

as covariates. Means were separated using Tukey's multiple comparison test (P<0.05).

Data were transformed using arcsine square root transformation to improve









homogeneity of variance assumptions (Zar 1999) by visual inspection of the plotted

residuals.

Cover, Richness, and Evenness

Separate analyses were conducted to evaluate the responses of native species

ground cover, canopy cover, total species richness, late succession species richness,

ruderal species richness, other non-native species richness (not including OWCF), and

species evenness to experimental treatments. For ground cover, separate analyses

were conducted for the 11 late succession species that occurred on tree islands

throughout the three year study. All ruderal and other non-native species occurred too

infrequently to be analyzed individually.

Data were analyzed using Proc Mixed ANCOVA (SAS 9.2) with time as the

repeated measure, treatment as a fixed factor, and transect lines and tree islands as

random factors. Variables were analyzed by treatment, time, and the treatment by time

interaction with pre-treatment conditions as covariates. Data were transformed using

arcsine square root transformation to improve homogeneity of variance assumptions

(Zar 1999) by visual inspection of the plotted residuals.

Shrub and Tree Survival

Analysis of tree survival pattern by species were assessed by multiple logistic

regression to model the dichotomous variable (0=dead, 1 =live) as a function of the

nominal variable treatment and ordinal variables height, d.b.h., foliage and interactions.

Proc Logistics (SAS 9.2) was used for all variables, and model significance was based

on chi-square values with P<0.05. Pairwise contrasts were used to compare treatment

responses that were significant at P<0.05.









Results


OWCF Ground Cover

All aerial herbicide treatments reduced OWCF coverage to less than 9% one year

after application, or an 88% to 99% decrease compared to untreated checks (Table 4-

2). Metsulfuron reduced coverage to a greater extent than glyphosate but the impact

was independent of concentration (P<0.05).

Two years after aerial treatments and one year following single ground treatments

OWCF cover was less than 20%, translating to a 67% to 97% decrease compared to

untreated checks (Table 4-3). As observed after a single aerial application, islands

treated with a single metsulfuron ground application had lower OWCF coverage than

those treated with glyphosate but no rate effect was observed for either herbicide

(P<0.05).

A second ground treatment, following aerial treatment and the initial ground

application, again resulted in much reduced OWCF cover compared to untreated

checks and, again, metsulfuron treated islands had lower cover than those treated with

glyphosate (P<0.05; Table 4-4). Cover the year following aerial application and a single

ground application was somewhat higher on all islands, treated and untreated, than the

year following aerial treatment only or the year following the second ground application.

Numerous new sporophytes <1.5 cm in height and some re-growth of OWCF were

observed on all treated islands during the final evaluation at year 4. OWCF was not

completely eliminated by any combination of treatments during this study. OWCF cover

continuously increased with a 41.3% increase in ground cover by the end of the 4-year

study on untreated islands.









Native Plant Ground Cover

Both Blechnum serrulatum (P<0.05; Table 4-5) and Smilax laurifolia (P<0.05;

Table 4-6) cover were up to 80% and 90% lower, respectively, on all herbicide treated

islands three years following aerial application and two annual ground treatments

compared to the untreated island controls. I observed no effect of herbicide or

concentration in cover of these species. The greatest reduction for both species

occurred the year following aerial application.

Osmunda cinnamomea cover was reduced by both metsulfuron treatments

compared to untreated islands (P<0.05; Table 4-7), but showed no response to

glyphosate.

Pre-treatment cover of Woodwardia virginica was only 0.6% to 1.5% and changed

little during the study. Three years after the aerial application and two ground

treatments, cover on tree islands treated with metsulfuron was not different than

untreated islands, while those islands treated with glyphosate had higher cover than

untreated islands (P<0.05; Table 4-8).

Vitas rotundifolia cover increased overall from pre-treatment levels. This species

became the most common native plant based on cover at three years post-treatment.

Cover on metsulfuron treated islands was not different than untreated islands and was

higher on those islands treated with glyphosate (P<0.05; Table 4-9).

Pre-treatment cover of Osmunda regalis was low, decreased overall following the

initial aerial herbicide applications, and remained low throughout the study. There were

differences in cover among metsulfuron treated islands and glyphosate and untreated

islands (P<0.05; Table 4-10), and it was completely absent from those islands treated

with metsulfuron.









Pre-treatment cover of Cladiumjamaicense, which was recorded almost

exclusively around edges of the tree islands, was less than 1% and remained rare

during the study. No changes in cover were observed during the study and no

differences were observed among herbicide treated or untreated islands (P=0.18; Table

4-11). It was, however, completely absent from glyphosate treated islands.

Shrub and Tree Ground Cover

Myrica cerifera cover, following all three herbicide applications, was less than

1.0% on glyphosate treated islands compared to over 6.0% on untreated islands while

cover on metsulfuron treated islands was not different than untreated islands (P<0.05;

Table 4-12). Persea palustris cover was lower on all herbicide treated islands

compared to untreated islands, except for the higher rate of metsulfuron, which was not

different than untreated islands (P<0.05; Table 4-13). Islands treated with the lower

rates of glyphosate had the least Persea palustris cover than any treated or untreated

islands. Ilex cassine cover was less on all herbicide treated islands compared to

untreated islands but there were no differences among herbicide treatments (P<0.05;

Table 4-14). Rapanea punctata cover was very low pre-treatment and no changes

were observed during the study (P=0.43; Table 4-15), however no Rapanea punctata

ground cover was observed on metsulfuron treated islands after pretreatment sampling.

Canopy Cover

One year following the three herbicide applications, canopy cover was lower on all

herbicide treated islands compared to untreated islands and lower on glyphosate

treated islands than on those treated with metsulfuron (P<0.05; Table 4-16). Canopy

cover was reduced 88% to 71% on glyphosate treated islands compared to untreated

islands but only 31% to 48% on metsulfuron treated islands.









Shrub and Tree Survival

Ilex Cassine

There were no differences (P=0.0963; Table 4-17) in the survival of Ilex cassine

between herbicide treated islands and untreated islands indicating this species, though

damaged, will survive those treatment rates of metsulfuron and glyphosate tested.

There were differences (P=0.0090; Table 4-18) in height of surviving Ilex cassine, with

plants 53 m in height more susceptible than plants >3 m. No other variables or

interactions were significant for Ilex cassine. On control islands, the survival rate of

tagged Ilex cassine was 100%.

Persea Palustris

Persea palustris was highly susceptible to glyphosate, but exhibited a high

tolerance to metsulfuron (P<0.0001 Table 4-17). There were interactions (P<0.0344)

for herbicide*foliage for Persea palustris (Table 4-19). Plants treated with glyphosate

with 575% foliage were less likely to survive compared to glyphosate treated plants with

>75% foliage or any treatment with metsulfuron. No other variables or interactions were

significant for Persea palustris. On control islands, the survival rate of tagged Persea

palustris was 100%.

Myrica Cerifera

Myrica cerifera was highly susceptible to glyphosate, but exhibited a high

tolerance to metsulfuron (P<0.0001; Table 4-17). There were differences (P=0.0217) for

percent foliage of Myrica cerifera (Table 4-20). Plants with 575% foliage were more

susceptible to herbicide treatment compared to plants >75% foliage. Significant

interactions were detected for herbicide*height (P<0.0001), herbicide*dbh (P<0.0001),

and herbicide*foliage (P<0.0001) for percent survival of Myrica cerifera. For all









significant interactions, plants treated with glyphosate <3 m in height, <8 cm in dbh and

having 575% foliage exhibited lower survival rates compared to plants treated with

metsulfuron. On control islands, the survival rate of tagged Myrica cerifera was 93%.

Rapanea Punctata

Rapanea punctata was highly susceptible to metsulfuron, but exhibited tolerance

to glyphosate, especially at the lower rate (P<0.05; Table 4-17). No other variables or

interactions were significant (Table 4-21). The highest survival percentage for Rapanea

punctata was 88% for the low rate of glyphosate. On control islands, the survival rate of

tagged Rapanea punctata was 100%.

Plant Richness Dynamics

General Trends

Sixteen climax species, 34 ruderal species, and five exotic (non-native) plant

species were documented on tree islands over the three year duration of this project

(Appendix A). General trends from pre-treatment (year 1) to three years post-treatment

(year 4) indicated no decline in climax species for any treatment, large increases in

ruderal species for all treatments, and small increases in other non-natives for

glyphosate treated tree islands (Table 4-22).

Total Species Richness

Differences in total species richness, including both native and non-native species,

were observed at three years post-treatment (Table 4-23). Total species richness was

higher on tree islands treated with the high rate of glyphosate compared to all other

herbicide treated and untreated islands (P<0.05).









Late Succession Species Richness

Lower late succession species richness was observed on tree islands treated with

the low rate of metsulfuron compared to all other treatments and untreated islands

(P<0.05; Table 4-24). There was no difference in late succession species richness

among metsulfuron (high rate), glyphosate (low or high rate) and untreated tree islands.

Late succession species not recorded at year 4 were rare at the start of the project and

were typically edge or gap species.

Ruderal Species Richness

Ruderal species richness was two to three times higher on those islands treated

with glyphosate compared to islands treated with metsulfuron and untreated islands

(P<0.05; Table 4-25). The largest increase in ruderal species was found on tree islands

treated with the high rate of glyphosate.

Other Non-Native Species Richness

The mean number of other non-native species never accounted for >1 species at

any time during the study. The number of non-native species was higher on islands

treated with the highest rate of glyphosate compared to other treatments and untreated

islands (P<0.05; Table 4-26). At three years post-treatment, seedlings of Melaleuca

quinquenervia and Schinus terebinthifolius were documented. Seedlings of Melaleuca

quinquenervia were observed at three years post-treatment on tree islands treated with

the low rate of metsulfuron, and both the low and high rates of glyphosate. Seedlings of

Schinus terebinthifolius were observed on tree islands treated with both rates of

glyphosate at the three years post-treatment.









Species Evenness

Species evenness was lower on glyphosate treated islands compared to

metsulfuron treated and untreated tree islands (P<0.05; Table 4-27). The low evenness

values for the glyphosate treated islands are attributed to large increases in Vitis

rotundifolia cover at three years post-treatment.

Discussion

Ground Cover

OWCF Management

Consistent with previous findings (Chapters 2,3), these results indicate that

metsulfuron provided significantly better control of OWCF than glyphosate for aerial,

ground follow-up and three consecutive herbicide treatments. However, hundreds of

new OWCF sporophytes were observed on tree islands for all glyphosate and

metsulfuron treated islands during the final evaluation. Limited information is available

on the viability of OWCF spores under lab or natural conditions. Spores of OWCF

maintained under dry, ambient conditions have been reported to germinate after four

years in agar (Michael Lott, Florida Atlantic University, personal communication) and

spores in experiment 2 (Chapter 3) exhibited 1.5% germination after seven years and

three months after being stored under the same conditions. The mean germination

rates of OWCF spores was 84-97% on natural soils from LNWR based on lab tests (Call

et al. 2007). This indicates that even with total eradication of mature OWCF, spores in

the soil and ground surface debris as well as those blown in by wind currents will

require long term management. Four biannual ground treatments over two years using

glyphosate, metsulfuron, triclopyr, and imazapic alone or in combination did not totally

eradicate OWCF at seven sites in central or southern Florida but most OWCF observed


100









was new growth from germinated spores (Jeffrey T. Hutchinson, unpublished data). In

contrast, Stocker et al. (2008) continued to find OWCF re-growth from rhizomes (<1%)

cover two years post treatment with triclopyr following bimonthly and biannual

treatments but reported no new growth from spores. This suggests that unless

infestations of OWCF are only a few square meters in size, re-treatment will be required

for some re-growth and new growth from spores.

In this study, aerial treatment of OWCF with metsulfuron reduced cover to <1.5%,

which is similar to the results observed for aerial treatment of bracken fern (Pteridium

aquilinum) with asulam (Pakeman et al. 2005). Both species are rhizomatous and can

re-sprout from surviving stems and fronds. For control of bracken, multiple treatments

were more effective than 1-2 treatments (Stewart et al. 2007).

Without follow-up treatments, OWCF will eventually recover on the tree islands to

pre-treatment levels. Based on the results of this study, it is possible that OWCF could

return to pre-treatment levels on tree islands in 5-6 years post treatment if no follow-up

treatments are preformed. In this study, OWCF increased in cover 1.4 to 17.5-fold on

tree islands receiving two treatments compared to three treatments. The increase in

OWCF cover at year 3 following the first ground treatment at year 2 may be due to

applicators over-looking young sporophytes. By the second ground application at year

3, these sporophytes may have become more conspicuous to the applicators and

resulted in lower cover at year 4. This indicates the difficulty in finding re-growth and

new sporophytes during re-treatment, especially in thick OWCF rachis mats. Stocker et

al. (2008) work indicates that re-treatments <6 months apart may be more beneficial to


101









native species as OWCF can be spot-treated with herbicide before it has the chance to

climb onto and twine around native vegetation.

An initial objective of this study was to evaluate the effects of three annual aerial

herbicide treatments with glyphosate and metsulfuron for control OWCF and the

impacts to native plants due to limited access and the high cost of ground treatments.

Twelve months post-treatment, the cover of OWCF was <8% for all treatments and

occurred in small clusters scattered throughout the tree islands, making an additional

aerial treatment unreasonable. An additional aerial herbicide treatment would have

resulted in additional losses of native ground and canopy cover, further changing the

structure and composition of the tree islands. Tree islands are important for providing

habitat heterogeneity in the wetland landscape of the northern Everglades and provide

food and refuge for multiple taxa of wildlife including migratory birds (Brandt and Black

2001; Brandt et al. 2003). Large scale aerial application of glyphosate over hundreds or

thousands of tree islands would greatly alter the structure of the tree islands and is not

recommended.

During the year two evaluations following the initial aerial treatment, I observed

that most of the live OWCF cover occurred <1-2 m from the edges of tree islands

indicating it was not treated. These low densities of OWCF may not have been evident

to the applicator from the air. Regardless, I recommend that the edges of the islands be

treated with metsulfuron during aerial application. Metsulfuron has no impact on

Cladiumjamaicense (Langeland and Link 2006), which is common along the edges, or

the native shrubs and trees that occur along the edges.


102









Other evaluations of ground treatments of OWCF in LNWR revealed that most of

the live OWCF documented on tree islands occurred along the edges at three and four

and half years post-treatment (Thomas and Brandt 2003; Barrett et al. 2006). In this

study, invasions of new OWCF sporophytes were common along the edges and

throughout the interior at year two and three following aerial treatments which reduced

both ground and canopy creating more openings. It appears that the combined loss of

ground and canopy cover following aerial application creates more suitable, open

conditions for OWCF spores to germinate in the interior portions of tree islands

compared to ground treatments. With ground treatments, there is minimal tree mortality

and loss of canopy cover. Increased spore germination and gametophyte development

of ferns can be attributed to increased light intensity and soil disturbance (Watkins et al.

2007). Gaps created from disturbance and loss of canopy can result in increased native

and non-native vine cover in hardwood forest of south Florida (Horvitz et al. 1998). On

Everglade's tree islands, a significant number of OWCF sporophytes were found at the

base of treefalls with reduced canopy cover following hurricanes (Lynch et al. 2009).

Plant Diversity Dynamics

The increase in species richness observed in this study following herbicide

treatment is not reflective of the vegetation typical of tree islands (Brandt and Black

2001; Brandt et al. 2003), but more reflective of a habitat modified by OWCF and

subsequent herbicide treatment. The lower number of ruderal species documented on

tree islands treated with metsulfuron may be due to limited canopy cover damage

following treatment. The total number of species recorded on tree islands prior to

treatments ranged from 15-21. This number of species is similar to number of species

(16-19) reported on tree islands with low infestations of OWCF (Brandt and Black 2001;


103









Brandt et al. 2003). However, by year 4 of the study, the number of species recorded

increased to 38-42 species on glyphosate treated tree islands and 28-31 species on

metsulfuron treated tree islands.

Tree islands treated with glyphosate exhibited a composition shift from OWCF and

late succession species (primarily native ferns) to ruderal species dominated by Vitis

rotundifolia. While Vitis rotundifolia is considered a late succession species typical of

tree islands, it was only a small component of cover at year 1, but became the dominant

native plant at year 4. Aerial treatments of bracken fern with asulam have also resulted

in a shift of vegetation changes from bracken and Ericaceae species, though in this

case to one dominated by early successional species such as mosses and grasses

(Pakeman et al. 2005). As suggested by Gordon (1998), this increase in species

richness is due to an increase in ruderal species on tree islands.

The mean number of late succession native species decreased by 1-2 species on

metsulfuron treated tree islands from year 1 to year 4, but remained the same or

increased slightly on tree islands treated with glyphosate. On tree islands treated with

metsulfuron, Osmunda regalis, 0. cinnamomea, and Smilax laurifolia became

uncommon at 4 years. For ruderal species, there was an increase of 4-6 species from

year 1 to year 4 on tree islands treated with glyphosate. On metsulfuron treated tree

islands, the increase in ruderal species was 2-3.

Evenness values for ground cover were 50.49 at all evaluation periods from year 1

to year 4 in this study, but lower for glyphosate treated islands compared to metsulfuron

treated islands. Brandt et al. (2003) reported the mean evenness value for ground

cover was 0.58 on tree islands with minimal OWCF cover. Species richness is relatively


104









low on tree islands in LNWR where OWCF is not present averaging eight native species

with Blechnum serrulatum being the most abundant ground cover (Brandt and Black

2001). The increase in the number of ruderal species, especially with the high rate of

glyphosate, likely did not affect evenness values as the cover of most ruderal species

was typically <5%. This indicates that as OWCF cover decreased following treatment,

there was a concomitant increase in Vitis rotundifolia, especially on glyphosate treated

islands, and decrease in native ferns.

Native Ground Cover

Both glyphosate and metsulfuron impacted the native ground cover on tree

islands. Aerial application of glyphosate resulted in more ruderal species at three years

post-treatment. Metsulfuron impacted native ferns more than glyphosate. Native ferns,

such as Blechnum serrulatum and Osmunda cinnamomea, have been shown to be

sensitive to metsulfuron at rates as low as 0.01 kg a.i./187 I/ha (Hutchinson and

Langeland 2008). With the exception of ruderal species, few recruits of late succession

native species were observed on treated islands. At three years post-treatment, some

small native ferns were observed growing on tussocks and at the base of dead trees,

but new growth of native ferns were far out-numbered by new OWCF sporophytes.

Community-level response to herbicide treatment is difficult to analyze as native

species cover has already been reduced by the invasive plant (Marrs 1985; Laufenberg

et al. 2005). The presence of OWCF on tree islands prior to treatment had likely

already reduced native ground cover and limited propagule production. On tree islands,

the combined impacts of OWCF (limited growth and propagule production of natives)

prior to treatment along with herbicide treatments (further reductions in native plant


105









abundance) and thick rachis mats will greatly inhibit and limit the re-colonization of

ground cover.

In this study, the only native species that was abundant at the end of the project

was the climbing vine, Vitis rotundifolia. In hurricane damaged tropical hardwood

forests of south Florida, management using cutting and herbicide of non-native vines

resulted in an increased recruitment of generalist native herbs, shrubs, and trees, but

not vines (Horvitz and Koop 2001). Native ferns were still present on all tree islands at

year 4, but the dominant native fern, Blechnum serrulatum, was reduced by 55-77% on

treated islands compared to pre-treatment conditions. The increase in Vitis rotundifolia

cover may have prevented further invasion of OWCF, other non-native plants such

Melaleuca quinquenervia and Schinus terebinthifolius, as well as limiting re-colonization

of native vegetation at three years post-treatment. The largest increases in Vitis

rotundifolia cover were most common over dead rachis mats of OWCF. Russell et al.

(1998) found that Dicranopteris linearis, a climbing fern native to the Old World tropics

and Polynesia, influenced forest floor light patterns, directed flora development patterns

and possibly prevents invasion of non-natives in Hawaiian rainforests. However, Vitis

rotundifolia is tardily deciduous which may create suitable conditions for germination of

OWCF spores in the winter. This aspect of Vitis rotundifolia may have also limited its

exposure to herbicides during winter treatments; which in turn, allowed it to thrive and

become the dominant plant at three years post-treatment due to increased sunlight. In

addition to the increased cover of Vitis rotundifolia, the rachis mat of dead OWCF may

have prevented the re-establishment of native plants. Rachis mats of OWCF can be >1

m thick (Pemberton and Ferriter 1998). On tree islands, many of the dead rachis mats


106









climbing 6-8 m into the canopy had a d.b.h. (diameter at 1.3 m) estimated to be >400

cm that were interconnected with other OWCF rachis and native shrubs and trees

forming a amorphous thicket. At the conclusion of the study, few native plants were

observed in areas where dead OWCF rachis mats and Vitis rotundifolia were present.

Prior to treatment, 16 native species characteristic of late succession tree islands

were documented. This number of late succession species is similar to the number

reported by on tree islands with little or no OWCF cover in LNWR (Brandt et al. 2003).

On tree islands heavily infested with OWCF, the abundance of native plants was

significantly lower (Brandt and Black 2001). After the initial aerial herbicide treatment of

OWCF, there was a decrease in ground cover of native late succession species. As

suggested by MacDougall and Turkington (2005), there are shifts from species

dominants (i.e. late succession species) to annual and perennial forbs, and graminoid

species that are functionally distinct (i.e. ruderal) in disturbed ecosystems. Annual

plants respond to the loss of canopy cover and increased sunlight, while perennial

plants take longer to recover from disturbance due to their interdependence on the

previous year's growth (Lindgren and Sullivan 2001).

All the native late succession species on tree islands are perennials, but exhibited

little recovery by the end of the study. It appears likely that native late succession

species such as ferns will require a long period of time to recover. At year 4, native

ferns were at approximately 30% of pre-treatment level, which may indicate a recovery

time of 9-10 years. Without follow-up treatment of OWCF, it will eventually spread

horizontally and vertically on the tree islands resulting in further suppression of native

vegetation. Spot treatment by ground applicators targeting OWCF re-sprouts and new


107









growth could limit non-target damage. Bimonthly and biannual ground treatments of

OWCF with triclopyr resulted in no impacts to native species cover, richness, evenness

or diversity (Stocker et al. 2008). Long term research greater is needed to determine

the recovery rate of native plant ground cover and OWCF in response to repeated

herbicide treatments.

Tree Survival and Canopy Cover

Limited tree mortality and preservation of the canopy was achieved with

metsulfuron treatments. The dominant shrub and tree species Myrica cerifera, Ilex

cassine, and Persea palustris all exhibited a high tolerance to metsulfuron. In

Everglades National Park, similar results were reported for aerial treatment with

metsulfuron with limited damage to Myrica cerifera and Persea palustris, but high

mortality to native ferns (Taylor 2006). The dramatic reduction in shrubs and trees on

islands aerially treated with glyphosate completely altered the structure and composition

of the islands. No tropical storms or hurricanes occurred during this study, so the

majority of the decrease in canopy cover on glyphosate treated islands was due to

aerial herbicide application.

Native shrub and tree species represented 22% of the ground cover on tree

islands in this study prior to treatment. This seems to indicate that the growth habit of

OWCF altered the growth form of shrubs and trees due to the twining nature of its

rachis over the limbs and boles of shrubs and trees. Most of the shrubs and trees

documented as ground cover exhibited bent or bowed limbs and boles due to near

complete entanglement with OWCF rachis. As suggested by Ugarte et al. (2006) for

large trees, the thick rachis mats of OWCF can alter growth patterns and increase the

chance of snapping and up-rooting. On tree islands with established OWCF, the growth


108









habit of OWCF with dense rachis mats climbing into the canopy likely causes a cascade

effect in which multiple large trees are toppled over. Numerous fallen trees were

observed on every tree island in this study.

The lack of invasion by non-native tree seedlings such as Melaleuca

quinquenervia and Schinus terebinthifolius at year 4 was surprising considering both

species are numerous in the refuge and surrounding areas. However, Melaleuca

quinquenervia and Schinus terebinthifolius seedlings were observed on several tree

islands and may become problematic over time, especially on tree islands in which

canopy was reduced to <20% cover. Hutchinson and Vankat (1997) reported that

canopy disturbance increased the potential for invasion of non-native plants.

Disturbance and loss of canopy trees from hurricane damage on LNWR tree islands

also increased the potential for OWCF invasion (Ugarte et al. 2006). The increased

availability of light may promote seed germination of both these invasive species. On

tree islands in LNWR that were aerially treated for Melaleuca quinquenervia with

imazapyr + glyphosate, high native tree mortality was observed and OWCF formed near

complete ground cover within 3-4 years post-treatment (Gayle Martin, LNWR, personnel

communication). Without persistent monitoring and follow-up treatments, it appears that

tree islands that have been treated for OWCF will again become dominated by invasive

plants.

The dynamics of the tree islands following herbicide treatment are complex. The

combined impacts of OWCF, hurricanes, and herbicide treatments greatly alter the

structure and composition of tree islands. Multiple disturbances have been suggested

to act in synergy to alter habitats (Hobbs and Huenneke 1992) which appears to be the


109









case with OWCF on tree islands. In addition, the size, frequency, and intensity of

disturbances may limit native plant recruitment and establishment (Hobbs and

Huenneke 1992).

Based on the results of this study, a conceptual model is proposed to describe the

effects of OWCF invasion and herbicide treatment on tree islands and emphasize the

difficulties in managing this invasive plant (Figure 4-2). Spores of OWCF are blown in

by wind currents, land in a suitable site and germinate. The twining rachis mat of

OWCF growing over native plants inhibits the growth of native vegetation suppressing

or limiting propagule production. Over time, the impacts of multiple layers of OWCF

fronds combined with thick inter-twining rachis mats greatly reduce native plant cover

and abundance. As spores are released, they fall nearby or could be carried in wind

currents to invade other sites.

When herbicide treatment occurs, OWCF cover is greatly reduced as well as much

of the remaining native ground cover. Further reductions in canopy cover result in

increased solar radiation and increased soil moisture. These conditions create suitable

sites for ruderal species, OWCF, and other non-native species to invade. Early

successional natives become the dominant plant cover, but numerous OWCF

sporelings occur throughout the tree island on tussocks, elevated soil mounds, and at

the base of dead trees. Management of OWCF is dependent on re-treatment which is

limited by access, funding, and management priorities. Providing that re-treatment of

OWCF does occur on a regular basis, it will likely take >10 years for a tree island to

return to its original state provided the canopy cover has remained relatively intact. If

there is substantial tree mortality, then recovery may take many decades.


110









In summary, the use of herbicides can effectively control OWCF on tree islands,

but results in substantial non-target damage to ground cover. Aerial applications of

metsulfuron limited shrub and tree mortality and reductions in canopy cover compared

to glyphosate treatments. Monitoring and follow-up treatments will be required for

control of OWCF on an annual basis. Without herbicide management, OWCF will

continue to form monocultures on tree islands and other habitats resulting in lower

native species richness, evenness, and canopy cover on tree islands. Follow-up

treatments will reduce OWCF growth and spore production, limiting the ability of OWCF

to infest new sites from wind blown spores. Funding limitations for long-term monitoring

and treatment along with limited accessibility make management of OWCF very difficult

in LNWR. The introduction of effective biocontrol agents is the best hope for long term

management of OWCF and maintenance of the structure and composition of tree

islands.

Management Implications

The isolated nature of OWCF infestations at LNWR make ground treatments

expensive, and with thousands of tree islands infested with OWCF, adequate funding

may not be available for follow-up treatments. The benefits of using metsulfuron for

aerial treatments of OWCF on Everglades tree islands are limited tree mortality and

preservation of canopy cover. Conversely, aerial treatments with glyphosate result in

high tree mortality and reductions in canopy cover. Both metsulfuron and glyphosate

resulted in substantial declines in ground cover. Recovery of trees and canopy cover

will require a much longer time, possibly decades, due to the constraints of secondary

growth, while ferns should recover faster due to a lack of secondary growth.


111









It is likely that OWCF cover can return to pre-treatment levels <6 years following a

single herbicide treatment. Biocontrol agents may be the best long term management

strategy, but may take decades for effective populations to become established. Until

biocontrol agents are discovered that significantly impact OWCF populations on a

landscape scale, the best management strategy for OWCF on tree islands is an initial

aerial application with metsulfuron over large infestations and annual monitoring and

follow-up ground treatments with herbicides of different modes of action.

It is not recommended that the same herbicide be used repeatedly because

OWCF exhibits characteristics of weeds that can rapidly develop herbicide resistance

(Hutchinson et al. 2007). This is especially true for the acetolactate synthase (ALS)

inhibitor herbicides, such as metsulfuron, where resistance in some weeds can occur

after several applications to the target weed (Lovell et al. 1996; Heap 1997). The use of

metsulfuron, glyphosate, and triclopyr should be used in rotation or in combinations for

control of OWCF to limit the potential for development of resistance. In addition, other

herbicides such as imazamox, imazapyr, imazapic, fluroxypyr, and aminopyralid should

be incorporated into management plans for follow-up treatments. These herbicides

have shown efficacy for control of OWCF under greenhouse conditions (Jeffrey T.

Hutchinson, unpublished data). Glyphosate, triclopyr, imazapyr, and imazamox are also

approved for use over standing water. Metsulfuron, fluroxypyr, and aminopyralid can

only be used for control of OWCF when no standing water is present. These seven

herbicides represent three modes of action that can be utilized to treat OWCF under dry

conditions or when standing water is present.


112









Table 4-1. Herbicide treatments and treatment frequency for aerial and ground herbicide application on tree islands in the
A.R.M. Loxahatchee National Wildlife Refuge.
Treatment rate


Herbicide
Metsulfuron
Metsulfuron
Metsulfuron
Metsulfuron
Glyphosate
Glyphosate
Glyphosate
Glyphosate
Untreated check
1 Spray to wet application


(kq a.i./187 I/ha)
Aerial
February 2006 (Year 1)
0.08
0.08
0.16
0.16
2.80
2.80
5.60
5.60
No treatment


(q a.i./l)
Ground 1
February-March 2007 (Year 2) February-March 2008 (Year 3)
0.03 No treatment
0.03 0.03
0.06 No treatment
0.06 0.06
5.20 No treatment
5.20 5.20
10.40 No treatment
10.40 10.40
No treatment No treatment


Table 4-2. Effects of aerial herbicide applications on Old World climbing fern ground cover one year after application
(mixed linear model ANOVA, F = 17.2, df = 4, P < 0.0001) with pre-treatment cover as the covariate.
Pre-treatment 1 year post-treatment 1
Treatment (kg a.i./ha) % Ground Cover (SE) % Ground Cover (SE)
Metsulfuron
0.08 67.3 (4.4)2 0.4 (0.1) a
0.16 62.0(3.2) 1.7 (0.7) a
Glyphosate
2.80 66.5(3.6) 7.0 (1.6) b
5.60 69.7 (2.4) 8.1 (1.3)b
Untreated check 46.0 (5.1) 56.9 (4.8) c
1 Different letters in last column indicate significant differences among treatments at P < 0.05 for aerial herbicide treatment
based on Tukey's adjusted least square means.
2 Means from 10 tree islands followed by standard error.


113








Table 4-3. Effects of aerial and a single ground treatment of herbicide on Old World climbing fern ground cover at one
year following ground treatment and two years following aerial application (mixed linear model ANOVA, F =
155.0, df = 4, P < 0.0001) with one year post treatment cover as the covariate.
1 year post-treatment 2 years post-treatment1
Treatment OWCF % Ground Cover (SE) OWCF % Ground Cover (SE)
Metsulfuron
0.08 kg a.i./ha aerial & 0.03 g a.i./l ground 0.4 (0.1)2 4.2 (0.6) a
0.16 kg a.i./ha aerial & 0.06 g a.i./l ground 1.7(0.7) 3.8 (0.9) a
Glyphosate
2.80 kg a.i./ha aerial & 5.20 g a.i./l ground 7.0 (1.6) 19.9 (2.6) b
5.60 kg a.i./ha aerial & 10.40 g a.i./l ground 8.1 (1.3) 14.3 (2.6) b
Untreated check 56.9 (4.8) 59.9 (3.0) c
1 Different letters in last column indicate significant differences among treatments at P < 0.05 for aerial herbicide treatment
based on Tukey's adjusted least square means.
2 Means of 10 tree islands followed by standard error.


Table 4-4. Effects of aerial and two annual ground treatments of herbicide on Old World climbing fern ground cover at
year 4 (repeated-measures ANCOVA, F = 100.2, df = 4, P < 0.0001) with year 1 cover as the covariate and time
(F = 303.3, df = 3, P < 0.0001).
OWCF- % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 59.8 (3.8)2 0.6 (0.2) 3.6 (1.2) 0.4 (0.1) a
Metsulfuron (0.16 & 0.06) 54.3 (3.9) 0.4(0.2) 3.5(0.9) 1.2 (0.9) a
Glyphosate (2.80 & 5.20) 66.3 (3.5) 9.0 (3.0) 16.6 (2.8) 5.7 (1.2) b
Glyphosate (5.60 & 10.40) 72.5 (2.1) 9.0 (1.7) 17.5 (4.6) 9.2 (2.1) b
Untreated check 46.0 (5.1) 57.0 (4.8) 60.0 (3.0) 65.0 (4.2) c
Time 1 a c b c
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.








Table 4-5. Effects of aerial and two annual ground treatments of herbicide on Blechnum serrulatum (swamp fern) ground
cover at year 4 (repeated-measures ANCOVA, F = 9.4, df = 4, P = 0.0002) with year 1 cover as the covariate
and time (F = 63.5, df = 3, P < 0.0001).
Blechnum serrulatum % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 17.9 (3.7)2 2.7(1.1) 9.1 (3.5) 8.0 (6.1) a
Metsulfuron (0.16 & 0.06) 19.5 (4.9) 0.9 (0.3) 7.0 (1.7) 7.0 (1.6) a
Glyphosate (2.80 & 5.20) 33.4 (7.7) 7.6(1.3) 9.3(2.8) 8.2 (2.0) a
Glyphosate (5.60 & 10.40) 24.5 (4.2) 1.8 (0.6) 3.0 (1.5) 5.4 (3.5) a
Untreated check 33.8 (5.5) 33.3 (6.2) 33.3 (7.2) 32.2 (6.5) b
Time 1 a b c b
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


Table 4-6. Effects of aerial and two annual ground treatments of herbicide on Smilax laurifolia (green brier) ground cover
at year 4 (repeated-measures ANCOVA, F = 12.5, df = 4, P < 0.0001) with year 1 cover as the covariate and
time (F = 149.1, df = 3, P < 0.0001).
Smilax laurifolia % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 8.4 (3.5)2 0.7 (0.2) 0.1 (0.1) 0.4 (0.2) a
Metsulfuron (0.16 & 0.06) 11.8 (3.2) 0.4 (0.2) 0.3 (0.2) 0.1 (0.1) a
Glyphosate (2.80 & 5.20) 17.2 (2.9) 0.3 (0.2) 0.2 (0.1) 1.8 (0.9) a
Glyphosate (5.60 & 10.40) 18.3 (2.6) 0.4 (0.1) 0.1 (0.1) 0.3 (0.1) a
Untreated check 12.0(1.2) 6.8(1.3) 5.3(1.1) 8.2 (1.0) b
Time 1 a b c b
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


115








Table 4-7. Effects of aerial and two annual ground treatments of herbicide on Osmunda cinnamomea (cinnamon fern)
ground cover at year 4 (repeated-measures ANCOVA F = 6.2, df = 4, P = 0.0020) with year 1 cover as the
covariate and time (F = 12.0, df = 3, P < 0.0001).
Osmunda cinnamomea % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 1.0 (0.6)2 0.2 (0.1) 1.2 (0.5) 0.4 (0.3) a
Metsulfuron (0.16 & 0.06) 0.3 (0.1) 0.2 (0.1) 1.2 (0.3) 1.0 (0.7) a
Glyphosate (2.80 & 5.20) 1.0 (0.4) 2.2 (0.7) 3.3 (0.7) 3.9 (1.0) b
Glyphosate (5.60 & 10.40) 1.6 (0.5) 5.0 (1.2) 4.3 (1.2) 5.5 (1.2) b
Untreated check 1.3(0.7) 7.5(2.1) 5.5(2.4) 5.6 (1.8) b
Time 1 a b b b
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


Table 4-8. Effects of aerial and two annual ground treatments of herbicide on Woodwardia virginica (Virginia chain fern)
ground cover at year 4 (repeated-measures ANCOVA, F = 4.0, df = 4, P = 0.0159) with year 1 cover as the
covariate and time (F = 0.9, df = 3, P = 0.4168).
Woodwardia virginica % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 0.6 (0.2)2 0.1 (0.1) 0.5 (0.3) 0.2 (0.1) a
Metsulfuron (0.16 & 0.06) 0.8 (0.2) 0.3 (0.1) 1.1 (0.3) 0.8 (0.2) a, b
Glyphosate (2.80 & 5.20) 0.2(0.1) 2.4(0.8) 1.1 (0.5) 2.3(1.2) b
Glyphosate (5.60 & 10.40) 1.2 (0.4) 2.4 (0.7) 0.7 (0.3) 2.4 (0.9) b
Untreated check 1.5 (0.6) 0.6 (0.2) 0.3 (0.2) 0.5 (0.2) a
Time 1 NS NS NS NS
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


116








Table 4-9. Effects of aerial and two annual ground treatments of herbicide on Vitis rotundifolia (muscadine) ground cover
at year 4 (repeated-measures ANCOVA, F = 6.0, df = 4, P = 0.0024) with year 1 as the covariate and time (F =
41.2, df = 3, P < 0.0001).
Vitis rotundifolia % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 3.4 (0.8)2 25.2 (5.9) 8.7 (1.9) 13.0 (4.0) a
Metsulfuron (0.16 & 0.06) 8.4 (2.6) 29.0 (6.6) 12.7 (3.8) 29.4 (8.7) a, b
Glyphosate (2.80 & 5.20) 10.2 (2.9) 16.9 (4.7) 42.7 (8.3) 59.4 (4.1) b
Glyphosate (5.60 & 10.40) 10.1 (2.0) 19.3 (5.5) 37.5 (8.2) 45.8 (7.2) b
Untreated check 7.2(2.9) 2.1 (0.8) 7.2(2.4) 10.6 (3.5) a
Time 1 a b b c
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


Table 4-10. Effects of aerial and two annual ground treatments of herbicide on Osmunda regalis (royal fern) ground cover
at year 4 (repeated-measures ANCOVA, F = 3.8, df = 4, P = 0.0195) with year 1 as the covariate and time (F =
9.5, df = 3, P < 0.0001).
Osmunda regalis % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 0.6 (0.4)2 0.0 (0) 0.1 (0.1) 0.0 (0) a
Metsulfuron (0.16 & 0.06) 1.1 (0.7) 0.0(0) 0.0(0) 0.0 (0)a
Glyphosate (2.80 & 5.20) 1.2 (0.6) 0.7 (0.2) 0.4 (0.2) 0.2 (0.1) b
Glyphosate (5.60 & 10.40) 2.4(1.1) 0.4 (0.2) 0.2 (0.1) 0.1 (0.1)b
Untreated check 0.4 (0.1) 0.7 (0.3) 0.5 (0.2) 0.5 (0.2)b
Time 1 a b b b
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.








Table 4-11. Effects of aerial and two annual ground treatments of herbicide on Cladiumjamaicense (sawgrass) ground
cover at year 4 (repeated-measures ANCOVA, F = 1.8, df = 4, P = 0.1749) with year 1 cover as the covariate
and time (F = 2.2, df = 3, P = 0.0943).
Cladium jamaicense % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 0.1 (0.1)2 0.1 (0.1) 0.4 (0.2) 0.4 (0.3)
Metsulfuron (0.16 & 0.06) 0.9 (0.4) 1.7 (0.9) 2.4 (1.3) 3.4 (2.1)
Glyphosate (2.80 & 5.20) 0.3 (0.1) 0.0 (0) 0.1 (0.1) 0.0 (0)
Glyphosate (5.60 & 10.40) 0.9 (0.4) 0.0 (0) 0.0 (0) 0.0 (0)
Untreated check 0.9(0.5) 0.8(0.5) 1.2(0.8) 1.1 (0.7)
Time 1 NS NS NS NS
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


Table 4-12. Effects of aerial and two annual ground treatments of herbicide on Myrica cerifera (wax myrtle) ground cover
at year 4 (repeated-measures ANCOVA, F = 10.5, df = 4, P < 0.0001) with year 1 cover as the covariate and
time (F = 10.5, df= 3, P < 0.0001).
Myrica cerifera % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 3.2 (0.6)2 3.9(1.5) 4.8(1.6) 2.8 (0.3)b
Metsulfuron (0.16 & 0.06) 3.4 (1.0) 1.9 (0.9) 3.8 (1.0) 6.2 (0.8) b
Glyphosate (2.80 & 5.20) 2.6 (0.6) 0.1 (0.1) 0.3 (0.1) 0.9 (0.3) a
Glyphosate (5.60 & 10.40) 2.2 (0.6) 0.2 (0.1) 0.4 (0.2) 0.7 (0.2) a
Untreated check 3.4(1.4) 3.9(1.3) 2.9(0.6) 6.4 (1.8) b
Time 1 a b b a
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


118








Table 4-13. Effects of aerial and two annual ground treatments of herbicide on Persea palustris (swamp bay) ground
cover at year 4 (repeated-measures ANCOVA, F = 8.1, df = 4, P = 0.0005) with year 1 cover as the covariate
and time (F = 15.4, df = 3, P < 0.0001).
Persea palustris % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 6.6 (1.7)2 9.0 (2.7) 9.2 (2.4) 3.4 (1.7) b
Metsulfuron (0.16 & 0.06) 11.2 (2.3) 9.4 (1.3) 11.4 (1.9) 10.1 (2.2) c
Glyphosate (2.80 & 5.20) 8.7 (0.7) 1.2 (0.7) 0.7 (0.5) 0.5 (0.2) a
Glyphosate (5.60 & 10.40) 11.4 (4.6) 0.9 (0.5) 1.6 (0.4) 3.3 (1.1) b
Untreated check 10.8(3.4) 8.6(1.4) 8.8(2.4) 6.8 (1.0) c
Time 1 a b b b
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


Table 4-14. Effects of aerial and two annual ground treatments of herbicide on Ilex cassine (Dahoon holly) ground cover
at year 4 (repeated-measures ANCOVA, F = 6.4, df = 4, P = 0.0018) with year 1 cover as the covariate and time
(F = 28.2, df = 3, P < 0.0001).
Ilex cassine % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 6.0 (1.9)2 2.1 (0.9) 1.0(0.5) 1.0 (0.6) a
Metsulfuron (0.16 & 0.06) 10.2 (2.0) 2.5 (0.8) 2.6 (0.5) 3.5 (1.9) a,b
Glyphosate (2.80 & 5.20) 11.3 (4.0) 1.2 (0.5) 0.8 (0.2) 1.8 (0.5) a,b
Glyphosate (5.60 & 10.40) 3.3 (1.9) 0.2 (0.1) 0.3 (0.2) 0.2 (0.1) a
Untreated check 6.5(1.8) 5.6(1.4) 3.3(0.8) 6.3 (1.4) c
Time 1 a b b b
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


119








Table 4-15. Effects of aerial and two annual ground treatments of herbicide on Rapanea punctata (myrsine) ground cover
at year 4 (repeated-measures ANCOVA, F = 1.0, df = 4, P = 0.4311) with year 1 cover as the covariate and time
(F = 4.2, df = 3, P = 0.0088).
Rapanea punctata % ground cover (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 0.8 (0.6)2 0.0 (0) 0.0 (0) 0.0 (0)
Metsulfuron (0.16 & 0.06) 0.2 (0.1) 0.0 (0) 0.0 (0) 0.0 (0)
Glyphosate (2.80 & 5.20) 1.0 (0.5) 0.5 (0.2) 0.4 (0.3) 1.4 (1.1)
Glyphosate (5.60 & 10.40) 1.6(1.1) 0.1 (0.1) 0.9 (0.5) 1.1 (0.7)
Untreated check 1.2 (0.9) 0.8 (0.5) 0.9 (0.5) 0.8 (0.4)
Time 1 a b b b
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


Table 4-16. Effects of aerial and two annual ground treatments of herbicide on tree canopy cover at year 4 (repeated-
measures ANCOVA, F = 7.5, df = 4, P < 0.0001) with year canopy cover as the covariate and time (F = 13.8; df
= 3, P < 0.0001).
% Canopy (SE)
Treatment Evaluation Period
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 41
Metsulfuron (0.08 & 0.03) 51 (9)2 51 (6) 44 (11) 31 (6) b
Metsulfuron (0.16 & 0.06) 44 (8) 52 (8) 43 (8) 40 (6) b
Glyphosate (2.80 & 5.20) 55 (5) 44 (4) 21 (5) 7 (3) a
Glyphosate (5.60 & 10.40) 42 (4) 33 (6) 13(3) 16 (4) a
Untreated check 69 (5) 71 (4) 65 (8) 58 (7) c
Time 1 a a b c
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


120








Table 4-17. Survival rates (%) of the four most common shrub and tree species (l/ex cassine, Persea palustris, Myrica
cerifera and Rapanea punctata) at three years post aerial herbicide treatment. Herbicide rates represent aerial
treatment.
% Tree survival (SE)
Three years post-trt
Herbicide (kg a.i.) Ilex cassine Persea palustris Myrica cerifera Rapanea punctata
Metsulfuron 0.08 80 (7)2 77 (6)b 65 (7)b 12 (8) a
Metsulfuron 0.16 89 (4) 93 (3)b, c 81 (4)ab 18 (8) a
Glyphosate 2.80 75(6) 6 (3)a 18 (6) 88 (13) b,c
Glyphosate 5.60 70(7) 20 (6)a 17 (5) 58 (10) b
Untreated check 100 (0) 100 (0) 0 93 (5) a 100 (0) C
1 Different letters in columns for tree species indicates significant difference at P < 0.05 for the three year evaluation
period based on logistic regression analysis.
2 Means of 50 Ilex cassine, 60 Persea palustris, 60 Myrica cerifera, and 20 Rapanea punctata replications per treatment
followed by standard error.


Table 4-18. Summary statistics for logistic regression analysis of tree variables as model predictors of Ilex cassine
(Dahoon holly) survival following herbicide treatment.
Parameters DF Wald Chi-Square P-value1
Overall model 7 33.18 < 0.0001
Herbicide 4 7.87 0.0963
Height 1 6.83 0.0090
D.B.H. 1 0.03 0.8563
Foliage 1 0.09 0.7655
Herbicide*Height 4 3.56 0.4685
Herbicide*DBH 4 3.96 0.4112
Herbicide*Foliage 4 6.93 0.1395
1 Significance level of P < 0.05 for tests of variables and their interaction.









Table 4-19. Summary statistics for logistic regression analysis of tree variables as model predictors of Persea palustris
(swamp bay) survival following herbicide treatment.
Parameters DF Wald Chi-Square P-value 1
Overall model 7 81.97 < 0.0001
Herbicide 4 75.31 < 0.0001
Height 1 0.01 0.9410
D.B.H. 1 0.72 0.3955
Foliage 1 0.28 0.5971
Herbicide*Height 4 5.67 0.2255
Herbicide*DBH 4 5.26 0.2613
Herbicide*Foliage 4 10.39 0.0344
1 Significance level of P < 0.05 for tests of variables and their interaction.


Table 4-20. Summary statistics for logistic regression analysis of tree variables as model predictors of Myrica cerifera
(wax myrtle) survival following herbicide treatment.
Parameters DF Wald Chi-Square P-value1
Overall model 7 75.91 < 0.0001
Herbicide 4 61.25 < 0.0001
Height 1 0.73 0.3919
D.B.H. 1 0.09 0.7596
Foliage 1 5.27 0.0217
Herbicide*Height 4 25.09 < 0.0001
Herbicide*DBH 4 24.82 < 0.0001
Herbicide*Foliage 4 30.70 < 0.0001
1 Significance level of P < 0.05 for tests of variables and their interaction.


122









Table 4-21. Summary statistics for logistic regression analysis of tree variables as model predictors of Rapanea punctata
(myrsine) survival following herbicide treatment.
Parameters DF Wald Chi-Square P-value 1
Overall model 7 16.98 0.0175
Herbicide 4 12.85 0.0120
Height 1 2.20 0.1377
D.B.H. 1 0.38 0.5402
Foliage 1 0.12 0.7322
Herbicide*Height 4 7.61 0.1069
Herbicide*DBH 4 8.34 0.0798
Herbicide*Foliage 4 4.46 0.3469
1 Significance level of P < 0.05 for tests of variables and their interaction.


123









Table 4-22. Total number of late succession, ruderal, non-native (including OWCF) and
all plant species recorded pre-treatment (year 0) to year three post treatment
for metsulfuron and glyphosate treated tree islands and untreated checks.
Herbicide rates in kg represent aerial treatment (year 0) and g represent
ground treatments (year 1 and 2).
Total number of Species
Late
Herbicide Year Succession Ruderal Exotic Total
Metsulfuron (0.08 kg & 0.03 g a.i.)
1 15 2 3 20
2 13 4 3 20
3 14 13 3 30
4 12 13 3 28
Metsulfuron (0.16 kg & 0.06 g a.i.)
1 14 4 2 20
2 13 6 1 20
3 15 13 2 30
4 13 16 2 31
Glyphosate (2.80 kg & 5.20 g a.i.)
1 14 2 1 17
2 12 3 1 16
3 14 14 3 31
4 14 20 4 38
Glyphosate (5.60 kg & 10.40 g a.i.)
1 14 6 1 21
2 14 7 2 23
3 13 15 3 31
4 14 24 4 42
Untreated check
1 14 0 1 15
2 12 2 1 15
3 14 4 1 19
4 13 2 1 16


124








Table 4-23. Effects of aerial and two annual ground treatments of herbicide on total species richness (including non-
native species) at year 4 (repeated-measures ANCOVA, F = 8.2, df = 4, P < 0.0001) with year 1 as the
covariate and time (F = 24.6, df = 3, P < 0.0001).
Treatment Total species richness (SE)
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4
Metsulfuron (0.08 & 0.03) 8.6 (0.4)2 6.5 (0.4) 8.1 (0.6) 8.1 (0.6) a
Metsulfuron (0.16 & 0.06) 8.9 (0.3) 5.9 (0.4) 9.5 (0.7) 9.2 (0.4) a,b
Glyphosate (2.80 & 5.20) 8.3 (0.4) 6.9 (0.5) 9.2 (0.9) 11.9 (0.7) b
Glyphosate (5.60 & 10.40) 8.9 (0.6) 8.1 (0.6) 10.3 (0.9) 14.0 (1.3) c
Untreated check 8.8 (0.4) 8.4 (0.3) 9.9 (0.3) 9.4 (0.6) a,b
Time 1 a b a c
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


Table 4-24. Effects of aerial and two annual ground treatments of herbicide on late succession species richness at year 4
(repeated-measures ANCOVA, F = 6.9, df = 4, P < 0.0001) with year 1 as the covariate and time (F = 10.6, df =
3, P < 0.0001).
Treatment Late succession species richness (SE)
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4
Metsulfuron (0.08 & 0.03) 7.4 (0.3)2 5.7 (0.3) 6.0 (0.4) 5.6 (0.3) a
Metsulfuron (0.16 & 0.06) 7.6 (0.2) 5.1 (0.4) 7.1 (0.4) 6.7 (0.3) b
Glyphosate (2.80 & 5.20) 7.3 (0.4) 5.7 (0.4) 6.0 (0.7) 7.1 (0.4) b,c
Glyphosate (5.60 & 10.40) 7.7 (0.6) 5.7 (0.6) 5.7 (0.7) 7.0 (0.7) b,c
Untreated check 7.8 (0.4) 7.1 (0.5) 8.4 (0.3) 8.1 (0.6) c
Time 1 a b c d
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


125








Table 4-25. Effects of aerial and two annual ground treatments of herbicide on ruderal species richness at year 4
(repeated-measures ANCOVA, F = 19.9, df = 4, P < 0.0001) with year 1 as the covariate and time (F = 70.5, df
= 3, P < 0.0001).
Treatment Ruderal species richness (SE)
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4
Metsulfuron (0.08 & 0.03) 0.1 (0.1)2 0.3 (0.1) 1.1 (0.2) 1.8 (0.2) a
Metsulfuron (0.16 & 0.06) 0.1 (0.1) 0.4(0.1) 1.3(0.3) 1.8 (0.3) a
Glyphosate (2.80 & 5.20) 0.0 (0.0) 0.3 (0.1) 2.1 (0.4) 3.7 (0.6) a,b
Glyphosate (5.60 & 10.40) 0.3 (0.1) 1.3 (0.4) 3.2 (0.9) 5.6 (0.9) b
Untreated check 0.0 (0.0) 0.2 (0.1) 0.6 (0.3) 0.3 (0.1) c
Time 1 a a b c
1 Different letters in last column and row indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


Table 4-26. Effects of aerial and two annual ground treatments of herbicide on non-native species (other than Old World
climbing fern richness at year 4 (repeated-measures ANCOVA, F = 5.4, df = 4, P = 0.0007) with year 1 as the
covariate and time (F = 3.9, df = 3, P = 0.0107).
Treatment Non-native species richness (SE)
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4
Metsulfuron (0.08 & 0.03) 0.1 (0.1)2 0.1 (0.1) 0.0 (0.0) 0.1 (0.1)a
Metsulfuron (0.16 & 0.06) 0.3 (0.1) 0.0 (0.0) 0.0 (0.0) 0.1 (0.1)a
Glyphosate (2.80 & 5.20) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.2 (0.1)a
Glyphosate (5.60 & 10.40) 0.0 (0.0) 0.1 (0.1) 0.4 (0.1) 0.4 (0.1)b
Untreated check 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0)a
Time 1 a a a b
1 Different letters in last column and rows indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.


126








Table 4-27. Effects of aerial and two annual ground treatments of herbicide on species evenness at year 4 (repeated-
measures ANCOVA, F = 5.0, df = 4, P = 0.0012) with year 1 as the covariate and time (F = 7.7, df = 3, P <
0.0001).
Treatment Species evenness (SE)
(aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4
Metsulfuron (0.08 & 0.03) 0.34 (0.03)2 0.44 (0.05) 0.49 (0.04) 0.44 (0.04) b
Metsulfuron (0.16 & 0.06) 0.40 (0.03) 0.44 (0.01) 0.49 (0.04) 0.39 (0.03) b
Glyphosate (2.80 & 5.20) 0.42 (0.03) 0.49 (0.05) 0.32 (0.05) 0.21 (0.02) a
Glyphosate (5.60 & 10.40) 0.38 (0.02) 0.45 (0.05) 0.30 (0.03) 0.27 (0.03) a
Untreated check 0.41 (0.03) 0.36 (0.01) 0.29 (0.02) 0.37 (0.04) b
Time 1 a b a c
1 Different letters in last column and rows indicate significant differences among treatments and time, respectively, at P <
0.05 for the three year evaluation period based on Tukey's adjusted least square means.
2 Means of 5 tree islands followed by standard error.































- Ground cover
= Canopy cover
= Tagged trees


Figure 4-1. Sampling design used to evaluate responses of ground cover (5 1.5 m), canopy cover (> 1.5 m), and tagged
trees on northern Everglade's tree islands treated with herbicides to control Old World climbing fern.




128














Tree
Island
I. ___


Additional
Management
problems


Loss of around and canopy cover
Reduction of OWCF
Reduction in native plant abundance
Increase in OWCF spore Qermination
Increase in solar radiation
Increase in soil moisture
Increase in ruderal species
Other non-native invade
Limited recruitment of native plants



Formation of thick OWCF rachis mat
Decrease in native plant cover/! abundance
Loss of canopy cover from tree fall
Decrease in solar radiation
Decrease in soil moisture
Decrease in native plant propaqules
Limited recruitment of native plants
Increased chance of fire


Herbicide


Altered structure
k and composition
of tree island


Figure 4-2. Conceptual model on the effects of OWCF on tree islands and sequence of events following herbicide
treatment.


129









CHAPTER 5
SUMMARY

The invasive potential of OWCF may redefine how "invasive potential" is defined

and may have resulted in a perpetual management problem in natural areas of Florida.

The spread of most invasive plants of Florida's natural areas can be linked to a long

history of various activities such as ornamental use, forage production or erosion

control. There is no clear documentation on how OWCF was introduced into Florida

(Nauman and Austin 1978; Pemberton and Ferriter 1998). The earliest documented

occurrences of live OWCF in Florida are from two herbarium reports from 1958

indicating the species was present in nurseries. It maybe that OWCF's presence in

Florida is due to just a few unintentional introductions. After 1958, it was not noted

again for a decade until it was discovered invading natural areas. Since that time, the

speed of invasion into natural wetlands has made OWCF one of the top two or three

most invasive plants in natural areas of Florida. The management of OWCF is difficult

because of the numerous spores it produces, its ability to colonize and thrive in remote

sites, and its extensive range in central and southern Florida. This study examined

various aspects of OWCF management during all phases of its life cycle using

herbicides.

Translocation studies using 14C radiolabeled herbicides indicated that limited

movement of glyphosate, metsulfuron, and triclopyr occurred in the rhizomes of OWCF.

In the 14C study, herbicides moved basipitally in the rachis, but movement of the

herbicides stopped at the rhizome and did move vertically through the rhizomes. The

key to successful control of OWCF is the elimination of the rhizomes because new

fronds emerge at irregular intervals on the rhizome. In established populations of


130









OWCF, the number of new fronds that arise from rhizomes may exceed several

hundred per plant. Complete foliar coverage with herbicide is required for successful

control of OWCF. This may be difficult in dense infestations of OWCF during the initial

treatment due to over-lapping fronds of OWCF that are inter-twined with native plants.

A follow-up treatment will be required within six months after the initial treatment for

adequate control. Occurrences of OWCF, less than a few square meters in size, may

be eliminated with a single herbicide treatment, but larger infestations will require

multiple treatments.

The spores of OWCF were highly susceptible to metsulfuron, but exhibited

tolerance to imazapyr, glyphosate, fluroxypyr, asulam, and triclopyr exposed in bathing

solution. It is unclear if metsulfuron applied to mature OWCF sporophytes will inhibit

spore development in the sori. Results of this study also indicated that OWCF

gametophytes are highly susceptible to metsulfuron. However, gametophytes exhibited

tolerance to metsulfuron at concentrations up to 0.16 kg a.i. and developed into

sporophytes at concentrations 5 0.04 kg a.i. These results indicate that OWCF exhibits

the potential for resistance to metsulfuron especially at lower concentrations.

Metsulfuron also exhibited residual activity on spore germination and growth in agar,

indicating it may provide longer control of OWCF.

Aerial and ground application of metsulfuron over Everglades tree islands was

more effective in reducing OWCF cover than aerial application of glyphosate. Both

metsulfuron and glyphosate treatments resulted in significant loss of native ground

cover. Evergreen trees and shrubs exhibited a high tolerance to metsulfuron, and

minimal impacts to the canopy cover were observed on tree islands treated with


131









metsulfuron. Aerial treatment with glyphosate on tree islands resulted in significant tree

mortality and loss of canopy cover. The dominant deciduous trees on tree islands,

Cephalanthus occidentalis and Salix caroliniana, suffered no damage to winter

application of either metsulfuron or glyphosate. It is recommended that aerial

application of metsulfuron be applied initially for control of OWCF on tree islands. After

the initial aerial treatment, additional aerial treatments are not recommended for at least

three years to prevent further non-target impacts. Follow-up ground treatments will be

required for OWCF re-growth, un-treated areas of OWCF missed during the initial

treatment, and numerous OWCF sporophytes that develop from spores following

treatment. Re-treatment of OWCF on tree islands by ground crews is recommended at

one year post aerial treatment and in future years.

Management Recommendations

1. Aerial treatment of OWCF on Everglades tree islands should be carried out using
metsulfuron.

2. Ground treatment of OWCF should be performed by rotating metsulfuron,
glyphosate, triclopyr and other herbicides to limit the potential for populations
becoming resistant.

3. Complete spray coverage of OWCF is essential to maximize herbicide
translocation into the rhizomes and prevent re-sprouting.

4. Re-treatment and long term monitoring will be required for adequate control of
OWCF.

5. Monitoring and re-treatment should be carried out 5 one year following the initial
treatment, but preferably at six months.

6. Aerial treatment of OWCF with metsulfuron should be considered for any
infestations of OWCF on Everglades tree islands that can be observed from the
air.

7. Sites where threatened and endangered plants and animals are known to exist will
require ground treatments. These sites should receive high funding priority.


132









8. Primary funding for control of OWCF should be consolidated to geographical
sections of Florida to limit the spread of spores in the areas where it occurs at the
highest densities. Currently, the Treasure Coast region in southeast Florida
should receive the highest funding for OWCF control. This geographical area
includes Martin, Palm Beach, and St. Lucie Counties. As OWCF coverage is
reduced in these counties, funding should be directed to the next geographical
area with the largest infestations of OWCF.

9. Secondary funding for control of OWCF should be committed for expanding
populations at the southern and northern ranges of its distribution in Florida. Dade
and Monroe counties at the southern range of OWCF's distribution should receive
priority funding for OWCF control since more endangered habitat and listed
species occur in these counties relative to northern counties.

10. Funding should be earmarked for monitoring and re-treatment of all OWCF sites
that receive initial treatments.



Additional Research

1. Evaluate the translocation patterns for the maximum herbicide rates of glyphosate,
metsulfuron, and triclopyr in OWCF.

2. Evaluate the efficacy of treatments using rotations of herbicides with different
modes of action in field trials.

3. Examine the viability of OWCF spores from sporophytes with mature fertile leaflets
treated with metsulfuron and other herbicides.

4. Evaluate the residual activity of metsulfuron and other herbicides on OWCF spore
germination and gametophyte survival using native soils.

5. Compare new sporophyte development at sites treated with metsulfuron and other
herbicides.

6. Compare the efficacy of herbicides on varying plant sizes of OWCF.

7. Evaluate the efficacy of aerial application of 0.04 kg a.i./187 I/ha of metsulfuron for
control of OWCF on tree islands and its effect on non-target vegetation.

8. Determine the time frame needed between the initial aerial application and a
second aerial application when metsulfuron can be applied over tree islands
without further reductions in ground and canopy cover and structural changes to
tree islands.

9. Compare the efficacy of aerially applied glyphosate and metsulfuron for control of
OWCF in deciduous cypress and bayhead swamp habitat.


133









10. Evaluate the effectiveness of aerial application for spot-treatment of OWCF on
Everglade's tree islands and other habitats.

11. Evaluate the potential of planting native plants in selected instances to reestablish
native plant diversity and competition with OWCF.

12. Evaluate triclopyr for aerial treatment of OWCF.

13. Continue testing novel or untested herbicides for efficacy on OWCF.


134








APPENDIX
PLANT SPECIES DOCUMENTED ON TREE ISLANDS AT A.R.M. LOXAHATCHEE
N.W.R. TREATED WITH METSULFRUON (0.08 AND 0.16 KG A.I.) AND
GLYPHOSATE (2.80 AND 5.60 KG A.I.) AT PRE-TREATMENT (YEAR 1) TO THREE
YEARS POST-TREATMENT. PRESENCE OF PLANTS INDICATED BY ASTERISK
(*), CLIMAX SPECIES (+), AND NON-NATIVE PLANTS (A). ALL OTHER PLANTS
ARE CLASSIFIED AS RUDERAL SPECIES.


135









Appendix


Plant Species
Andropogon virginicus L.



Asclepias incarnate L.



Bidens laevis (L.)



+ Blechnum serrulatum Rich.



Boehmeria cylindrical (L.) Sw.



+ Cephalanthus occidentalis L.



Ceratopteris pteridoides (Hook.) H ieron.


Year
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4


Metsulfuron
0.06 kg 0.12 kg


Glyphosate
3.0 kg 6.0 kg


136


Control
Untreated











Appendix cont.


Plant Species
+ Cladiumjamaicense Crantz


Year
1


Metsulfuron
0.06 kg 0.12 kg
*


Glyphosate
3.0 kg 6.0 kg
*


Crinum americanum L.


Cyperus haspan L.



Drosera brevifolia Pursh



A Emilia fosbergii Nicolson


Erechtites hieracifolia (L.) Raf. Ex DC.



Eriocaulon compressum Lam.


Control
Untreated
*










Appendix cont.


Plant Species
Eupatorium capillifolium (Lam.) Small



Galium tinctorium L.



Habenaria floribunda Lindl.



Hypericum fasciculatum Lam.


Year
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4


+ Ilex cassine L.


Lachnanthes caroliana (Lam.) Dandy


A Ludwigia peruviana (L.) H. Hara


Metsulfuron
0.06 kg 0.12 kg


*


Glyphosate
3.0 kg 6.0 kg


138


Control
Untreated












Appendix cont.


Plant Species
Ludwigia repens J.R. Frost




A Lygodium microphyllum (Cav.) R. Br.




A Melaleuca quinquenervia (Cav.) S.T. Blake




Mikania scandens (L.) Willd.




+ Myrica cerifera L.




+ Nephrolepis biserrata (Sw.) Schott




+ Osmunda cinnamomea L.


Year
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4


Metsulfuron
0.06kg 0.12kg
*
*
*
*
*
*


Glyphosate
3.0 kg 6.0 kg
*
*
*
*
*
*
*
*


139


Control
Untreated

*
*

*
*
*
*











Appendix cont.


Plant Species
+ Osmunda regalis L.




Panicum hemitomon Schult.




Parthenocissus quinquefolia (L.) Planch.




+ Peltandra virginica (L.) Schott & Endl.




+ Persea palustris (Raf.) Sarg.




Phytolacca americana L.




Pluchea rosea R. K. Godfrey


Year
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4


Metsulfuron
0.06 kg 0.12 kg
*
*
*
*


Glyphosate
3.0 kg 6.0 kg
*
*
*
*
*
*


140


Control
Untreated
*
*
*
*










Appendix cont.


Plant Species
Polygonum densiflorum Meisn.



Pontedaria cordata L.



+ Rapanea punctata (Lam.) Lundell



Rhus copallinum L.



Rhynchospora inundata (Oakes) Fernald



Rhynchospora microcarpa Baldwin ex A. Gray



Rhynchospora tracyi Britton


Year
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4


Metsulfuron
0.06 kg 0.12 kg
*


Glyphosate
3.0 kg 6.0 kg


Control
Untreated










Appendix cont.


Plant Species
Rhynchospora chalarocephala Fernald & Gale



Saccharum giganteum (Walter) Pers.



+ Salix caroliniana Michx.



Sarcostemma clausum (Jacq.) Roem. & Schult.



A Schinus terebinthifolius Raddi



Scoparia dulcis L.



+ Smilax laurifolia L.


Year
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4


Metsulfuron
0.06 kg 0.12 kg


Glyphosate
3.0 kg 6.0 kg


142


Control
Untreated











Appendix cont.


Plant Species
+ Thelypteris interrupta (Willd.) K. Iwats


Year
1


Metsulfuron
0.06 kg 0.12 kg
*


Glyphosate
3.0 kg 6.0 kg


Triadenum virginicum (L.) Raf.


Utricularia spp.



+ Vitis rotundifolia Michx.


+ Woodwardia virginica (L.) Sm.



Xyris ambigua Beyr. ex Kunth


143


Control
Untreated
*

*









LIST OF REFERENCES


Anonymous. 2007. Accelerated invasive species control plan for A.R.M. Loxahatchee
National Wildlife Refuge: 2007-2008. Unpublished report, South Florida Water
Management District, West Palm Beach, Florida

Ballesteros, D. and C. Walters. 2007. Water properties in fern spores: sorption
characteristics relating to water affinity, glassy states, and storage stability.
Journal of Experimental Botany 58:1185-1196.

Banks, J. A. 1999. Gametophyte development in ferns. Annual Review of Plant
Physiology and Plant Molecular Biology 50:163-186.

Barrett, M. A., L. A. Brandt, and B. Thomas, Jr. 2006. Monitoring ground treatments of
Old World climbing fern (Lygodium microphyllum) on the Arthur R. Marshall
Loxahatchee National Wildlife Refuge: a follow-up report. Wildland Weeds 10:4-
7.

Beckie, H. J., I. M. Heap, R. J. Smeda, and L. M. Hall. 2000. Screening forherbicide
resistance in weeds. Weed Technology 14:428-445.

Beckner, J. 1968. Lygodium microphyllum, another fern escaped in Florida. American
Fern Journal 58:93-94.

Biehl, T. 2004. Lygodium infestations in the Lake Wales Ridge. Wildland Weeds 7:4-5.

Boughton, A. J. and R. W. Pemberton. 2009. Establishment of an imported natural
enemy, Neomusotima conspurcatalis (Lepidoptera; Crambidae) against an
invasive weed, Old World climbing fern, Lygodium microphyllum, in Florida.
Biocontrol Science and Technology 19: 769-772.

Brandt, L. A. 2004. Effectiveness of different aerial spray for control of Lygodium
microphyllum on tree islands in the A.R.M. Loxahatchee NWR 24 month
monitoring period. Unpublished report, USFWS, A.R.M. Loxahatchee NWR,
Boynton Beach, Florida.

Brandt, L. A. and D. W. Black. 2001. Impacts of the introduced fern, Lygodium
microphyllum, on the native vegetation of tree islands in the Arthur R. Marshall
Loxahatchee National Wildlife Refuge. Florida Scientist 64:191-196.

Brandt, L. A., K. M. Portier and W. M. Kitchens. 2000. Patterns of change in tree
islands in the Arthur R. Marshall Loxahatchee National Wildlife Refuge from 1950
to 1991. Wetlands 20:1-14.

Brandt, L. A., D. Ecker, I. G. Rivera, A. Traut and F. J. Mazzotti. 2003. Wildlife and
vegetation of bayhead islands in the A.R.M. Loxahatchee National Wildlife
Refuge. Southeastern Naturalist 2:179-194.


144









Brown, V. M. 1984. A biosystematic study of the fern genus Lygodium in eastern North
America. M. S. Thesis. Biology Department, University of Central Florida.
Orlando, Florida.

Call, E. M., L. A. Brandt and D. L. Deangelis. 2007. Old World climbing fern (Lygodium
microphyllum) spore germination on natural substrates. Florida Scientist 70:55-
61.

Canfield, R. H. 1941. Application of the line interception method in sampling range
vegetation. Journal of Forestry 39:388-394.

Carroll, E. W., 0. J. Schwarz, and L. G. Hickok. 1988. Biochemical studies of
paraquat-tolerant mutants of the fern Ceratopteris richardii. Plant Physiology
87:651-654.

Chachalis, D. and K. N. Reddy. 2004. Pelargonic acid and rainfall effects on
glyphosate activity in trumpetcreeper (Campsis radicans). Weed Technology
18:66-72.

Clark, D. W. 2002. The effect of Old World climbing fern [Lygodium microphyllum
(Cav.) R. Brown] on south Florida cypress [Taxodium distichum (L.) Rich.]
swamp plant and insect community structure. M. S. Thesis. Agronomy
Department, University of Florida. Gainesville, Florida.

Clark, H. M. 1936. The morphology and anatomy of Lygodium japonicum. American
Journal of Botany 23:405-413.

Cole, A. W., and G. E. Coats. 1973. Tall morningglory germination response to
herbicide and temperature. Weed Science 21:443-446.

Conway, E. 1953. Spore and sporeling survival in bracken (Pteridium aquilinum (L.)
Kuhn). Journal of Ecology 41: 289-294.

Crone, E. E., M. Marler, and D. E. Pearson. 2009. Non-target effects of broadleaf
herbicide on a native perennial forb: a demographic framework for assessing and
minimizing impacts. Journal of Applied Ecology 46: 673-682.

De la Cretaz, A. L. and M. J. Kelty. 1999. Establishment and control of hay-scented
fern: a native invasive species. Biological Invasions 1:223-236.

Doyle, P. and M. Stypa. 2004. Reduced herbicide rates a Canadian perspective.
Weed Technology 18:1157-1165.

DuPont Corporation. 2003. DuPont Escort XP Herbicide Special Local Need 24(c)
Label. Available online < http://www.cdms.net/LDat/ld5QT008.pdf >. Accessed 4
January 2010.


145









Egley, G. H. and R. D. Williams. 1978. Glyphosate and paraquat effects onweed seed
germination and seedling emergence. Weed Science 26: 249-251.

Fairbrother, A. and L. A. Kapustka. 2001. Low-dose, high-potency herbicides: a
historical perspective of environmental concerns to frame the issues. Pp. 1- 17
in S. A. Ferenc, ed., Impacts of Low-Dose, High-Potency Herbicides on
Nontarget and Unintended Plant Species. SETAC Press, Pensacola, Florida.

Fernandez, H. and M. A. Revilla. 2003. In vitro culture of ornamental ferns. Plant Cell,
Tissue, and Organ Culture 73:1-13.

Ferriter, A. and T. Pernas. 2006. An explosion in slow motion: tracking the spread of
Lygodium microphyllum in Florida. Wildland Weeds 9:7-9.

[FLAS] University of Florida Herbarium. 2008. Gainesville, FL. Available online <
http://www.flmnh.ufl.edu/herbarium/ >. Accessed 12 December 2009.

Flinn, K. M. 2006. Reproductive biology of three fern species may contribute to
differential colonization success in post-agricultural forests. American Journal of
Botany 93:1289-1294.

[FSU] Florida State University Herbarium. 2008. Tallahassee, Florida. Available online
< http://herbarium.bio.fsu.edu/index.php >. Accessed 12 December 2008.

[FTG] Fairchild Tropical Gardens Herbarium. 2008. Miami, Florida. Available online<
http://www.virtualherbarium.org/default.html >. Accessed 12 December 2009.

Gandiaga, S., J. C. Volin, E. L. Kruger, and K. Kitajima. 2009. Effects of hydrology on
the growth and physiology of an invasive exotic, Lygodium microphyllum (Old
World climbing fern). Weed Research 49:283-290.

Gardner, S. C. and C. E. Grue. 1996. Effects of Rodeo and Garlon 3A on nontarget
wetland species in Central Washington. Environmental Toxicology and
Chemistry 15:441-451.

Geiger, D. R. and H. D. Bestman. 1990. Self-limitation of herbicide mobility by
phytotoxic action. Weed Science 38:324-329.

Gillman, F. S. 2007. The ecological significance of the herbaceous layer in temperate
forest ecosystems. Bioscience 57: 845-858.

Goolsby, J. A. 2004. Potential distribution of the invasive Old World climbing fern,
Lygodium microphyllum in North and South America. Natural Areas Journal
24:351-353.

Goolsby, J. A., A. D. Wright, and R. W. Pemberton. 2003. Exploratory surveys in
Australia and Asia for natural enemies of Old World climbing fern, Lygodium
microphyllum: Lygodiaceae. Biological Control 28:33-46.


146









Goolsby, J. A., P. J. De Barro, J. R. Makinson, R. W. Pemberton, D. M. Hartley and D.
R. Frohlich. 2006. Matching the origin of an invasive weed for selection of a
herbivore haplotype for a biological control programme. Molecular Ecology
15:287-297.

Gordon, D.R. 1998. Effects of invasive, non-indigenous plant species on ecosystem
processes: lessons from Florida. Ecological Applications 8:975-989.

Groninger, J. W. and L. H. McCormick. 1991. Effects of sulfometuron on hay-scented
fern spore germination. Canadian Journal of Forest Research 21:942-943.

Harms, C. T. and J. J. DiMaio. 1991. Primisulfuron herbicide-resistant tobacco cell
lines. Applications of fluctuation test design to in vitro mutant selection with plant
cells. Journal of Plant Physiology 137:513-519.

Haughn, G. and C. R. Somerville. 1987. Selection for herbicide resistance at the whole
plant level. Pp. 98-107 in H. M. LeBaron, R. 0. Mumma, R. C. Honeycutt, and J.
H. Duesing, eds., Biotechnology in Agriculture Chemistry. American Chemical
Society, Washington, D.C.

Heap, I. 1997. The occurrence of herbicide resistance weeds worldwide. Pesticide
Science 51:235-243.

Henry, J. A., K. M. Portier and J. Coyne. 1994. The Climate and Weather of Florida.
Pineapple Press, Inc., Sarasota, Florida.

Hickok, L. G. and R. M. Kiriluk. 1984. Effects of auxins on gametophyte development
and sexual differentiation in the fern Ceratopteris thalictroides (L.) Brongn.
Botanical Gazette 145:37-42.

Hickok, L. G., T. R. Warne, and R. S. Fribourg. 1995. The biology of the fern
Ceratopteris and its use as a model system. International Journal of Plant
Sciences 156:332-345.

Hobbs, R. J. and S. E. Humphries. 1995. An integrated approach to the ecology and
management of plant invasions. Conservation Biology 9:761-770.

Hobbs, R. J. and L. F. Huenneke. 1992. Disturbance, diversity, and invasion:
implications for conservation. Conservation Biology 6:324-337.

Hofstra, D. E., P. D. Champion, and T. M. Dugdale. 2006. Herbicide trials for the
control of parrotsfeather. Journal of Aquatic Plant Management 44:13-18.

Horvitz, C. C. and A. Koop. 2001. Removal of non-native vines and post-hurricane
recruitment of tropical hardwood forests of Florida. Biotropica 33:268-281.


147









Horvitz, C. C., J. B. Pascarella, S. McMann, A. Freedman, and R. H. Hofstetter. 1998.
Functional role of invasive non-indigenous plants in hurricane-affected
subtropical hardwood forests. Ecological Applications 8:947-974.

Hoss, N. E., K. AI-Khatib, D. E. Peterson, and T. M. Loughin. 2003. Efficacy of
glyphosate, glufosinate, and imazethapyr on selected weed species. Weed
Science 51:110-117.

Hutchinson, T. F. and J. L. Vankat. 1997. Invasion by Lonicera maackii (Amur
honeysuckle) in southwestern Ohio forests: community invisibility and effects on
tree seedlings and herbs. Conservation Biology 11:1117-1124.

Hutchinson, J.T. and K.A. Langeland. 2006. Survey of control measures on Old World
climbing fern (Lygodium microphyllum) in southern Florida. Florida Scientist
69:217-223.

Hutchinson, J. T. and K. A. Langeland. 2008. Response of selected nontarget native
Florida wetland plant species to metsulfuron methyl. Journal of Aquatic Plant
Management 46:72-76.

Hutchinson, J.T., G. E. MacDonald and K. A. Langeland. 2007. The potential for
herbicide resistance in non-native plants in Florida's natural areas. Natural
Areas Journal 27:258-263.

Hutchinson, J., A. Ferriter, K. Serbesoff-King, K. Langeland, and L. Rodgers (eds.).
2006. Old World Climbing Fern (Lygodium microphyllum) Management Plan for
Florida. South Florida Water Management District. West Palm Beach, FL.
Available online < http:// www.fleppc.org >. Accessed 10 December 2008.

Kalnay, P. A. and S. Glenn. 2000. Translocation of nicosulfuron and dicamba in hemp
dogbane (Apocynum cannabinum). Weed Technology 14:476-479.

Keary, I. P., C. Thomas, and E. Sheffield. 2000. The effects of the herbicide asulam on
the gametophytes of Pteridium aquilinum, Cryptogramma crispa, and Dryopteris
filix-mas. Annals of Botany 85:47-51.

Khan, A. A. 1981. Effect of leaf position and plant age on the translocation of 14C-
assimilates in onion. Journal of Agriculture Science 96:451-455.

Kirkwood, R. C., R. Hetherington, T. L. Reynolds, and G. Marshall. 2000. Absorption,
localization, translocation, and activity of glyphosate in barnyardgrass
[Echinochloa crus-galli (L) Beauv.]: influence of herbicide and surfactant
concentration. Pest Management Science 56:359-367.

Klekowski, E. J., Jr. 1969. Reproductive biology of the Pteridophyta III. A study of the
Blechnaceae. Botanical Journal of the Linnean Society 62:153-169.


148









Koger, C. H. and K. N. Reddy. 2005. Glyphosate efficacy, absorption, and
translocation in pitted morningglory (Ipomoea lacunose). Weed Science 53:277-
283.

Kohler, E. A., C. S. Throssell, and Z. J. Reicher. 2004. 2,4-D rate response,
absorption, and translocation of two ground ivy (Glechoma hederacea)
populations. Weed Technology 18:917-923.

Krebs, C.J. 1989. Ecological Methodology. Harper Collins Publishers, New York, NY.

Kurumatani, M., K. Yagi, T. Murata, M. Tezuka, L. N. Mander, M. Nishiyama, and H.
Yamane. 2001. Isolation and identification of antheridiogens in the ferns,
Lygodium microphyllum and Lygodium reticulatum. Bioscience, Biotechnology,
and Biochemistry 65:2311-2314.

Lakela, 0. and R. W. Long. 1976. Ferns of Florida. Banyan Books, Miami, FL.

Langeland, K. A. and M. L. Link. 2006. Evaluation of metsulfuron methyl for selective
control of Lygodium microphyllum growing in association with Panicum
hemitomon and Cladium jamaicense. Florida Scientist 69:149-156.

Langeland, K. A., H. M. Cherry, C. M. McCormick and K. A. Craddock Burks. 2008.
Identification and Biology of Nonnative Plants in Florida's Natural Areas. IFAS
Communication Services. University of Florida. Gainesville, Florida.

Laufenberg, S. M., R. L. Sheley, J. S. Jacobs, and J. Borkowski. 2005. Herbicide
effects on density and biomass of Russian knapweed (Acroptilon repens) and
associated plant species. Weed Technology 19:62-72.

Le Duc, M. G., R. J. Pakeman, and R. H. Marrs. 2003. Changes in the rhizome system
of bracken subjected to long-term experimental treatment. Journal of Applied
Ecology 40:508-522.

Le Duc, M. G., R. J. Pakeman, P. D. Putwain, and R. H. Marrs. 2000. The variable
responses of bracken fronds to control treatments in Great Britain. Annals of
Botany 85 (Sup. B): 17-29.

Lemmon, P. E. 1957. A new instrument for measuring forest overstory density.
Journal of Forestry 55:667-668.

Lindgren, P. M. F. and T. P. Sullivan. 2001. Influence of alternative vegetation
management treatments on conifer plantation attributes: abundance, species
diversity, and structural diversity. Forest Ecology and Management 142:163-182.

Lott, M. S., J. C. Volin, R. W. Pemberton, and D. F. Austin. 2003. The reproductive
biology of the invasive ferns Lygodium microphyllum and L. japonicum
(Schizaeaceae): implications for invasive potential. American Journal of Botany
90:1144-1152.


149









Loveless, C. M. 1959. A study of the vegetation in the Florida Everglades. Ecology
40:1-9.

Lovell, S.T., L.M. Wax, M.J. Horak and D.E. Peterson. 1996. Imidazolinone and
sulfonylurea resistance in biotypes of common waterhemp (Amaranthus rudis).
Weed Science 44:789-794.

Lynch, R. L., H. Chen, L. A. Brandt and F. J. Mazzotti. 2009. Old World climbing fern
(Lygodium microphyllum) invasion in hurricane caused treefalls. Natural Areas
Journal 29:210-215.

MacDougall, A.S. and R. Turkington. 2005. Are invasive species the drivers or
passengers of change in degraded ecosystems? Ecology 86:42-55.

Main, C. L., J. E. Beeler, D. K. Robinson, and T. C. Mueller. 2006. Growth,
reproduction, and management of Chinese yam (Dioscorea oppositifolia).Weed
Technology 20:773-777.

Marrs, R.H. 1985. The effects of potential bracken and scrub control herbicides on
lowland Calluna and grass heath communities in East Anglia, UK. Biological
Conservation 32:13-32.

Marrs, R.H., S. W. Johnson, and M. G. Le Duc. 1998. Control of bracken and
restoration of heathland. VII. The response of bracken rhizomes to 18 years
continuous bracken control or 6 years of control followed by recovery. Journal of
Applied Ecology 35:748-757.

McIntyre, G. I. 1962. Preliminary studies on the translocation of 14C-labelled
herbicides in bracken (Pteridium aquilinum). Weed Research 2:51-59.

Moldenke, H. N. 1944. A contribution to our knowledge of the wild and cultivated flora
of Florida 1. American Midland Naturalist 32:529-590.

Moran, R.C. 2004. A Natural History of Ferns. Timber Press, Inc., Portland, Oregon.

Mueller, R. J. 1982. Shoot morphology of the climbing fern Lygodium (Schizaeaceae):
general organography, leaf initiation, and branching. Botanical Gazette 143:319-
330.

Mueller, R. J. 1983. Indeterminate growth and ramification of the climbing leaves of
Lygodium japonicum. American Journal of Botany 70:682-690.

National Hurricane Center. 2009. 2004 Atlantic Hurricane Season. National Oceanic
and Atmospheric Administration, National Weather Service, Miami, Florida.
Available online < http://www.nhc.noaa.gov/ >. Accessed 2 October 2009.


150









[NWRS] National Wildlife Refuge System. 2010. The United States Fish and Wildlife
Services National Wildlife Refuges. Available online <
http://www.fws.gov/refuges/ >. Accessed 15 June 2010.

Nauman, C. E. and D. F. Austin. 1978. Spread of the exotic fern Lygodium
microphyllum in Florida. American Fern Journal 68:65-66.

Norsworthy, J. K., N. R. Burgos, and L. R. Oliver. 2001. Differences in weed tolerance
to glyphosate involve different mechanisms. Weed Technology 15:725-731.

[NPS] National Park Service. 2010. Everglades: International Designations. Available
online< http://www.nps.gov/ever/parknews/internationaldesignations.htm >.
Accessed 9 March 2010.

Page, C. N. 1979. The diversity of ferns an ecological perspective. Pp. 10-56, In: A.
F. Dyer (ed.), The Experimental Biology of Ferns. Academic Press, New York.

Pakeman, R. J. and R. H. Marrs. 1992. The conservation value of bracken, Pteridium
aquilinum (L.) Kuhn, dominated communities in the UK, and an assessment of
the ecological impact of bracken expansion or its removal. Biological
Conservation 62:101-114.

Pakeman, R. J., M. G. LeDuc, and R. H. Marrs. 2000. Bracken distribution in Great
Britain: strategies for its control and the sustainable management of marginal
lands. Annals of Botany 85 (Supp B):37-46.

Pakeman, R. J., J. L. Small, M. G. Le Duc and R. H. Marrs. 1998. An assessment of
aerially applied asulam as a method of log-term bracken control. Journal of
Environmental Management 53:255-262.

Pakeman, R. J., R. H. Thwaites, M. G. Le Duc and R. H. Marrs. 2000. Vegetation re-
establishment on land previously subject to control of Pteridium aquilinum by
herbicide. Applied Vegetation Science 3:95-104.

Pakeman, R. J., J. L. Small, M. G. Le Duc and R. H. Marrs. 2005. Recovery of
moorland vegetation after aerial spraying of bracken (Pteridium aquilinum (L.)
Kuhn) with Asulam. Restoration Ecology 13:718-724.

Pemberton, R.W. 1998. The potential of biological control to manage Old World
climbing fern (Lygodium microphyllum), an invasive weed in Florida. American
Fern Journal 88:176-182.

Pemberton, R. W. 2003. Northward range extension in Florida of the invasive fern
Lygodium microphyllum (Lygodiaceae). Sida 20:1761-1761.

Pemberton, R.W. 2006. Taking down climbing fern biological control of Lygodium
microphyllum. Wildland Weeds 9:10-11.


151









Pemberton, R. W. and A. P. Ferriter. 1998. Old World climbing fern (Lygodium
microphyllum), a dangerous invasive weed in Florida. American Fern Journal
88:165-175.

Petrov, P. and R. H. Marrs. 2000. Follow-up methods for bracken control following an
initial glyphosate application: the use of weed wiping, cutting and reseeding.
Annuals of Botany 85 (Sup. B):31-35.

Pettitt, J. M. 1979. Ultrastructure and cytochemistry of spore wall morphogenesis. Pp.
213-252 in A. F. Dyer, ed., The Experimental Biology of Ferns. Academic Press,
New York.

Prather, T. S., J. M. DiTomaso, and J. S. Holt. 2000. History, mechanisms, and
strategies for prevention and management of herbicide resistant weeds.
Proceedings of the California Weed Science Society 52:155-163.

Preston, C. and S. B. Powles. 2002. Evolution of herbicide resistance in weeds: initial
frequency of target site-based resistance to acetolactate synthase-inhibiting
herbicides in Lolium rigidum. Heredity 88:8-13.

Ranker, T. A. and H. A. Houston. 2002. Is gametophyte sexuality in the laboratory a
good predictor of sexuality in nature? American Fern Journal 92:112-118.

Ray, T. B. 1986. Sulfonylurea herbicides as inhibitors of amino acid biosynthesis in
plants. Trends in Biochemical Sciences 11:180-183.

Reddy, K. N. 2000. Factors affecting toxicity, absorption, and translocation of
glyphosate in redvine (Brunnichia ovata). Weed Technology 14:457-462.

Rice, P. M., J. C. Toney, D. J. Bedunah, and C. E. Carlson. 1997. Plant community
diversity and growth form responses to herbicide applications for control of
Centaurea maculosa. Journal of Applied Ecology 34: 1397-1412.

Roberts, D. 1996. The monster of Hobe Sound or "Why does Lygodium microphyllum
like it so well here." Pp. 29-30 in M. Bodle, ed., Proceedings from Climbing
(Twining) Ferns and Invasive Exotic Vine Workshop. South Florida Water
Management District, West Palm Beach, Florida.

Roberts, D. 1998. Old World climbing fern research and mitigation at Jonathan
Dickinson Park. Resource Management Notes 9:30-32.

Roberts, R. E., R. 0. Woodbury, and J. Popenoe. 2006. Vascular plants of Jonathan
Dickinson State Park. Florida Scientist 69:288-327.

Russell, A. E., J. W. Raich, and P. M. Vitousek. 1998. The ecology of the climbing fern
Dicranopteris linearis on the windward Mauna Loa, Hawaii. Journal of Ecology
86:765-779.


152









[SAS] Statistical Analysis Software. 2002. SAS for Windows, Version 9.2. Cary, NC:
SAS Institute.

Schneller, J. J. 2008. Antheridiogens. Pp. 134-158 in T. A. Ranker and C. H. Haufler,
eds., Biology and Evolution of Ferns and Lycophytes. Cambridge University
Press, United Kingdom.

Schneller, J. J., C. H. Haufler, and T. A. Ranker. 1990. Antheridiogen and natural
gametophyte populations. American Fern Journal 80:143-152.

Smith, A. R., K. M. Pryer, E. Schuettpelz, P. Korall, H. Schneider, and P. G. Wolf. 2006.
A classification of extant ferns. Taxon 55:705-731.

Smith, A. R., K. M. Pryer, E. Schuettpelz, P. Korall, H. Schneider, and P. G. Wolf. 2008.
Fern classification. Pp. 417-467 in T. A. Ranker and C. H. Haufler, eds., Biology
and Evolution of Ferns and Lycophytes. Cambridge University Press, United
Kingdom.

Soltis, D. E. and P. S. Soltis. 1992. The distribution of selfing rates in homosporous
ferns. American Journal of Botany 79:97-100.

Stewart, G. B., A. S. Pullin and C. Tyler. 2007. The effectiveness of Asulam for
bracken (Pteridium aquilinum) control in the United Kingdom: a meta-analysis.
Environmental Management 40:747-760.

Stocker, R. K., A. Ferriter, D. Thayer, M. Rock, and S. Smith. 1997. Old World climbing
fern hitting south Florida below the belt. Wildland Weeds 1:6-10.

Stocker, R. K., R. E. Miller, Jr., D. W. Black, A. P. Ferriter, and D. D. Thayer. 2008.
Using fire and herbicide to control Lygodium microphyllum and effects on a pine
flatwoods plant community in south Florida. Natural Areas Journal 28:144-145.

Swami, P. and V. Raghavan. 1980. Control of morphogenesis in the gametophyte of a
fern by light and growth hormones. Canadian Journal of Botany 58:1464-1473.

Taylor, J. 2006. Management of Old World climbing fern in Everglades National Park.
Pp. 67-69 in Hutchinson, J., A. Ferriter, K. Serbesoff-King, K. Langeland, and L.
Rodgers, eds., Old World Climbing Fern (Lygodium microphyllum) Management
Plan for Florida. South Florida Water Management District, West Palm Beach,
FL. Available online < http:// www.fleppc.org >. Accessed 10 December 2009.

Thomas, B. 2006. A.R.M. Loxahatchee National Wildlife Refuge. Pp. 73-78 in
Hutchinson, J., A. Ferriter, K. Serbesoff-King, K. Langeland, and L. Rodgers,
eds., Old World Climbing Fern (Lygodium microphyllum) Management Plan for
Florida. South Florida Water Management District, West Palm Beach, FL.
Available online < http:// www.fleppc.org >. Accessed 10 December 2009.


153









Thomas, B. Jr. and L. A. Brandt. 2003. Monitoring ground treatments of Old World
climbing fern (Lygodium microphyllum) on the Arthur R. Marshall Loxahatchee
NWR. Wildland Weeds 6:9-11.

Tobe, J. D., K. C. Burks, R. W. Cantrell, M. A. Garland, M. E. Sweeley, D. W. Hall, P.
Wallace, G. Anglin, G. Nelson, J. R. Cooper, D. Bickner, K. Gilbert, N. Aymond,
K. Greenwood, and N. Raymond. 1998. Florida Wetland Plants An
Identification Manual. Florida Department of Environmental Protection.
Tallahassee, Florida.

Toth, L. A. 2009. Hydrological restoration of wetlands on the Kissimmee River
Floodplain: complementarity of an herbicide treatment of the exotic vine
Lygodium microphyllum. Florida Scientist 72:103-120.

Tranel, P. J., and T. R. Wright. 2002. Resistance of weeds to ALS-inhibiting
herbicides: what have we learned? Weed Science 50:700-712.

Tryon, R. M. and G. Vitale. 1977. Evidence for antheridiogen production and its
mediation of a mating system in natural populations of fern gametophytes.
Botanical Journal of the Linnean Society 74:243-249.

Twiss, E. M. 1910. The prothallia of Aneimia and Lygodium. Botanical Gazette
49:168-181.

Ugarte, C. A., L. A. Brandt, S. Melvin, F. J. Mazzotti and K. G. Rice. 2006. Hurricane
impacts on tree islands in Arthur R. Marshall Loxahatchee National Wildlife
Refuge, Florida. Southeastern Naturalist 5:737-746.

Unland, D. R., K. AI-Khatib, and D. E. Peterson. 1999. Interactions between imazamox
and diphenylethers. Weed Science 47:462-466

Veerasekaran, P., R. C. Kirkwood, and W. W. Fletcher. 1977. Studies on the mode of
action of asulam in bracken (Pteridium aquilinum L. Kuhn) I. Absorption and
translocation of (14C) asulam. Weed Research 17:33-39.

Vencill, W. K. (ed.). 2002. Herbicide Handbook (8th ed.). Weed Science Society of
America. Lawrence, Kansas.

Voeller, B. 1971. Developmental physiology of fern gametophytes. Bioscience 21:266-
270.

Vogelmann, T., A. Bassel, and J. Miller. 1981. Effects of microtubule-inhibitors on
nuclear migration and rhizoid differentiation in germinating fern spores (Onoclea
sensibilis). Protoplasma 109:295-316.

Volin, J. C., M. S. Lott, J. D. Muss, and D. Owens. 2004. Predicting rapid invasion of
the Florida Everglades by Old World climbing fern (Lygodium microphyllum).
Diversity and Distributions 10:439-446.


154









Walker, E. R. and L. R. Oliver. 2008. Translocation and absorption of glyphosate in
flowering sicklepod (Senna obtusifolia). Weed Science 56:338-343.

Watkins, J. E., Jr., M. K. Mack, and S. S. Mulkey. 2007. Gametophyte ecology and
demography of epiphytic and terrestrial tropical ferns. American Journal of
Botany 94:701-708.

Wetzel, P.R., A.G. van der Valk, S. Newman, D.E. Gawlik, T.T. Gann, C.A. Coronado-
Molina, D.L. Childers, and F.H. Sklar. 2005. Maintaining tree islands in the
Florida Everglades: nutrient redistribution is the key. Frontiers in Ecology and
the Environment 3:370-376.

Wikstrom, N., P. Kenrick, and J. C. Vogel. 2002. Schizaeaceae: a phylogenetic
approach. Review of Palaeobotany and Palynology 119:35-50.

Williams, C.B. 1964a. Patterns in the Balance of Nature. Academic Press, London,
United Kingdom.

Williams, R.D. 1964b. Assimilation and translocation in perennial grasses. Annals of
Botany 28:419-426.

Wilson, K. A. 2002. Continued pteridophyte invasion of Hawaii. American Fern Journal
92:179-183.

Wolf, S. 1993. Effect of leaf age on photosynthesis, carbon transport and carbon
allocation in potato plants. Potato Research 36:253-262.

Woodmansee, S. W. and J. L. Sadie. 2005. New occurrences of Schizaea pennula
Sw. in Florida. American Fern Journal 95:84-87.

Woodmansee, S. W., K. Bradley, and S. Hodges. 2005. Systematic Reconnaissance
Flights and Exotic Plant Species Mapping at Selected National Wildlife Refuges
in Florida. Final Report. Submitted to U.S. Fish and Wildlife Service, Sanibel,
Florida.

Wu, Y. K. Rutchey, N. Wang, and J. Godin. 2006. The spatial pattern and dispersion of
Lygodium microphyllum in the Everglades wetland ecosystem. Biological
Invasions 8:1483-1493.

Wunderlin, R. P. and B. F. Hansen. 2003. Guide to the Vascular Plants of Florida, 2nd
edition. University Press of Florida, Gainesville, Florida.

Wunderlin, R. P., and B. F. Hansen. 2008. Atlas of Florida Vascular Plants. Institute for
Systematic Botany, University of South Florida, Tampa. [S. M. Landry and K. N.
Campbell (application development), Florida Center for Community Design and
Research]. Available online < http://www.plantatlas.usf.edu/>. Accessed: 12
December 2008.


155









Young, F. L., D. R. Gealy, and L. A. Morrow. 1984. Effect of herbicides on germination
and growth of four grass weeds. Weed Science 32: 489-493.

Zar, J. H. 1999. Biostatisical Analysis, 4th Edition. Prentice Hall International Inc.,
Englewood Cliffs, New Jersey.


156









BIOGRAPHICAL SKETCH

Jeffrey T. Hutchinson was born in Salisbury, North Carolina. He graduated from

West Rowan High School in 1980 and attended N.C. State University from 1980-1982.

In 1983, he attended Western Carolina University for two semesters. From 1984-1990

he served as an electrical technician in the United States Marine Corps and was

stationed in South Carolina, Tennessee, Georgia, California, Hawaii, Japan and Korea.

From 1990-1992, Jeff attended Central Florida Community College in Ocala and

received an A.S. in biology in 1992. From 1992-1995, he attended the University of

Florida and received a B.S. in wildlife ecology with a minor in forestry in 1995. From

1996-1998, he attended the University of Kentucky and received his M.S. in forestry in

1998 under the direction of Dr. Michael Lacki. His thesis focused on characterizing the

roost sites and habitat characteristics of forest dwelling bats. Jeff worked as a district

biologist for the Florida Park Service from 1998-2002 and as land manager at Archbold

Biological Station from 2002-2004. In 2004, Jeff began his dissertation work on OWCF

under the direction of Dr. Kenneth Langeland at the University of Florida's Center for

Aquatic and Invasive Plants. His career interests include all aspects of natural resource

conservation.


157





PAGE 1

PHYSIOLOGICAL CHARACTERISTICS OF HERBICIDES AND MANAGEMENT OF OLD WORLD CLIMBING FERN (LYGODIUM MICROPHYLL UM) By JEFFREY T. HUTCHINSON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010 1

PAGE 2

2010 Jeffrey T. Hutchinson 2

PAGE 3

To my wife, Thea, thanks for the love and support 3

PAGE 4

ACK NOWLEDGMENTS First and foremost, I thank my wife, Thea, my most trusted friend, companion, supporter, and cosmic mate for 23+ years fo r her love and support. I am forever grateful and indebted to Dr Kenneth Langeland for offering me the opportunity to pursue a Ph.D. in weed science. Dr. Langeland allowed me a unique opportunity to work independently, but always maintained an open door policy. My committee members all greatly contributed to the knowledge I learned over the last 5+ years. Dr. Robert St amps took several days of his time to show me the basic culture techniques of fern propagation. Dr. Doria Gordon provided very insightful comments that I hope are exhi bited throughout much of this dissertation. Dr. MacDonald offered great insight and comments on the design of experiments as well as allowing me to use his lab. Dr. William Ha ller offered insightful comments on different herbicides to screen. It is nearly impossible to recall all the assistance and support I received over the last 5+ years for field sites. Dr. Laura Brandt, Stefani Melvin, Gayle Martin, Dr. Mark Barrett, Jeremy Conrad, Lisa Jameson, Jon Wallace and all staff members at A.R.M. Loxahatchee N.W.R. provided th e best support possible from 2005-2009. Kevin Maier, George Pelt and Don Napier were very helpful with the initial setup of plots on tree islands. Gayle Martin was a constant contact and went well beyond the call of duty many times when I needed help. I cant think of a better place to do field research than A.R.M. Loxahatchee N.W.R. Dr. John Volin (now at the University of Connecticut) and his staff at Florida Atlantic University provided valuable insight on culture methods of OWCF spores. Dr. Michael Kane and his staff at the University of Florida provided valuable insight on 4

PAGE 5

various methods that could be us ed to handle and count OWCF spores. Dr. Mete Yilmaz at the University of Florida gladly allowed me to use his lab on a moments notice, going well beyond the call of duty expected of him. Dr. Fr ank Chapman allowed me to use about every piece of equipment in his lab without ever making a minor grievance, and at the same time offering me very informative advice along the way. I am very thankful for the advice Dr. Chapman gave me. Bob Querns, David Mayo, Aimee Copper an d Pedro Mendez at the University of Florida provided assistance in the lab and field, and I am grateful fo r their support. Bob Querns was very helpful with the translo cation project and his help was greatly appreciated. Thea Hutchinson provided assistanc e in the field multiple times digging up OWCF plants in south Florida. Many staf f and faculty members in the University of Florida Fisheries Department allowed me to use their equipment and provided practical advice, and I am grateful to all of them. I would like to thank the U.S. Fish and Wild life Service, Florida Fish and Wildlife Conservation Commissions Bureau of Invasive Plants, South Florida Water Management District, St. Johns River Wa ter Management District, and the Florida Exotic Pest Plant Council for providing f unding. Without the fina ncial support of these agencies, these projects would not have been possible. Finally, I want to thank my parents, Melvia Hall and Cecil Hutchinson for their support over the years. My in-laws, now deceased, Helen and Amurat Bogushevitz were a constant source of moral and financ ial support, which made life easier. My crazy dogs, Maggie and Lola, made life fun and stress-free. I will a lways be grateful for the time I got to spend with Phillip Myers, a great friend and a big source of inspiration. 5

PAGE 6

TABL E OF CONTENTS page ACKNOWLEDG MENTS..................................................................................................4LIST OF TABLES............................................................................................................9LIST OF FI GURES ........................................................................................................12ABSTRACT ...................................................................................................................14CHAPTER 1 INTRODUC TION....................................................................................................172 ABSORPTION AND TRANSLOCATION OF GLYPHOSATE, METSULFURON, AND TRICLOPYR IN OLD WORLD CLIMBING FERN (LYGODIUM MICROPHYLLUM)..................................................................................................27Introducti on.............................................................................................................27Materials and Methods............................................................................................30Plant Mate rial...................................................................................................30Absorption and Tr anslocati on...........................................................................31Experimental Design and Data An alysis ..........................................................32Results and Discussion...........................................................................................33Visual Effects of He rbicides on OWCF .............................................................33Autoradiog raphy...............................................................................................34Absorpti on........................................................................................................35Transloca tion....................................................................................................363 TOLERANCE OF OLD WORLD CLIMBING FERN SPORES AND GAMETOPHYTES TO HE RBICIDE EXPO SURE...................................................46Introducti on.............................................................................................................46Materials and Methods............................................................................................48Source of Spores..............................................................................................48Spore Prepar ation............................................................................................49Spore Co unts...................................................................................................49Germination Medium ........................................................................................49Spore Inoculation.............................................................................................50Germination Rates...........................................................................................51Sporophyte Developm ent.................................................................................51Experiment 1 Tolerance of Ga metophytes to Me tsulfuro n.............................52Experiment 2 Tolerance of Spores to Selected Herbicides............................53Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium......54Result s....................................................................................................................55Experiment 1 Tolerance of Ga metophytes to Me tsulfuro n.............................55 6

PAGE 7

Experiment 2 Tolerance of Spores to Selected Herbicides ............................56Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium......57Discussio n..............................................................................................................58Experiment 1 Tolerance of Ga metophytes to Me tsulfuro n.............................58Experiment 2 Tolerance of Spores to Selected Herbicides............................60Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium......624 REPEATED HERBICIDE APPLICATION FOR CONTROL OF OLD WORLD CLIMBING FERN AND THE EFFECTS ON NON-TARGET VEGETATION ON EVERGLADES TREE ISLAND S............................................................................83Introducti on.............................................................................................................83Materials and Methods............................................................................................87Study Si te.........................................................................................................87Evaluation Periods ...........................................................................................88Herbicide Tr eatment.........................................................................................88Ground Cove r...................................................................................................89Canopy Cover..................................................................................................89Plant Spec ies...................................................................................................90Native Shrubs and Tr ees..................................................................................91Statistical Analys is............................................................................................91OWCF..............................................................................................................91Cover, Richness, and Evenness......................................................................93Shrub and Tree Surviv al...................................................................................93Result s....................................................................................................................94OWCF Ground Cover.......................................................................................94Native plant ground cove r..........................................................................95Shrub and tree gr ound cover.....................................................................96Canopy co ver.............................................................................................96Shrub and Tree Surviv al...................................................................................97Ilex cassine................................................................................................97Persea palus tris.........................................................................................97Myrica ce rifera...........................................................................................97Rapanea punc tata......................................................................................98Plant Richness Dynamic s.................................................................................98General tr ends...........................................................................................98Total species richnes s...............................................................................98Late succession spec ies ric hness..............................................................99Ruderal species richne ss...........................................................................99Other non-native spec ies richness.............................................................99Species ev enness....................................................................................100Discussio n............................................................................................................100Ground Cove r.................................................................................................100OWCF managem ent................................................................................100Plant diversity dynamics ..........................................................................103Native ground cover.................................................................................105Tree Survival and Canopy Cove r...................................................................108 7

PAGE 8

Management Implic ations............................................................................... 1115 SUMMARY ...........................................................................................................130Management Recomm endations ....................................................................132Additional Re search.......................................................................................133APPENDIX PLANT SPECIES DOCUMENTED ON TREE ISLANDS AT A.R.M. LOXAHATCHEE N.W.R. TREAETED WITH METSULFRUON (0.08 AND 0.16 KG A.I.) AND GLYPHOSATE (2. 80 AND 5.60 KG A.I.)........................................135LIST OF REFE RENCES.............................................................................................144BIOGRAPHICAL SKETCH ..........................................................................................157 8

PAGE 9

LIST OF TABLES Table page 2-1 Sources of material used in OWCF 14C translo cation st udy..............................422-2 Radioactivity (percent of applied) and standard error for total recovery, leaf wash, absorbed, treated leaf, and translo cation following application of 14Clabeled metsulfuron, glyphosate, and triclopyr to OWCF (all treatments combined) ...........................................................................................................432-3 Percent 14C-metsulfuron absorption, translocation, and distribution in OWCF for five treatment methods. Values repr esent the means of 10 replications.......432-4 Percent 14C-glyphosate absorption, transloca tion, and distribution in OWCF for five treatment methods. Values repr esent the means of 10 replications.......442-5 Percent 14C-triclopyr absorption, translocati on, and distribution in Old World climbing fern for five treatment methods. Values represent the means of 10 replicatio ns.........................................................................................................443-1 Site, county, managing agency and geogr aphical pre-positioning system (GPS) coordinates of locations wher e OWCF spores we re collected.................653-2 Collection site and age of OWCF spor es in weeks for each experiment............663-3 Herbicide concentrations used to eval uate effects on Old World climbing fern spore germination. ..............................................................................................663-4 Effects of various concentrations of metsulfuron on germination of Old World climbing fern spores and sporophyte survival .....................................................673-5 Inhibition rates (I50 and I95) for germination of Old Wo rld climbing fern spores following exposure to six different herbi cides based on non-linear regression...684-1 Herbicide treatments and treatment frequency for aerial and ground herbicide application on tree islands in the A.R.M. Loxahatchee National Wildlife Refuge. .............................................................................................................1134-2 Effects of aerial herbicide applicat ions on Old World climbing fern ground cover one year after applicat ion with pre-treatment cover as the covariate......1134-3 Effects of aerial and a single ground treatment of herbicide on Old World climbing fern ground cover at one y ear following ground treatment and two years following aerial applic ation with one year post treatment cover as the covariat e...........................................................................................................114 9

PAGE 10

4-4 Effects of aerial and two annual ground treatments of herbicide on Old World climbing fern ground cover at year 4 wit h year 1 cover as the covariate and time...................................................................................................................1144-5 Effects of aerial and two annual ground treatments of herbicide on Blechnum serrulatum (swamp fern) ground cover at year 4 with year 1 cover as the covariate and time............................................................................................1154-6 Effects of aerial and two annual ground treatments of herbicide on Smilax laurifolia (green brier) ground cover at year 4 with year 1 cover as the covariate and time............................................................................................1154-7 Effects of aerial and two annual ground treatments of herbicide on Osmunda cinnamomea (cinnamon fern) ground cover at year 4 with year 1 cover as the covariate and time............................................................................................1164-8 Effects of aerial and two annual ground treatments of herbicide on Woodwardia virginica (Virginia chain fern) ground cover at year 4 with year 1 cover as the cova riate and ti me........................................................................1164-9 Effects of aerial and two annual ground treatments of herbicide on Vitis rotundifolia (muscadine) ground cover at year 4 with year 1 as the covariate and time ............................................................................................................1174-10 Effects of aerial and two annual ground treatments of herbicide on Osmunda regalis (royal fern) ground cover at year 4 with year 1 as the covariate and time...................................................................................................................1174-11 Effects of aerial and two annual ground treatments of herbicide on Cladium jamaicense (sawgrass) ground cover at year 4 with year 1 cover as the covariate and time............................................................................................1184-12 Effects of aerial and two annual ground treatments of herbicide on Myrica cerifera (wax myrtle) ground cover at year 4 with year 1 cover as the covariate and time............................................................................................1184-13 Effects of aerial and two annual ground treatments of herbicide on Persea palustris (swamp bay) ground cover at year 4 with year 1 cover as the covariate and time............................................................................................1194-14 Effects of aerial and two annual ground treatments of herbicide on Ilex cassine (Dahoon holly) ground cover at year 4 with year 1 cover as the covariate and time............................................................................................1194-15 Effects of aerial and two annual ground treatments of herbicide on Rapanea punctata (myrsine) ground cover at year 4 wit h year 1 cover as the covariate and time ............................................................................................................120 10

PAGE 11

4-16 Effects of aerial and two annual gr ound treatments of herbicide on tree canopy cover at year 4 with year canopy cover as the covariate and time.......1204-17 Survival rates (%) of the four most common shrub and tree species ( Ilex cassine, Persea palustris Myrica cerifera and Rapanea punctata ) at three years post aerial herbi cide treatm ent................................................................1214-18 Summary statistics for logistic regre ssion analysis of tree variables as model predictors of Ilex cassine (Dahoon holly) survival following herbicide treatment ..........................................................................................................1214-19 Summary statistics for logistic regre ssion analysis of tree variables as model predictors of Persea palustris (swamp bay) survival following herbicide treatment ..........................................................................................................1224-20 Summary statistics for logistic regre ssion analysis of tree variables as model predictors of Myrica cerifera (wax myrtle) survival following herbicide treatment ..........................................................................................................1224-21 Summary statistics for logistic regre ssion analysis of tree variables as model predictors of Rapanea punctata (myrsine) survival following herbicide treatment ..........................................................................................................1234-22 Total number of late succession, ruderal, non-native (including OWCF) and all plant species recorded pre-treatment (y ear 0) to year three post treatment for metsulfuron and glyphosate treated tr ee islands and untreated checks......1244-23 Effects of aerial and two annual gr ound treatments of herbicide on total species richness (including non-native spec ies) at year 4 with year 1 as the covariate and time............................................................................................1254-24 Effects of aerial and two annual gr ound treatments of herbicide on late succession species richness at year 4 wit h year 1 as the covariate and time..1254-25 Effects of aerial and two annual ground treatments of herbicide on ruderal species richness at year 4 with year 1 as the covari ate and time.....................1264-26 Effects of aerial and two annual ground tr eatments of herbicide on non-native species (other than Old World climbing fe rn richness at year 4 with year 1 as the covariate and time ......................................................................................1264-27 Effects of aerial and two annual ground treatments of herbicide on species evenness at year 4 with year 1 as the covariat e and time ................................127 11

PAGE 12

LIST OF FIGURES Figure page 1-1 Number of OWCF herbarium record s per decade from the University of Florida, Florida Sate University, the Univ ersity of South Flor ida, and Fairchild Tropical Ga rdens................................................................................................262-1 Dried plants and autoradi ographs of OWCF treated ba sally with metsulfuron, glyphosate, and triclopy r.....................................................................................453-1 Location of collection sites of Old Wo rld Climbing Fern spores in Florida..........693-2 Germination rates of Old World climbing fern spores of different age groups....703-3 Germination rates (%) of Old World climbing fern spores collected from different location in Flor ida.................................................................................713-4 Survival of Old World climbing fern gametophytes 30 days after (60 total days in agar + nutrient medium) exposur e to metsulfuron in agar medium........723-5 Survival of Old World climbing fe rn sporophytes 120 days (60 days in agar + nutrient medium and 60 days in potted so il) after exposure to metsulfuron.....733-6 Old World climbing fern sporophyte dry weight (grams) at 120 days (60 days in agar + nutrient medium and 60 days in potted soil) after exposure to metsulfuron on agar medium and transplanted to potting so il............................743-7 Effects of metsulfuron on Old World climbing fern spore germination 30 days post expos ure.....................................................................................................753-8 Effects of glyphosate on Old Worl d climbing fern spore germination 30 days post expos ure.....................................................................................................763-9 Effects of imazapyr on Old World climbing fern spore germination 30 days post expos ure.....................................................................................................773-10 Effects of triclopyr on Old World c limbing fern spore germination 30 days post expos ure.....................................................................................................783-11 Effects of fluroxypyr on Old Worl d climbing fern spore germination 30 days post expos ure.....................................................................................................793-12 Effects of asulam on Old World clim bing fern spore germination 30 days post exposu re.............................................................................................................80 12

PAGE 13

3-13 Germination rates of Old World clim bing fern spores placed on an agar mixed with metsulfuron (0.4 g a.i. / l) or agar al one (untreated checks) 15 and 45 days post-treatment from spores collected at 12 locati ons............................813-14 Germination of Old World climbing fern spores collected Indian River Community College (IRCC) and Jonathan Dickinson State Park (JDSP) placed on an agar media mixed with metsulfu ron (0.4 g a.i. / l) or agar alone (untreated checks) 45 days post-tr eatment ........................................................824-1 Sampling design used to evaluate responses of ground cover ( 1.5 m), canopy cover (> 1.5 m), and tagged trees on northern Everglades tree islands treated with herbicides to cont rol Old World clim bing fern ....................1284-2 Conceptual model on the effects of OWCF on tree islands and sequence of events following herbici de treatment ................................................................129 13

PAGE 14

Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Pa rtial Fulfillment of the Requirements for t he Degree of Doctor of Philosophy PHYSIOLOGICAL CHARACTERISTICS OF HERBICIDES AND MANAGEMENT OF OLD WORLD CLIMBING FERN (LYGODIUM MICROPHYLLUM) By Jeffrey T. Hutchinson August 2010 Chair: K. A. Langeland Major: Agronomy Absorption and translocation patterns were determined in Old World climbing fern [Lygodium microphyllum (Cav.) R. Br.; OWCF] using the three herbicides most commonly used for management of this plant by land managers in Florida. Using 14Clabeled herbicides, absorption and translocation were evaluated for glyphosate, metsulfuron, and triclopyr in OWCF using fi ve different application scenarios (cut and spray, basal spray, 25% foliar spray, 50% foliar spray, and 100% foliar spray). Triclopyr was absorbed to the greatest extent (60.3%) of applied radioactively compared to glyphosate (31.2%) and metsulfuron (19.8%). The majority of radioactivity remained in treated leaves for all herbicides with only small percentages of the absorbed radioactivity being detected in other plant parts. All three herbicides translocated acropetally and basipitally to some extent. Radioactivity, for the mo st part, translocated evenly throughout the plants but the greatest amount of radioactivity derived from triclopyr occurred in rhizomes when the cut and spray and basal applications were used. The radioactivity in rhizomes der ived from glyphosate was greater in those treated using cut and spray. Based on autor adiographs, there was limited horizontal 14

PAGE 15

movement of any herbicide in the rhizomes of OWCF which may explain why re-sprouts are observed several weeks following treatment. OWCF gametophytes were highly su sceptible to metsulfuron with 1.4% survival at concentration 27 mg a.i./l. Inhibition conc entrations of OWCF gametophytes exposed to metsulfuron were 6.1 and 26.4 mg a.i./l for the I50 and I95 values, respectively. Survival of treated gamet ophytes that developed into sporophytes was 0.014% at concentration 27 mg a.i./l, but no sporophytes developed at concentrations 432 mg a.i./l. Inhibition concentrations of OWCF sporophytes were 5.6 and 24.1 mg a.i./l for the I50 and I95 values, respectively. Bas ed on the total number of OWCF gametophytes treated that developed into sporophytes, the tolerance level for a concentration of 216 mg a.i./l metsulfuron is 2.0 x 10-5. This indicates that the potential for tolerant OWCF populations is high using lower concentrations as no gametophytes developed into sporophytes at the standard field treatment concentration of 432 mg a.i./l. Spores of OWCF were highly suscept ible to metsulfuron, but exhibited tolerance to imazapyr, glyphosate, flur oxypyr, asulam, and triclopyr ex posed in bathing solution. Spore germination rates were 0.4% for spores ex posed to 0.1 g a.i./l of metsulfuron, but 0% for rates 0.2 g a.i./l. Reduction in spore germination was observed with all other concentrations of herbicides tested, ranging fr om 10.4% with triclopy r (40.0 g a.i./l) to 42.6% with asulam (4.2 g a.i./l) compared to 47.9% germination for untreated checks. Metsulfuron also exhibited residual ac tivity on OWCF gametophyte development, growth and survival on agar containing a c oncentration of 0.4 g a.i./l of metsulfuron compared to untreated checks. These results may explain why greater control has been achieved with metsulfuron for control of OWCF. 15

PAGE 16

Aerial and ground applications of metsulfuron on Everglades tree islands wer e more effective in reducing OWCF co ver than aerial and gr ound application of glyphosate. Both herbicides resulted in si gnificant non-target dam age to ground cover. There was a reduction in native climax s pecies ground cover and an increase in ruderal species ground cover at the end of the study. High tree survival and limited reductions in canopy cover were observed on tree isl ands treated with metsulfu ron, while low tree survival and large decreases in canopy co ver were observed on tree islands treated with glyphosate. Based on the results of this study, metsulfuron should be aerially applied to for initial control of OWCF and followed up with ground treatments using glyphosate or herbicides wit h other modes of action. 16

PAGE 17

17 CHAPTER 1 INTRODUCTION Old World climbing fern [ Lygodium microphyllum (Cav.) R. Br; OWCF] is native to Australia, Africa, Asia and Oceania (Pemberton 1998). Within its native range, OWCF is found in wet tropical and subt ropical habitat. OWCF belongs to the order Schizaeales and the climbing fern family, Lygodiaceae, which is comprised of approximately 25 species (Smith et al. 2006). Fossil records i ndicate that the family is at least 65-100 million years old (Wikstrm et al. 2002). The family is c haracterized by protostelic subterranean rhizomes, a flexible and twining rachis, leaves that are two to four pinnate, solitary sporangia on the abaxial surface co vered by an indusium, and indeterminate growth to > 20m (Page 1979; Smith et al. 2006 and 2008). Species of Lygodium are most common in open swampy forests and forest edges (Page 1979). Spores of Lygodiaceae are 80 m, tetrahedral in shape with three layers, the epispore, exine, and intine (Twiss 1910; Cla rke 1936). OWCF has an average of 133 sori on each fertile leaflet with an average of 215 spores per sorus (Volin et al. 2004). Spore production in OWCF occurs throughout t he year in south Flor ida, but is highest from September to November (Volin et al. 2004). Germination of OWCF spores requires moisture and occurs in five to seven days following hydration (Twiss 1910). Upon germination, OWCF spores develop into male, female or hermaphroditic haploid gametophytes and reach sexual maturity in fi ve to eight weeks (Lott et al. 2003; Gandiaga et al. 2009). Sexual differentia tion among gametophytes is determined by antheridiogens (Kurumatani et al. 2001) produced by the fastest growing female gametophytes. This promot es male gametophyte development which facilitates

PAGE 18

outcrossing and prevents inbreeding (Schne ller 2008). After fertilization, the conspicuous diploid sporophytes are form ed at 13-14 weeks (Gandiaga et al. 2009). The rachis of Lygodium spp. arise from subterranean rhizomes, and exhibit continuous apical growth, circumnutation, and indeterminate growth (Clarke 1936; Mueller 1982). The climbing habit of OW CF is characterized as a searcher morphology in which the rachis expands with minimal pinnae expansion to reduce the weight of the plant until it finds support (Muller 1983). Once the rachis finds support, the pinnae expand while the rachis continues to expand acropetally in a twining nature (Muller 1983). This growth habit allows OWCF to growth horizontally along the ground and shrub layers and vertically into the canopy. Pinnae along expanding rachis divide into two pinnules with a dormant bud in between, and the terminal bud along an expanding rachis will become ac tive if the apical tip is damaged (Clark 1936). Cross sections of OWCF pinnae include the presenc e of an upper epidermis, mesophyll cells, and lower epidermis (Clark 1936), indicating similarity to the leaves of angiosperms. OWCF has spread rapidly across the landscape of southern Florida and is increasing its range into central Florida (Ferriter and Pernas 2006; Pemberton and Ferriter 1998; Pemberton 2003). The distribut ion of OWCF on Everglade tree islands is correlated with the density and spat ial distribution of tree island s, indicating that the fern is likely still in an early stage of invasion (Wu et al. 2006). In a survey of Florida flor a from 1927-1930, no reports of Lygodium spp. were documented (Moldenke 1944). Floral surveys of Water Conservation Areas 2 and 3 in the central Everglades failed to document any Lygodium spp. in the late 1950s 18

PAGE 19

(Loveless 1959), but presently OWCF is co mmon on many tree i slands in the northern Everglades and coastal prairies in the western Everglades. The origin of the OWCF population in south Florida is believed to be Queensland, Australia (Goolsby et al. 2006). OWCF was first reported in the United States in cultivation from nurseries in Davenport (Pol k County), Florida (Florida Division of Plant Industry, Record #0-3753) and Delray Beach (Palm Beach County), Florida (University of Florida Herbarium, FLAS P5100) in 1958. It was first reported as escaped and invading natural areas in the vicinity of J upiter in northern Palm Beach County, Florida (Beckner 1968). OWCF was reported to be rare in Florida as late as 1976 when sporulating specimens were recorded in Ma rtin County (Lakela and Long 1976). By 1978, OWCF was reported to be naturalized and invasive in Palm Beach and Martin Counties with cover at one site in Palm Beach County estimated to be 7.4 ha (Nauman and Austin 1978), the first indicati on of the invasive potential of this fern. Initially, OWCF was confined to wet, disturbed areas such as canals, rivers, ditches and disturbed swamps with all sites having standing water for much of the year (Nauman and Austin 1978). By 1984, OWCF had invaded 21% of t he wetlands in Jonathan Dickinson State Park, Martin County (Brown 1984). In 2003, the area coverage of OWCF in sout h Florida (all areas south of the north section of Lake Okeechobee) was estimat ed to be 48,878 ha (Ferriter and Pernas 2006). In a 2004 aerial survey of Highlands and Polk Counties along the Lake Wales Region of south-central Fl orida, 213 locations with Lygodium spp. were documented, with some infestations of OWCF as large as 40-80 ha (Biehl 2004). By 2005, OWCF was found as far north as Hillsborough County on the west coast and Volusia County on 19

PAGE 20

the east coast (Wunderlin and Hansen 2008) In December 2008, an unreported population of OWCF (FLAS 22 5459) was documented in northern Lake Count y, Florida, ca. 25 km south of the Mari on County line. This population of OWCF represents a 40 km range increase to the north in central Florida relative to the population reported by Pemberton (2003) in east Orlando. Bas ed on herbarium records (FLAS 2008; FSU 2008; FTG 2008; Wunderlin and Hansen 2008), there appears to have been a rapid expansion of OWCF in the 1990s (Figure 1-1). The herbarium records suggest a 30 year lag time between the introduction of OWCF and the fern becoming invasive. From 1960-1973, all herbarium records for OWCF occurrences were from Martin County, except a single herbarium record from Highland County in 1973 (FLAS 2008; FSU 2 008; FTG 2008). From 1975-1979, all herbarium records were documented from Palm Beach County (FLAS 2008; FSU 2008; FTG 2008), with a single herbarium record fr om Polk County (FLAS 2008). In 1981, OWCF was documented in Collier County (W underlin and Hansen 2008) with additional records in Highlands and Polk Counties (FTG 2008; Wunderlin and Hansen 2008). From 1991-1996, OWCF was documented in Charlotte, Desoto, Hardee, Manatee, and Lee Counties (FLAS 2008; FSU 2008; FTG 2008; Wunderlin and Hansen 2008), indicating a range increase from east to west Florida. In 1997, a range increase was documented in Brevard County along the southeast coast of Florida (Wunderlin and Hansen 2008). In 1999, herbarium records indicate OWCF had moved further north along the southwest coast of Florida with documentations from Hillsborough County (Wunderlin and Hansen 2008). Additional reco rds of OWCF were recorded in Brevard County along the east coast of Florida in 2002 (FTG 2008; Wunderlin and Hansen 20

PAGE 21

2008). Documentation of OWCF in Sumt er County from 2008 (Wunderlin and Hansen 2008) and from Lake County in 2009 indicates that OWCF is slowly moving northward in Florid a. In addition to invading further north into c entral Florida, the pot ential distribution of OWCF, based on climate modeling in the New World, indicates that climate and environmental conditions are suitable for ex pansion into the southern tip of Texas, Caribbean islands, Central Amer ica, and large parts of Sout h America (Goolsby 2004). Therefore, OWCF exhibits the potential to invade areas in the new World tropics with high diversity. Rapid spread of OWCF across the landsca pe of southern Florida in less than 52 years since it was first documented is likel y due in part to the production of numerous wind-blown spores that can potentially trav el many kilometers (Pemberton and Ferriter 1998). A single fertile pinnae can potentially produce 28,600 spores (Volin et al. 2004), and a single spore is capable of intragametophy tic selfing under la boratory conditions. Thus, one spore can theoretically establish a popul ation at a distant lo cation far from the parent plant (Lott et al. 2003). Sampling for airborne OWCF spores at one location in Martin County, Florida result ed in an average of 724 spores/m3/hour being collected (Pemberton and Ferriter 1998). The rapid invasion of OWCF in central and southern Florida can be attributed to multiple factor s such as copious spore production, long distance dispersal of spores, mixed mating system, rapid growth, vertical and horizontal growth pattern, and a lack of natural enemies. Intragametophytic selfing appears to be ra re in most species of homosporous ferns (Soltis and Soltis 1992), but may have cont ributed to the rapid spread of OWCF in 21

PAGE 22

Florida (Lott et al. 2003). Further evidence that selfing may facilitate the spread of some ferns into new locations is provided by Flinn (2006). Tryon and Vitale (1977) were the first to observe an antheridio gen system in natural populations of Asplenium pimpinellifolium Fee. and Lygodium heterodoxum Kze. Antheridiogens have also been hypothesized to inhibit interspecific competit ion among other fern species (Schneller et al. 1990). Furthermore, OWCF gametophytes r each sexual maturity in five weeks and produce sporophytes from five to 12 weeks, which is highly advantageous in the seasonal wet and dry cycle of sout hern Florida (Lott et al. 2003). OWCF has invaded a variety of mesic and hy dric natural and disturbed habitats in central and southern Florida (Pemberton and Ferriter 1998; Hutchinson and Langeland 2006). Natural habitats invaded by OWCF include cypress swamps, bayhead swamps, hardwood swamps, pine flatwood s, Everglades tree islands, coastal prairies, and to a lesser degree marsh, lake edges, and mangrove edges. Environmental factors correlated with the presence of OWCF are the length and depth of inundation and soil water content (Volin et al. 2004). The fe rn is also commonly found in any type of disturbed habitat with mesic or hydric soils. Three native plants most commonly associated with OWCF are swamp fern (Blechnum serrulatum ), false nettle ( Boehmeria cylindrica ), and royal fern ( Osmunda regalis ) (Volin et al. 2004). These three species are obligate or facultative indicators of we tland habitat (Tobe et al. 1998), indicating that OWCF could potentially invade any wetland habitat in southern or central Florida. OWCF has invaded some of the least distur bed and isolated habitat in Florida such as Everglades tree islands, coastal prairi es, cypress swamps, bay swamps, and pine flatwoods. 22

PAGE 23

On Everglades tree islands, OWCF has reduced the abundanc e of native plants and created more homogeneous tree islands dominated by OWCF (Brandt and Black 2001). Lower native plant cover, richness and diversity were documented in habitats heavily infested with OWCF compared to similar habitats in which no OWCF was present (Clark 2002). In addition, OWCF is a severe threat to many threatened and endangered plants (Woodmansee and Sadle 2005; Hutchinson et al. 2006). Two of the most common land managemen t activities conducted by natural resource agencies in Florida are prescri bed fire and invasive plant treatment with herbicides. Management of OWCF has primarily relied on ground or aerial applications of glyphosate {N-(phosphonomethyl) glycine}, metsulfuron {Methyl 2-[[[[(4-methoxy-6methyl-1,3,5-triazin-2-yl)amino]-carbonyl]am ino]sulfonyl]benzoate}, and triclopyr {3,5,6trichloro-2-pyridinyloxyacetic acid} (H utchinson and Langeland 2006). However, successful management of OWCF requires l ong-term monitoring and multiple follow-up treatments (Hutchinson et al. 2006). Managers of public l ands in Florida have reported that lack of personnel, limited f unding, other duties, site access, and OWCF on adjacent lands as limiting factors in managing the fern and most are hoping for identification of successful biocontrol agents (Hutchinson an d Langeland 2006). The majority of large scale herbicide application for control of OW CF on public lands is conducted by private contractors through funding from the Florida Fish and Wildlife Conservation Commissions Invasive Plant Management Section. Manual (i.e., hand-pulling, cutting) and me chanical control (i.e., mowing) can reduce above ground biomass, but OWCF re-sprouts from rhizomes in 6-7 days. Manual removal is expensive, labor intensiv e, and time consuming (Roberts 1998), in 23

PAGE 24

addition to potentially moving the spores to other non-infested sites. Due to the hydric soils typical of the habitat in which OWCF oc curs mechanical treatment is unfeasible in natural areas due to significant so il disturbance and accessibility. Prescribed fire does not control OWCF (Rober ts 1998), which recovers to its initial cover in 12-24 months (Roberts 1996). There is cursory evidence that prescribed fire can result in OWCF invading a sit e where it was not initially present prior to the fire (Hutchinson et al. 2006). Lipids, present as food reserves in many fern spores, make fern spores more ignitable (Moran 2004), which could explain the intense heat generated as the fire spreads upwards on the rachis mats where most of the fertile leaflets are present. The pyrogenic nature of this plant can easily spread fire, prescribed or natural, into wetland habitats that seldom burn. As a result, OWCF has altered the prescribed burn programs in so me natural areas (Roberts et al. 2006). Following a fire, interspecific competition is decreased and OWCF re-sprouts rapidly and further infests the habitats (Goolsby et al. 2003). However, St ocker et al. (2008) found that in small plots, prescribed fire followed by biannual herbicide treatments resulted in < 1% cover of OWCF after three years. Flooding of OWCF under greenhouse conditions resulted in reduced growth rates, but with a concomitant increase in the number of OW CF fertile leaflets (Gandiaga et al. 2009). Herbicide treatment with glyphosate followed by flooding was successful in controlling OWCF in hydrologicall y restored wetlands (Toth 2009). Multiple biocontrol agents are being ev aluated for control and three species ( Austromusotima camptozonale Floracarus perrepare and Neomusotima conspurcatalis ) have been released in southeastern Florida (Goolsby et al. 2003; 24

PAGE 25

Pemberton 2006). One species, Neomusotima conspurcatalis has shown promise by defoliating 1.4 ha of OWCF among three rel ease sites (Boughton and Pemberton 2009). Herbicides are currentl y the only option av ailable to manage and ha lt the spread of OWCF infestations in Florida. Herbicide control of OWCF has shown some success but OWCF has never been eliminated from a heavily infested site and recovers after a single herbicide application (Thomas and Brandt 2003; Langeland and Link 2006; Stocker et al. 2008). It is unclear if the three most comm only used herbicides (glyphosate, metsulfuron, and triclopyr) for OWCF control are translocating into the rhizomes of OWCF, which in theory, would pr event re-sprouts. No information exists on the effects or tolerance of OWCF spores and gametophytes to herbicides. There are no known studies that have examined the effi cacy of repeated applications of glyphosate and metsulfuron for control of OWCF and their effects on native plants on Everglades tree islands. The objectives of this research were: 1) examine herbicide translocation in OWCF, 2) assess the impact of herbicides on the su rvival of OWCF spores and gametophytes, and 3) evaluate the efficacy of herbicides on OWCF for multiple ground and aerial herbicide applications. Additional research was conducted to determine the impacts of herbicides on native plant richness and diversity for objective 3. 25

PAGE 26

26 0 5 10 15 20 25 30 35 1950's1960's1970's 1980's1990's2000'sDecadeOWCF Herbarium Records Figure 1-1. Number of OWCF herbarium records per decade from the University of Florida (FLAS, 2008), Florida Sate University (FSU, 2008), the University of South Flor ida (Wunderlin and Hansen 2008) and Fairchild Tropical Gardens (FTG, 2008). FTG herbariums records include all herbarium record s from the University of Miami.

PAGE 27

27 natural and cultivated areas. Non-native areas in many parts of the world (Langel ferns are considered weedy in agriculture 1999; LeDuc et al. 2000). Old World clim (Cav.) R. B mesic and hydric natural areas in central Nauman and Austin 1978). This natural areas by land managers in cent Langeland 2006). treatments are required to r and Langeland 2006; Langeland and Li where OWCF has been eradicat achieved with annual treatment (Hutchinson effective in eliminating mature plants, new present in the soil or blown in from nearby coverage is which mak mixes, it took 2 to 3 herbicide applications study plots, but new growth fr CHAPTER 2 ABSORPTION AND TRANSLOCATION OF GLYPHOSATE, METSULFURON, AND TRICLOPYR IN OLD WORLD CLIMBI NG FERN (LYGODIUM MICROPHYLLUM) Introduction Ferns (Class Pteridopsida) both non-native and native, can cause problems in ferns have become problematic in natural and et al. 2008; Wilson 2002) and some native and forestry sites (De la Cretaz and Kelty bing fern [OWCF; Lygodium microphyllum r.] is a highly invasive plant t hat can alter the struct ure and composition of and southern Florida (Brandt and Black 2001; species is now considered one of the worst threats to ral and southern Florida (Hutchinson and Chemical control of OWCF in natural areas can be achieved, but consecutive educe and maintain populations at low levels (Hutchinson nk 2006). Presently, t here are no known cases ed from a natural area, t hough suppression has been et al. 2006). While herbicide treatments are growth will develop from OWCF spores plants. In field sites infested with OWCF, often >75% over several ha and the fern can grow > 15 m into the canopy es treatment difficul t. In field studies using several herbicides and tank every six months to eliminate OWCF from the om spores was still present (J T. Hutchinson and K. A.

PAGE 28

Langeland, unpublished data). In some cases, multiple re -treatments may not be possible due to funding, limit ed personnel, and inaccessibility. Consequently, the fern has proven to be extremely difficult to control once it has become established and begins to reproduce. Management of OWCF with herbicides is conducted year-round in southern Florida by multiple local, state, feder al and private land management agencies. Except when occasional freezing temperatures o ccur, OWCF is an evergreen that grows vigorously throughout the year in sout hern Florida, wher e the average annual temperature is 23 C with < 2 days/year w hen temperatures drop below 0 C (Henry et al. 1994). The most frequently used herbici des for OWCF management in Florida are glyphosate, metsulfuron, and triclopyr (H utchinson and Langeland 2006). Application rates for glyphosate and metsulfuron reported by land managers vary from 7.0 to 17.5 g ai / l and 0.1 to 0.2 g ai / l, respectively, fo r spot treatments to i ndividual infestations. Land managers use triclopyr less frequently to control OWCF in Florida, but reported rates varying from 22.0 to 55.0 g ai / l. Glyphosate and triclopyr have been used to manage OWCF for more than 15 years. Mets ulfuron has only been used for four years since a Special Local Need 24c Label was ob tained for use on OWCF in wet areas of Florida (EPA SLN No. FL-030010). Despite the spread of OWCF across the landscape of central and southern Florida, relatively little research has been conducted to determine the most effective treatment method for control. Stocker et al. (1997) reported that treatments of glyphosate and triclopyr resulted in 100% aboveground mortality of OWCF at rates as low as 1.5% product, but neither herbicide co mpletely eradicated the plant as re-growth 28

PAGE 29

was observed for both treatment s after one year. In the Art hur R. Marshall Loxahatchee National Wildlife Refuge in Palm Beach C ount y, effective control of OWCF has been achieved for > 3 years post-treatment in research plots by cutting the vertically ascending rachis at 1.5 m above ground level and foliar spraying the lower portion with glyphosate at 0.12 kg ai 3.8 l -1 (Thomas and Brandt 2003). OWCF plots treated with foliar sprays of metsulfuron had the same abov e-ground dry weight biomass as control plots at 2.5 years posttreatment (Langeland and Li nk 2006). Glyphosate and metsulfuron have been the most effective herbicides to date to control OWCF, but results are inconsistent with regard to l ong-term management (Hutchinson et al. 2006). Triclopyr is one of the most common herbi cides used in natural areas for management of invasive plants, but it is unknown if this herbicide can be used effectively for control of OWCF. The ability of OWCF to resprout from rh izomes following herbicide treatment is an indication that herbicides have limited trans location to the rhizomes or throughout the entire rhizome of an individual fern. Some ferns have high root to shoot ratios (LeDuc et al. 2003), and roughly 50% of the carbon seq uestered by OWCF is allocated to belowground root mass (John Volin, personnel co mmunication). Application techniques that maximize herbicide translocation to the rhizomes of OWCF should result in greater or complete control of the fern. Studies to determine herbicide uptake and translocation in OWCF should provide a better understanding of managem ent strategies for long-term control. The objective of this research was to determine herbicide upta ke and translocation patterns as a function 29

PAGE 30

of herbicide placement in OWC F for gly phosate, metsulfuron, and triclopyr using 14Clabeled compounds. Materials and Methods Plant Material Experiments were replicated twice using OWCF plants grown from two different sources. For the first experiment, OWCF was grown from rhizomes in Okeelanta muck soil collected form Allapattah Flats, Martin County, Florida. Rhizome sections of 103 cm2 were cut with a machete from mature OW CF, and placed in 0.5-L plastic pots (12.7 cm in depth). Rhizomes were excavated with nat ive soil intact. All fronds (live and dead) from the rhizomes were cut at soil level, and the rhizomes allowed to re-sprout. Plants were grown for 90 days to a height of 1 m prior to treatment. The experiment was repeated with OWCF grown from spores. Soil excavated below mature OWCF plants was placed in shallow 2.5-L tubs and maintained under saturated conditions. Soil was collected from the same site as rhizomes in Martin County, Florida. The spores were a llowed to germinate and form gametophytes; fertilization to sporophytes occurred ther eafter. Approximately 6 months after germination, the sporophytes, 7.6 cm in hei ght, were transferred into 0.5-L pots in Okeelanta muck soil and grown to 1 m in height prior to treatment. Plants were grown in a greenhouse at t he University of Florida campus in Gainesville and exposed to a natural photoper iod (10.5 to 14.0 hours sunlight) under a 50% shade cloth with a temper ature range of 21 to 37 C. The first experiment was conducted from March to June 2005, and the second experiment was conducted from January to October of 2005. No additional nutrient supplements were added to the 30

PAGE 31

ferns over the duration of t he study. Pots were placed in 7.5-L rectangular tubs with water depth maintained at 2 to 3 cm below soil level and watered 3 to 4 times per week. Absorption and Translocation OWCF plants were treated with glyphosate1, metsulfuron2, and triclopyr3 herbicides at rates of 1.67 kg ae ha -1, 0.10 kg ai ha -1, and 1.20 kg ae ha -1, respectively. Sources of material are list ed in Table 2-1. Five different application scenarios were used for each herbicide as follows: 1) plants were cut 6 cm from the base of the plant and the remaining foliage was sprayed (denoted as cut and spray), 2) foliage was treated from 0 to 6 cm above the bas e of the plant (denoted as basal spray), 3) foliage was treated from 0 to 25 cm abov e the base of the pl ant (denoted as 25% foliar), 4) foliage was treat ed 0 to 50 cm above the base of the plant (denoted as 50% foliar), and 5) 100% of the pl ant was treated (denoted as 100% foliar). Herbicides were applied with a CO2 sprayer at 172 kPa with a singl e TeeJet 11003 flat fan nozzle4. All treatments included no n-ionic surfactant5, 0.5 v/v, at a spray volume of 370 L ha-1. Uptake and translocation were determined using 14C-labeled glyphosate6 (specific activity 1998.0 kBq mg-1), metsulfuron7 (specific activity 1845.2 kBq mg-1), and triclopyr8 (specific activity 895.4 kBq mg-1). Immediately after sp raying, application of 5 L droplets 14C-radiolabeled material was made to t he adaxial surface of five separate pinnae or rachis on each plant. Spotting of r adiolabeled material was made at 6 cm (cut and spray, and basal), 25 cm (25% foliar), 50 cm (50% foliar) and 100 cm (100% foliar). This resulted in a total of 12.12 kBq mg-1 (0.33 Ci) 14C glyphosate, 11.84 kBq mg-1 (0.32 Ci) 14C metsulfuron, and 12.95 kBq mg-1 (0.35 Ci) 14C triclopyr per plant. 31

PAGE 32

At nine day s after treatment (DAT), plants (including rhizomes) were removed from pots, radiolabeled leaflets excised and the soil washed from roots using detergent and water. Treated areas on leaflets or rachis were excised and washed in five sequential 1ml aliquots of deionized water applied to eac h radiolabeled spot to determine percent uptake. A total of 25 ml of rinsate (combi ned total for all five spots) was obtained for each treatment. Unabsorbed 14C from the leaf wash solu tion was quantified with liquid scintillation spec trometry (LSS). 9 Whole plants were placed on blotter paper in plant presses and oven-dried at 70 C for 72 hours. Dried plants were covered with X-ray film10 and stored 40 days at room temperature to obtain autoradiographs. Fo llowing autoradiography the plants were sectioned into rhizomes and stolons, lower 50% rachis and leaves, and upper 50% rachis and leaves. Radioactivity was dete rmined by LSS. Translocation was determined by measuring radioactivity in individual pl ant parts. Plant tissues were finely ground through a 0.5 mm screen using a Wiley Mill11 and 0.15 to 0.20 g of ground tissue was oxidized to recover 14C using a Harvey OX-500 biological oxidizer12 with a proprietary 14C-trapping liquid scintillation cocktail13. Absorbed radioactivity is defined as the total 14CkBq obtained from all plant parts and presented as the percent of applied radioactivity for each herbicide Percent recovery is the absorbed radioactivity plus the amount recovered from leaf-rinses. Translo cated radioactivity is reported as the percent of absorbed radioactivity in plant parts (top, bottom, rhizomes) and presented as a percent of absorbed radioactivity. Experimental Design and Data Analysis Treatments were arranged in a complete ly randomized design. Each treatment was replicated five times and the experiment was repeated using plants grown from 32

PAGE 33

rhizomes (n = 25) and spores (n = 25). Ther e was no significant (P > 0.05) treatment by experiment interaction between rhizome and spore grown plant s, therefore data represent the average of both experiments. Total recovery leaf w ash data, absorbed and translocated radioactivity were comb ined among treatment scenarios for each herbicide and subjected to analysis of variance (ANOVA) using Proc GLM (SAS14) procedure, and means were separated using Fi shers protected LSD test at the 5% level of probability. Radioactivity in plant parts was analyz ed by herbicide and treatment scenario separately to determine patterns of trans location for each herbicide. Data were subjected to ANOVA and means were separated using Fishers protected LSD test at the 5% level of probability. Translocation was analyzed for each treatment method for percentage radioactivity in top 50%, bo ttom 50% and rhizomes using an ANOVA with means separated using Fishers protected LSD test at the 5% level of probability. For translocation data with the cut and spray method, a paired t-test using SAS was used to compare bottom 50% and rhizomes at the 5% level of probability. Results and Discussion Visual Effects of Herbicides on OWCF OWCF treated with triclopyr exhibited chlorosis, epinasty, and necrosis nine DAT (days after treatment) on 30-40% of OWCF leaflets for a ll treatments scenarios. There were no observable effects on OWCF treated with glyphosate or me tsulfuron at nine DAT. This may indicate that triclopyr is absor bed or translocated at a faster rate relative to glyphosate or metsulfur on, or the rate at which symptoms of the different mechanisms of actions are expressed are different. In field trials with triclopyr at similar rates, necrosis was observed in < 4 hours in OWCF (J. T. Hutchinson, personal 33

PAGE 34

observation). Rapid burning and tissue necro sis by tri clopyr in OWCF may increase absorption, but could inhibit translocation by limiting entry of the her bicide into vascular tissue. It was discovered that rapid wiltin g and yellowing in purple loosestrife treated with triclopyr stimulated new growth (G ardner and Grue 1996). Hofstra et al. (2006) suggested that lower concentrations of tr iclopyr may result in more effective translocation in parrotsfeather compared to higher rates and prevent basal stem resprouts. Autoradiography The general pattern observed from autoradiographs indicated most of the 14C material remained in the treated leaflets or rachis. For all three herbicides, some basipetal movement was observed for cut and spray and basal spray (Figure 2-1). However, basipetal movement with 100% fo liar spray for any herbicide was minimal with almost no movement of 14C material into the rhizom es. With 25% and 50% foliar spray, herbicide movement was primarily ac ropetal with some basip etal movement for all three herbicides. Basipetal movement into the rhizomes was observed for cut and spray and basal spray treatments of triclopyr and glyphosat e. However, there was no evidence of horizontal movement for triclopyr or gl yphosate along rhizomes in autoradiographs. Basal treatment with triclopyr also indicat ed some acropetal translocation (Figure 1c). Autoradiographs of OWCF treat ed with metsulfuron indicated minimal translocation of the herbicide into the rhizomes. These result s corroborate field observations of all three herbicides in which OWCF is highly suscept ible to herbicide treatment, but re-sprouting occurs within several weeks to months pos t-treatment. Re-sprout s from OWCF treated 34

PAGE 35

with herbicide is likely due to limited translocation into the rhizomes and no horizontal translocation within the rhizomes. Absorption There was no signific ant difference in to tal recovery between treatments or plant part other than the treated l eaf among herbicides (Table 2-2). Total recovery was 67% for metsulfuron, 65% for glyphosate, and 69% fo r triclopyr. However, differences were observed among herbicides in leaf wash with only the lowest percentage, 9%, washed from leaves treated with triclopyr follo wed by 34% for glyphosate and 47% for metsulfuron. Accordingly, differences in absorption were observed among herbicides (Table 2-2). The highest level of absorption wa s observed for triclopyr (60%) followed by glyphosate (31%) and metsulfuron (20%). Total translocation was different for triclopyr (14%) compared to glyphosate (6%) and metsulfuron (5%). Overall, more of the applied triclopyr (87%) was absorbed compared to only 48% of glyphosate and 30% of metsulfuron. Howe ver, 23% of the applied metsulfuron was translocated out of the treat ed leaflets compared to 23% of triclopyr and 20% of glyphosate. This indicates that translocation out of the treated leaflet is a better indicator of the effectiveness of a herbicide in reac hing its target site and blocking specific physiological processes inside cells. There wa s no difference in treatment scenarios for absorption of metsulfuron and tr iclopyr, but there were differences between the cut and spray (38%) and 100% foliar spray (27%) wit h glyphosate (Tables 2-3 to 2-5). Reddy (2000) reported 22% absorption of glyphosate in redvine [ Brunnichia ovata (Walt.) Shinners] and Chachalis and Reddy (2004) reported 20% absorption of glyphosate in trumpetcreeper [Campsis radicans (L.) Seem. ex Bureau]. These values are similar to the 31% absor ption of glyphosate observ ed in OWCF. While the data 35

PAGE 36

indic ates that triclopyr and glyphosate are absorbed in greater amounts than metsulfuron by OWCF, increased rates of her bicide absorption do not necessarily mean better control, but the relationship is speciesspecific (Norsworthy et al. 2001; Chachalis and Reddy 2004). Norsworthy et al. (2001) found that some species have more tolerance for glyphosate even at higher absorption rates. Translocation Radioactivity was detected in all porti ons of treated plants for all herbicide treatments indicating that all herbicides were translocated to some extent (Table 2-3 to 2-5). The majority of radioac tivity remained in treated leaves for all herbicides with only small percentages of the absorbe d radioactivity being detected in other plant parts (top 50%, bottom 50%, and rhizomes). Averaged ov er all treatment methods for each herbicide, more radioactivity remained in the triclopyr (47%) treated leaves compared to metsulfuron (15%) and glyphosate (25%) (Tabl e 2-2). Necrosis at 9 DAT and the large amount of radioactivity remaining in treated leaves indicates that translocation may have been limited with triclopyr. In a review of phytotoxic action on herbicide translocation, it was suggested that herbici des may inhibit phloem translocation and herbicide distribution (Geiger and Bestman 1990). Our result s indicate that this may have been the case in this study where in hibition of normal carbon metabolism and phloem transport occurred in OWCF. Radioactivity was detected in the upper plant portions for all three herbicides (top 50%) when lower portions of plants (basal 25% foliar, 50% foliar) were treated indicating that all three herbicides moved acr opetally (Table 2-3 to 2-4). Radioactivity was also detected in rhizomes in all treatm ent scenarios for all herbicides indicating that all three herbicides moved basip itally. While some significant differences in radioactivity 36

PAGE 37

in plant par ts, other than treated leaves, were observed, these differences were small and radioactivity, for the most part, was evenly distributed among the top 50%, bottom 50%, and rhizomes. The greatest amount of radioactivity move ment into the rhizomes was from the cut and spray and basal applica tions for both triclopyr and glyphosate. This might be explained because the herbicide is applied to older leaves at the base. In many plant species, photoassimilates are tr anslocated from older leaves into sinks (rhizomes, tubers, bulbs, etc.) (Williams 1964b; Khan 1981; Wolf 1993). Veerasekaran et al. (1977) found that asulam applied to ma ture fronds of bracken fern resulted in greater translocation into the rhizom es. In contrast, translocation of 14C glyphosate in pitted morningglory ( Ipomoea lacunosa L.) indicated that herbicide movement did not differ whether the herbicide was applied to the top, middle, or bottom portion of the plant (Koger and Reddy 2005). In addition, when using the cut and spray method to treat OWCF, herbicides can diffuse basipitally in the rhizomes through cuts in the rachis. Translocation of a given herbicide varies among species. Basipetal movement was reported to be 49% for 14C-glyphosate in trumpetcreeper (Chachalis and Reddy 2004), which is substantially greater than t he 4% that we observ ed with OWCF. Similar to observations for OWCF, limited basipetal movement was reported for stoloniferous ground ivy ( Glechoma hederacea L.) treated with 2, 4-D (Kohler et al. 2004) and ivyleaf morningglory [ Ipomoea hederacea (L.) Jacq.] treated with gl yphosate, imazethapyr, or glufosinate (Hoss et al. 2003). These results indicate that translocation into the rhizomes is greater with the cut and spray and basal treatment methods using triclopyr a nd glyphosate where herbicide is concentrated on the older leav es at the base of the plant The cut/spray method is 37

PAGE 38

the most commonly used method for treat ing OWCF growing high up into trees (Hutchinson and Langeland 2006). OWCF has extensive rhizomes, especially in long established populations. It can be assumed that translocation of herbicide th rough the rhizomes to prevent re-sprouting is needed for long-term control. Veerasekaran et al. (1977) reported th at asulam (4.4 to 8.8 kg a.i. ha -1) eliminated 90% of the shallow active frond buds in bracken fern ( Pteridium aquilinum (L.) Kuhn), but only eliminated 52% of the dormant buds on deeper rhizome branches. Translocation of herbici de from the treated leaves may not be an indication of the efficacy of the herbicide (Kalnay and Glenn 2000). This is especially true with low volume, high potency acetolactate synthase inhibiting herbicides such as metsulfuron in which small amounts of herbicide are toxic to plants (Fairbrother and Kapustka 2001). Based on aut oradiographs of OWCF, once herbicide translocates to the rhizomes, there is little or no horizontal translocation along the rhizome. Koger and Reddy (2005) suggested that control of pitt ed morningglory is more likely affected by rate than spray coverage. Reddy (2000) s uggested that higher rates of glyphosate will be required for effective control of redvine due to its deep rooted rootstocks where new sprouts emerge following treatment. Likewise, the use of higher herbicide rates for cut and spray and basal treatments may be requir ed for greater translocation into the rhizomes of OWCF. Conversely, it is also possible that t he herbicides tested may interfere with phloem transport at the junction of t he rachis and rhizome. This c ould limit horizontal movement of herbicides through the rhizome and increa sed rates may not result in greater translocation in to rhizomes. Pakeman et al (2000) reported variable control of bracken 38

PAGE 39

fern with aerial applic ation of asulam in wh ich ca. 75% of the si tes sprayed exhibited bracken recovery. This corresponds with r eports from land managers in Florida who have reported inconsistent results w hen treating OWCF with glyphosate and metsulfuron (Hutchinson et al. 2006). Research has shown that glyphosate inhibits phloem transport in plants 24 to 48 h after treatment (Geiger and Bestman 1990; Kirkwood et al. 2000; Walker and Oliver 2008). These results, which show limited translocation in OWCF may indicate that triclopyr and metsulfuron also interfere with physiological processes that block phloem transport in OWCF. There are no studies of herbicide trans location in OWCF. Other authors have reported limited basipetal movement of herbicides in other vines (Unland et al. 1999; Hoss et al. 2003; Kohler et al. 2004). Unlike in bracken fern, where radiolabeled asulam accumulated in rhizomes (Veerasekar an et al. 1977), I found no observable accumulation of radiolabeled material in the rhizomes with the three herbicides tested in OWCF. Based on autoradiographs, 2,4-D movement into the rh izomes of bracken fern was greatest when the herbicide was applie d to immature fronds (McIntyre 1962). Treatment of immature fronds of OWCF could only be acco mplished where the fern has first invaded or on new sprouts following herbicide treat ment, cutting, or fi re. Main et al. (2006) found that triclopyr disp layed excellent basipetal translocation in Chinese yam ( Dioscorea oppositifolia L.) resulting in the elimination of tubers. In pitt ed morning glory, movement of 14C-glyphosate to the roots was highes t for 100% foliar treatment (Koger and Reddy 2005). Control of OWCF is depe ndent on controlling the rhizomes, and all three herbicides used in this study exhibi ted some basipetal movement into the 39

PAGE 40

rhizomes. Howev er, autoradiographs rev ealed limited movement through the rhizomes indicating that all rachises must be treated to prevent re-sprouts. Limited field research on OWCF is availabl e for the three herbicides tested in this study. The greater effectiveness of the cut and spray treatment for basipetal movement into the rhizomes I found are consistent wit h the results of Thomas and Brandt for field treatment of OWCF (2003). Metsulfuron at rates of 0.04 and 0.08 kg ai ha -1 has been shown to provide control of OWCF fo r 12 months using 100% foliar treatment (Langeland and Link 2006). Based on these results, metsulfuron may be effective for ground treatments using the cut and spray, ba sal, or 100% foliar treatments. Aerial applications of metsulfuron at rates of 0.08 and 0.16 kg ai ha -1 over OWCF resulted in > 90% mortality at 1 year post treatment (J. T. Hutchinson and K. A. Langeland, unpublished data). Stocker et al. (1997) reported that spot treatments with triclopyr (0.13 kg ai 3.8 l -1) foliar sprayed and band-sprayed from the base of OWCF to 1.2 m above ground resulted in 100% browning of OWCF, but re-growth was observed and they suggested limited translocation. Using prescribed fire and/or tr iclopyr, it was found that bimonthly and biannual tr eatments of OWCF were succe ssful in reducing OWCF cover to < 1% but re-sprouts were still doc umented after 3 years (Stocker et al. 2008). The results of this study suggest that cu t and spray and basal treatments with triclopyr should result in greater basipetal movement into the rhizomes. These results revealed limited movement of the herbicides tested through the rhizomes of OWCF, which explains why re-sprouts from treated OWCF are commonly observed in the field < 12 mo post-treat ment. In a greenhouse study on OWCF using twice the labeled rates of metsulfuron (0.16 kg ai) and glyphosate (11.20 kg ai), no re40

PAGE 41

sprouts were observed 52 weeks post-treatm ent, indicating that herbicide rates of metsulfuron and glyphosate may need to be increas ed for initial treatments to effectively control the fern (Hutchin son and Langeland 2008). The current treatment methods (cut and spray and 100% foliar spray) used by applicators in Florida should continue to be used to manage OWCF, but re-treatment will be required. Additional research is needed to determine if basal treatments (i.e., band spraying) are effective under field conditions since our results revealed some her bicide movement to the rhizomes using this method. Glyphosate and metsulfuron ar e the most commonly used herbicides to treat OWCF in Florida, but these results indicate that triclopyr exhibits greater translocation to the rhizomes. Field res earch is needed to determine the efficacy of triclopyr and herbicide combinations on OWCF for both ground and aerial treatments. Research on herbicide rates above those us ed in this project are also needed to determine their efficacy on controlling OWCF. 41

PAGE 42

Table 2-1. Sources of material used in OWCF 14C translocation study. 1 Glyphosate herbicide, Monsanto Agricult ural Company, 800 N. Lindbergh Blvd., St. Louis, MO 63167 (Roundup Ultramax, 50.2 a.i.) 2 Metsulfuron herbicide, DuPont Corporat ion, 1007 Market St., Wilmington, DE 19898 (Escort XP, 60% a.i.) 3 Triclopyr herbicide, SePRO Corporation, 11550 N. Meridian St., Suite 600, Camel, IN 46032 (Renovate 3, 44.4% a.i.) 4 TeeJet MidTech Southeast, P.O. Box 832, Tifton, GA 31793 5 Entry non-ionic surfactant, Monsanto Co rporation, 800 N. Li ndbergh Blvd., St. Louis, MO 63167 6 14C-glyphosate, Amersham Life Scienc es Inc., 3350 N. Ridge Ave., Arlington Heights, IL 6004 7 14C-metsulfuron, gift from DuPont Corpor ation, 1007 Market St ., Wilmington, DE 19898 8 14C-triclopyr, gift from Dow AgroSciences LLC, 9330 Zionsvill e Rd., Indianapolis, IN 46268 9 Packard Tricarb 1600CA Liquid Scintilla tion Analyzer, Packard Instrument Co., 800 Research Parkway, Meriden, CT 06450 10 Kodak X-OMAT XAR-5 film, Sigma-Aldr ich, 3050 Spruce St., St. Louis, MO 63103 11 Wiley Mill. Arthur W. Thomas Com pany, Philadelphia, PA (No model number) 12 R. J. Harvey Biological Oxidizer, Model OX-500, R. J. Harvey Instrument Co., 123 Patterson Street, Hillsdale, NJ 07642 13 R. J. Harvey 14-Carbon Cocktail (pr oprietary blend of biscumene and PPO (diphenyloxazole) in mixed xylenes), R. J. Harvey Instrument Co., 123 Patterson Street, Hillsdale, NJ 07642 14 SAS Institute, 2002-2003, SAS software, Version 9.1, SAS Institute, Cary, NC 57513 42

PAGE 43

43 Table 2-2. Radioactivity (percent of applied) and standard error for total recovery leaf wash, absorbed, treated leaf, and translocation following application of 14C-labeled metsulfuron, glyphosat e, and triclopyr to OWCF (all treatments combined). Total Leaf Treated Herbicide Recovery Wash Absorbed Leaf Translocated % Recovered Metsulfuron 66.7 (4.9)1 46.9 (4.7) 19.8 (0.6) 15.2 (0.9) 4.6 (0.8) Glyphosate 65.2 (4.7) 34.0 (5.5) 31.2 (1.8) 25.0 (1.3) 6.2 (0.4) Triclopyr 69.0 (4.3) 8.7 (1.4) 60.3 (3.4) 46.7 (2.0) 13.6 (1.1) LSD (0.05%) NS 6.2 4.4 4.1 2.9 1 Means of 10 replications followed by standard error. Table 2-3. Percent 14C-metsulfuron absorption, translocation, and distri bution in OWCF for five treatment methods. Values represent the means of 10 replications. % 14C-metsulfuron recovered Movement out of Treated Leaves Leaf Treated Total Top Bottom LSD 1 Treatment Wash Absorbed Leaves Trans located50% 50% Rhizomes (0.05%) Cut/Spray 32.2 19.1 15.8 3. 3 N/A 0.9 2.4 P = 0.03 Basal 41.4 20.5 16.4 4. 1 0.6 1.3 2.2 1.5 25% Foliar 49.4 21.8 16.1 4.7 0.7 2.5 1.5 1.5 50% Foliar 51.6 17.9 10.3 7.6 3.2 2.4 2.0 NS 100% Foliar 60.1 20.7 17. 3 3.4 0.6 1.3 1.5 NS LSD (0.05%) 14.4 NS 5. 8 3.1 1.7 NS NS 1 Significance test comparing % DPM in top 50%, bottom 50% and rhizomes within rows. For cut/spray treatment, significance test was a paired t-test (P < 0.05).

PAGE 44

Table 2-4. Percent 14C-glyphosate absorption, translocati on, and distribution in OWCF for fi ve treatment methods. Values represent the means of 10 replications. % 14C-glyphosate recovered Movement out of Treated Leaves Leaf Treated Top Bottom LSD 1 Treatment Wash Absorbed Leaves Trans located50% 50% Rhizomes (0.05%) Cut/Spray 15.4 37.5 31.3 6. 2 N/A 1.8 4.4 P = 0.06 Basal 30.4 28.8 22.5 6. 3 1.4 1.8 3.1 NS 25% Foliar 48.4 32.2 25.0 7.2 1.7 3.3 2.2 NS 50% Foliar 38.2 30.6 24.5 6.1 2.8 1.1 2.2 NS 100% Foliar 37.4 26.7 21. 7 5.0 0.4 1.3 3.3 1.4 LSD (0.05%) 19.1 10.1 8. 0 NS 1.8 NS 1.8 1 Significance test comparing % DPM in top 50%, bottom 50% and rhizomes within rows. For cut/spray treatment, significance test was a paired t-test (P < 0.05). Table 2-5. Percent 14C-triclopyr absorption, translocati on, and distribution in Old World climbing fern for five treatment methods. Values represent t he means of 10 replications. % 14C-triclopyr recovered Movement out of Treated Leaves Leaf Treated Top Bottom LSD 1 Treatment Wash Absorbed Leaves Trans located50% 50% Rhizomes (0.05%) Cut/Spray 5.8 51.1 38.4 12. 7 N/A 4.6 8.1 P = 0.003 Basal 6.9 53.3 36.2 17. 1 2.8 7.0 7.3 4.1 25% Foliar 11.1 64.6 52.0 12.6 4.1 5.9 1.9 NS 50% Foliar 12.8 64.3 49.4 14.9 5.5 6.3 1.5 NS 100% Foliar 6.8 68.3 57. 7 10.6 4.2 9.7 0.7 NS LSD (0.05%) 5.9 NS 11.1 NS NS NS 3.3 1 Significance test comparing % DPM in top 50%, bottom 50% and rhizomes within rows. For cut/spray treatment, significance test was a paired t-test (P < 0.05). 44

PAGE 45

45 a) b) c) Figure 2-1. Dried plant s (left) and autoradiographs (right) of OWCF treated basally with a) metsulfuron, b) glyphosate, and c) tric lopyr. Arrows on left indicate site of treatment and arrows on right indicate spotted pinnae.

PAGE 46

CHA PTER 3 TOLERANCE OF OLD WORLD CLIMBING FERN SPORES AND GAMETOPHYTES TO HERBICIDE EXPOSURE Introduction Few studies have used spores and gametophytes of ferns in herbicide screening trials. Spores are single-celled, without an embryo, and have limited food reserves compare to seeds which are comprised of 1000s of cells, contain an embryo and greater food reserves (Moran 2004). In addi tion, spores germinate into gametophytes which require moisture for fertilization prio r to growing into sporophytes while seeds germinate directly into a new sporophyte generation. No previous studies have examined t he effects of herbicides on Old World climbing fern ( Lygodium microphyllum ; OWCF) gametophytes and spores. Gametophytes of Lygodium japonicum exposed to 2,4-D at concentrations > 10.0 mg / l exhibited decreased growth rates and rema ined as filaments (Swami and Raghavan 1980). Gametophytes have been used to study the physiological basis of tolerance in the paraquat-tolerant fern, Ceratopteris richardii Brongn. (Carroll et al. 1988). 2,4-D was shown to affect gametophyte growth, cell size, antheridia number, and rhizoid development in Ceratopteris thalictroides (L.) Brongn. (Hickok and Kiriluk 1984). Gametophytes of Pteridium aquilinum [L. (Kuhn)] exposed to asulam at 100 g / l suffered 100% mortality, and exhibited lowe r levels of mortality with decreasing concentrations (Keary et al. 2000). Spores of Onoclea sensibilis treated with colchicine, a microtubule inhibitor, resulted in single ce ll division without further growth (Vogelmann et al., 1981). High propagule production co mbined with a high germination rate may accelerate herbicide resistance in ferns because susce ptible plants are selected against, leaving 46

PAGE 47

resistant genotypes in the population (Prather et al. 2000). Volin et al. (2004) estimated that single fertile pinnae of OWCF could produce 28,600 spores. Germination rates of OWCF spores on natural soils have been repor ted to range 84 97 % (Call et al. 2007). Spore dispersal in OWCF has been recorded to be 724 spores / m3 / hour which explain how the fern has rapidly invaded natural ar eas of south and central Florida (Pemberton and Ferriter 1998). Spores and gametophytes are easily cultured in the laboratory and are genetically identical to sporophytes (Voeller 1971; Keary et al. 2000; Fernandez and Revilla 2003). The use of gametophytes and spores could prov ide insight into the tolerance level of OWCF to various herbicides because mult iple populations and different ages can be tested under controlled conditions. Detection of tolerance levels in which OWCF spores and gametophytes survive treatment could be used to set a minimu m rate for field herbicide application. In 2003, metsulfuron received a special local need 24 (c) registration for application to treat OWCF (Dupont Corporation 2003). Since 2003, metsulfuron has been used to treat OWCF with aerial and gro und-based applications (Hutchinson et al. 2006). The acetolactate synthase (ALS) inhibi ting herbicide class, sulfonylurea, which includes metsulfuron, inhibits branc hed chain amino acids and was introduced commercially in 1982. Since 1982, more specie s have become resistant to this class of herbicides than any other class of herbi cides, with some weedy plants developing resistance in < 5 years (Tranel and Wright 2002). Resistance to herbicides with an ALS mode of action occurs through the selection of a biotype with an amino acid substitution of the ALS enzyme rendering the herbicide i neffective. The ALS gene is nuclear 47

PAGE 48

inherited being transmitted in pollen, spor es, and seeds (Tranel and Wright 2002), so OWCF spores and gametophytes could be used fo r evaluation of herbi cide tolerance. With the heavy relianc e on metsulfuron, coupled with the known problems with resistance and the tremendous reproductive capac ity of OWCF, the objectives of this research were to assess the susceptibi lity and/or tolerance level of OWCF gametophytes to metsulfuron, and OWCF spor es to metsulfuron and other herbicides with different modes of actions. Materials and Methods Source of Spores Mature fertile fronds of OWCF were collected from 12 locations in natural areas of central and southern Flor ida from populations that had never been treated with herbicide (Table 3-1; Figure 3-1). Fronds we re collected from > 6 OWCF plants at each site and combined for one composite sample per site. Fronds were placed in open plastic bags and allowed to air dry at ambient temperature. Age of the spores ranged from 1 to 380 weeks for experiment 1, 20 to 29 weeks for experiment 2, and 4 to 432 weeks for experiment 3 (Table 3-2). Spore age, in weeks, is the time elapsed between collection of the fertile leaf lets from live plants and the in itiation of each experiment and varies among experiments and the collection time of spores. Ranker and Houston (2002) suggested using spores of different a ge classes would be more reflective of a natural age structure. The use of OWCF spores from different populations and various ages should provide a more comparative view of the response of OWCF spores to metsulfuron under natural conditions. 48

PAGE 49

Spore Preparation Spores from each location were isolated from leaf material by pas sing through a 63 m sieve (Hubbard Scientific Co., Northbrook, I ll.; Part # 3070-6). After sieving, 0.02 (SE=0.001) g portions of spores were wei ghed into pre-weighed 1.5 ml Thumbs-Up Microtubes (Diversified Biotec h, Boston, MA) on a Mettler analytical balance (MettlerToledo, Inc., Columbus, OH; Part # AE 100). One ml of 1% bleach solution in distilled water was pipetted into each microtube cont aining the spores. The microtubes and spores were vortexed for 2 minutes and then centrifuged for 2 minutes at 1000 rpm. The liquid was decanted and the pr ocess repeated once. After spores were surfaced sterilized, the liquid was decanted and 1 ml of distilled water was added to the microtubes. Spore Counts Spores were counted by vortexing the microtube containing the spores in 1 ml of distilled water for 3 minutes to fully suspend the spores in the solution. Immediately after the spore suspensi on was vortexed, a 0.1 l solution was pipetted into each chamber of a hemacytomer (Hausser Scientific, Horsham, PA.; Bright Line Counting Chamber, Part # 1490). Spores were counted in 10 1 mm2 grids of the hemacytomer under a microscope at 4x power and averaged to estimate spore density. The mean number of spores was multiplied by 10,000 to estimate the number of spores per 0.02 grams, which equated to approxim ately 175,000 (SE = 10,200) spores. Germination Medium A stock solution of modified Parker / Thompsons basal nutrient medium containing macroand micronutrients, and a modified iron solution (Klekowski, 1969; Hickok et al., 1995) was prepared in distilled water based on the methods of Lott et al. 49

PAGE 50

(2003). The nutrient solution wa s titrated to a pH of 6.0 us ing concentrated hydrochloric acid and placed on a heated stir plate. For experiment 3, metsulfuron was added to the solution for a concentration of 0.4 g a.i. / l (equivalent to a field application rate of rate of 0.08 kg a.i. / 187 l). Ten grams (1% w/v) of agar were added to the nutrient solution and heated at 125 C on a stir plate until the agar and nutrients were clear and in solution. The agar / nutrient solution was then autoclaved for 30 minutes. Within 15 minutes of autoclaving, 100 Petri dishes containing the agar / nutrient solution were prepared under sterile conditions using a Laminar flow hood by pipetting 10 ml of agar / nutrient solution into each dish. The dishes were allowed to solidify over 3 hours and subjected to 1 hour of ultraviolet light for further sterilization. Petri dishes were sealed in sterilized plastic bags and st ored at 4 C until inoculation with spores. Spore Inoculation Under a laminar flow hood, 1 ml microtubes containing distilled water and 0.02 grams of spores were suspended in solution for two minutes using a vortex mixer. Immediately after mixing, a 0.2 ml aliquot suspension of spor es was pipetted into a Petri dish. Each dish was slowly swirled to distribute the spores on the agar medium. The remainder of the microtube was vortexed again and pipetted onto 4 additional dishes. Approximately 35,000 spores were pipetted into each Petri dish with the number of replications varying among experiments as described below. Each Petri dish was sealed in parafilm. The Petri dishes were placed under cool-white fluorescent light (80 mol/m2/s1) for a photoperiod of 16/8 hours with a mean temperature of 25.0 (SE = 0.02) C. Temperature was recorded at 30 minute intervals using a Hobo-Temp datalogger (Onset Computer Co rporation, Bourne, MA). 50

PAGE 51

Germination Rates Germinated spores were counted 15, 30 and/or 45 days after inoculation under a microscope (Fisher Scientific Micromaster II Video Microscope) at 4x magnific ation depending on the experiment. Germination ra tes were calculated by dividing the number of germinated spores by the total number of spores within 20 5 mm2 reticule grids (Fisher Scientific, Cat. # 12-561-RG1) and converting to percent. Spores were counted as germinated if the pr imary prothallia cell had em erged from the spore. The sexuality of gametophytes was not quantified, but the majority of all gametophytes were male (filamentous, plate-like, and amerisit c) while <5% were females containing a meristem notch. Sporophyte Development Sixty days post-treatment and a total of 90 days, all sporophytes that developed from treated gametophytes were transferred into 0.15 liter pots containing commercially purchased humus. Sporophytes were scraped from the agar with a spatula and placed in a small depression (2 cm wide x 0.05 cm deep) in the humus. A small amount of loose rock-wool was incorporated into the humus to raise sporophytes 0.05 cm above humus level. At 60 days following the potting of sporophytes in humus, all sporophytes were counted and then cut at soil level. Percent survival of sporophytes for each treatment was determined by dividing the num ber of live sporophytes per treatment by the number of gametophytes that survived each treatment. All cut OWCF plant material was placed in paper bags and dried at 90 C fo r seven days to determine dry weight biomass. 51

PAGE 52

Experiment 1 Tolera nce of Gametophytes to Metsulfuron Thirty days after being placed on agar, gametophytes were exposed to metsulfuron at concentrations of 27, 54, 108, 216, 432, and 864 mg a.i. / l (equivalent to field application concentrations of 0.005, 0.10, 0.20, 0.40, 0.80, and 0.16 kg a.i. / 187 l) or 0.5% v/v surfactant (Dy neAmic, Helena Chemical Co.). Surfactant was not added to metsulfuron treatments. Untreated checks contained no metsulfuron or surfactant. Metsulfuron was added to each Petri dish under sterile conditions by pipetting 1.0 ml aliquots of herbicide solutions into the Petri dishes. This volume ensured that all gametophytes were completely exposed to the herbicides. Germination rates were determined by dividing the number of gam etophytes by the number of spores and expressing as percent. Each treatment was replicated eight times and the experiment was repeated four times for 32 samples per treatment. Germination rates were determined separ ately for age group (<10, 11-20, 21-40, 70-100, 101-200, and >300 weeks) and loca tion 30 days prior to treatment. Germination rate was determined for each treatment (includi ng surfactant and untreated checks) prior to treatment (30 days) and 30 days post-treatment. The site with spores >300 weeks old (DuPuis Reserve) was elimi nated from further analysis due to low mean germination rate (1.5%). S porophyte survival was determined by counting the number of individual sporophytes that developed in potted humus at 120 days post-treatment. Individual sporophytes were cut at soil leve l 120 days post-treatment to determine dryweight. All data were subjected to analysis of variance (ANOVA) using Proc GLM procedure, and means were separated using Fi shers protected LSD test at the 5% level of probability (SAS, 2002). Data from all four expe riments were combined and 52

PAGE 53

separated by herbicide because there was no treatment by experiment interaction (P > 0.05). Non-linear regression (Proc NLIN; SAS 2002 ) was used to calculate a doseresponse curve for each treatment rate based on gametophyte survival 30 days posttreatment and sporophyte survival and dr y weight 120 days post-treatment. The exponential decay equation [y = ae -(bx)] was used for non-linear regression analysis. Inhibition concentrations (I50 and I95) were calculated for gametophyte survival at 30 days post-treatment and spor ophyte survival at 120 days post-treatment. Growth reduction concentrations (GR50 and GR95) were calculated for sporophyte dry weight biomass 120 days post-treatment. Experiment 2 Tolerance of Spores to Selected Herbicides OWCF spores were placed for 1, 3, 6, 12 and 24 hours in 1 ml aliquots of solutions containing one of five concentrations of six herb icides (Table 3-3) or 0.5% v/v surfactant (DyneAmic, Helena Chemical Co.). Un treated checks contained no herbicide or surfactant. Each treatment was replicat ed two times and the experiment was repeated three times for six samples per treatment. Following exposure, the herbicide solution was decanted and the spores washed with distill ed water three times. Treated spores were placed on an agar/nutrient media and monitored for germination after 30 days. Germination rates were determined after 30 days and means, standard error, and 95% confidence intervals were calculated. Germination rates were determined by dividing the number of gametophytes by the number of spores and expressed as percents. Data for exposure time from each experim ent was combined because there were no differences among exposure times or treatment by time interaction for each herbicide (P > 0.05). 53

PAGE 54

Non-linear regression (Proc NLIN; SAS 2002 ) was used to calculate a doseresponse curve for each herbicide based on percent germination at 30 days posttreatment (SAS, 2002). The exponential decay equation [y = ae -(bx)] was used for nonlinear regression analysis. I nhibition concentrations for I50 and I95 were calculated for percent spore germination at 30 days pos t-treatment for each herbicide. Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium Suspensions of OWCF spores were pipetted onto an agar/nutrient media containing 0.4 g a.i. / l metsulfu ron (equivalent to a field applicat ion rate of 0.08 kg a.i. / 187 l / ha). In the first expe riment, spores from 12 locati ons that varied in age were combined and tested (Table 3-2). Each treat ment was replicated eight times with an equal number of untreated checks and the exper iment was conducted twice for a total of 16 samples. In the second experiment, spores from two locations (IRCC and JDSP) were placed on agar/nutrient media (untreat ed check) or agar/nutrient media in which metsulfuron was incorporated to test indivi dual populations of the same age. The ages of spores were 46 and 36 weeks for the IRCC and JDSP sites, respectively. Each treatment was replicated eight times with an equal number of untreated checks and the experiment was conducted twice for a total of 16 samples per location. Percent spore germination was count ed after 15 and 45 days for the first experiment, and at 45 days fo r the second and third experiments. Differences between untreated checks and agar/metsulfuron samples we re analyzed with a paired t-test at P < 0.05 (SAS 2002). 54

PAGE 55

Results Experiment 1 Tolera nce of Gametophytes to Metsulfuron Germination of OWCF spores 300 weeks of age was lower than all other age groups (P = 0.0005) but there was no diffe rence in germination rates among other age groups (Figure 3-2). Germination rates were not different among the more southerly locations with a trend of lower germination in more northerly collection sites (Figure 33). There was no difference in germination rates of OWCF spores 30 days pretreatment (Table 3-4). Thirty days post-treat ment (60 total days on agar), survival was reduced to 1.4% by all metsulfuron concentrati ons or surfactant compared to 36.6% survival of untreated checks (Table 3-4). Exposure to metsulfuron greatly reduced gametophyte survival (Figure 3-4). Inhibiti on concentrations were 6.1 and 26.4 mg a.i. / l for the I50 and I95 values, respectively. The survival rate of OWCF gametophytes after 30 days exposure to metsulfuron at 27 mg a.i. / l indicates that 14,000 out of 1 million gametophytes survive treatment, while the survival rate of OWCF gametophytes treated with 432 mg a.i. / l of metsulfuron drops to 4,000 out of 1 million. Survival of sporophytes treated with metsulfuron or surfactant only 120 days post treatment was lower compar ed to untreated sporophytes (Table 3-4). No OWCF sporophytes developed following any treatment with metsulfuron at concentrations 432 mg a.i. / l metsulfuron. Sporophyte development in su rfactant treatments exhibited higher survival rates than all metsulfuron tr eatments suggesting that the herbicide has an added effect. Sporophyte survival dec lined with increasing concentrations of metsulfuron (Figure 3-5). Inhibition concentrations for OWCF sporophytes treated with metsulfuron were 5.6 and 24. 1 mg a.i. / l for the I50 and I95 values, respectively. 55

PAGE 56

Dry weight biomass of OWCF sporophyte s treated with herbicide or surfactant only 120 days post treatment was lower com pared to untreated checks (Table 3-4). Sporophytes treated with surfactant exhibited higher dry weight biomass compared to all concentrations of metsulfuron. Simila r to sporophyte survival, sporophyte weight declined at increasing concentrations of mets ulfuron (Figure 3-6). Growth reduction concentrations of OWCF dry weight biom ass treated with metsulfuron were 7.7 and 33.1 mg a.i / l for the I50 and I95 values, respectively. Based on the total number of gametophytes treated, the potential for a treated gametophyte to develop into a sporophyte would be 140 out of 1 million and 20 out of 1 million fo r metsulfuron at concentrations of 27 and 216 mg a.i. / l, respectively. Experiment 2 Tolerance of Spores to Selected Herbicides OWCF spores exposed to all concentrati ons of metsulfuron had lower germination rates compared spores exposed to the other five herbicides tested in inhibiting OWCF spore germination at 30 days (Figures 3-7 to 3-12). Germination was 0.4% for spores exposed to 0.1 g a.i. / l metsulfuron and germination was completely inhibited at higher concentrations. There was no difference between the germination rates of spores exposed to surfactant, the lowest two concen trations of asulam and glyphosate, and the lowest rate of fluroxypyr compared to unt reated checks. The average germination rate in untreated checks and surfactant was 47.9% (SE = 4.6) and 44.7% (SE = 5.8), respectively. Reduction in spore germination was observed with all other concentrations tested, ranging from 10.4% with triclopyr (40.0 g a.i. / l) to 42.6% with asulam (4.2 g a.i. / l). Spores treated with herbicides other than metsulfuron transitioned into two-dimens ional gametophytes. Gameto phytes that developed from spores treated with 0.1 g a. i. / l metsulfuron remained filamentous after 30 days. 56

PAGE 57

Spores were highly sensitive to metsulfuron with I50 and I95 concentrations of 0.014 and 0.063 g a.i. / l, respectively, and > 1000 -fold more sensitive to metsulfuron compared to I95 concentrations of any other herbici de tested (Table 3-5). Germination declined exponentially with increasing concentrations of metsulfuron (Figure 3-7). The five other herbicides tested resulted in a gradual decrease in spore germination with increasing concentration (Figures 3-8 to 312), but no herbicide completely inhibited germination. The I95 concentration (69.7 g a. i. / l) of imazapyr, which is another ALS inhibitor, was > 1100-fold great er than metsulfuron. The I95 concentrations for glyphosate, imazapyr, tricl opyr, fluroxypyr, and asulam are higher than the maximum label rate of these herbicides. Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium Germination of OWCF spores from 12 locations of varying age exposed for 15 days on agar with metsulfuron (0.4 g a.i. / l) was not different than those on untreated agar (P=0.85, Figure 3-13). After 45 days of exposure, survival of gametophytes was 76% lower on agar with metsulfuron compared to those on untreated agar (P=0.0001). The length and width of OWCF gametophyte s grown on untreated checks were 3.5 x and 4.0 x greater in length and width, respectively, than germinated OWCF spores on agar + metsulfuron at 45 days post-expos ure (data not shown). Gametophyte development on untreated checks exhibited tw o-dimensional shapes, while those grown on agar + metsulfuron remained filamentous or plate-like. Germination rates of OWCF spores colle cted from IRCC of the same age were 18% higher for spores germi nating on untreated agar compar ed to spores growing on agar + metsulfuron medium (P=0.001, Figu re 3-14). Germination rates of spores collected from JDSP of the same age were 43% higher for spores germinating on 57

PAGE 58

untreated checks relative to spore s growi ng on agar + metsulfuron medium (P=0.0002, Figure 3-14). Discussion Experiment 1 Tolerance of Gametophytes to Metsulfuron Findings of this study indicate that 30 day old OWCF gametophytes are highly susceptible to metsulfuron at concentrations 27 mg a.i. / l. Si milar results of high mortality have been reported for fern gametophytes exposed to asulam and 2,4-D (Hickok and Kiriluk 1984; Keary et al. 2000) Thirty days post-treatment, gametophytes exhibited some tolerance, 0.3 to 1.2 % surviv al, with all concentrations of metsulfuron. The surfactant used in this study acted as a contact herbicide with hi gh initial mortality to gametophytes, but did not affect subs equent spore germination. No new spore germination was observed at 30 days post-tr eatment for all c oncentrations of metsulfuron. Sporophyte development 120 days following treatment of gametophytes was 0.014% for all concentrations of metsul furon. No sporophytes developed from gametophytes treated with metsul furon at concentrations 432 mg a.i. / l, indicating that these concentrations may be the most effectiv e in inhibiting the ALS enzyme in mature OWCF. This concentration (432 mg a.i. / l) is lower than the most common concentration (450 mg a.i./l) of metsulfuron used for aerial treatment of OWCF in Florida, but higher than the mo st common concentration (90 mg a.i./l) of metsulfuron used for ground treatments (Hutchinson et al. 2006). Volin et al. (2004) estimat ed that approximately 3 x 1010 OWCF spores could be produced in heavily infested sites 1 ha if 1% of OWCF had fertile leaflets. Based on aerial estimates, approximately 43,000 ha in southern Florida are infested with OWCF (Ferriter and Pernas 2006). Theoretically, if 10% of 43,000 ha are heavily infested with 58

PAGE 59

OWCF and the entire area is tr eated with 216 mg a.i. / l me tsulfuron, the number of tolerant sporophytes that would de velop would be approximately 4 x 108 based on an average germination rate of 37% with 0.009 % of these gametophytes developing into sporophytes. This indicates high potentia l for OWCF to develop tolerance to metsulfuron at concentrations 216 mg a.i. / l. In this study, the tolerance rate of metsulfuron treated OWCF gametophytes that developed into sporophytes was 2.0 x 10-5 treated with 216 mg a.i. / l. This toler ance is greater than the estimated resistance rate of Arabidopsis thaliana at 1 x 10-9 (Haughn and Somerville 1987) and Nicotiana tabacum at 2.7 x 10-8 (Harms and DiMaio 1991), but about equal to Lolium rigidum at 2.2 x 10-5 to 1.2 x 10-4 (Preston and Powles 2002) for other ALS herbicides. Sulfonylurea herbicides, such as metsulfuron, act at the cellular level by inhibiting acetolactate synthase, the first enzyme in t he biosynthesis of the branched chain amino acids valine, leucine, and isoleucine (Ray 1986). The effectiveness of metsulfuron on OWCF gametophytes would be due to the conc entration within the cells regardless of the amount of active ingredient used. In bioassays of OWCF sporophytes, 0.06 mg / ml of the ALS enzyme was extracted from leaf tissue and 100 g / l of metsulfuron was needed to inhibit the enzyme in OWCF (Atul Pu ri and Jeffrey Hutchinson, unpubl. data). The amount of ALS enzyme in OWCF appears to be low; thus, only small amounts of metsulfuron would be needed to inhibit plant growth. Herbicide resistance for any plant c an only be validated by treating second generation plants in pot bioassays (Beckie et al. 2000). None of the surviving OWCF sporophytes treated with metsulfuron developed sporangium 1.5 y ears post-treatment from re-sprouts following cu tting (for dry weight biom ass determinations) 120 days 59

PAGE 60

following th e initial treatment. This may indicate that metsulfuron inhibits normal development and sporangium formation in OWCF sporophytes after gametophytes have been exposed to the herbicide. Herbicides have been reported to alter phenology and reproduction traits in plants without effect ing leaf area or biomass (Crone et al. 2009; Rice et al. 1997). The results of this study indicate that a small percentage of OWCF may possess an ALS gene that imparts tolerance to me tsulfuron at concentrations under 432 mg a.i./l. Under field conditions, in which the same population of OWCF is treated multiple times with metsulfuron, this could result in resistant populations. T he use of sub-lethal concentrations of herbicides can provide inc onsistent control and select for metabolic resistance in plants (Doyle and Stypa 2004). Minimum treatment concentrations of OWCF with metsulfuron should be 432 mg a.i. / l and herbici des with other modes of action should be used in rotation with metsulfuron. Experiment 2 Tolerance of Spores to Selected Herbicides No herbicide except metsulfuron comple tely inhibited OWCF spore germination suggesting that metsulfuron translocated th rough the spore wall. It is unclear how metsulfuron would affect spore germinati on during field application. The herbicide concentrations used in this study are cons iderably greater than would be expected to contact OWCF spores in a natural setting a ssuming that only a small percentage of the applied herbicide translocates into the spores. However, if metsul furon is translocated into the tapetum of t he sporangium of fertile leaflets, the herbicide could inhibit spore development and result in non-viable s pores. Concentrations of metsulfuron 63 mg a.i. / l would have to translocate into the fertile leaflets to inhibit 95% of OWCF spores. 60

PAGE 61

Herbicide translocation was limited in OW CF with the majority of the herbicide remaining at the spot of application (Chapter 2). Treatment of OWCF with radiolabeled metsulfuron at a concentration of 270 mg a.i. / l (0.10 kg a.i. / 370 l) indicated that 17.3%, or approximately 46.7 mg a.i. / l, of the applied herbic ide remains at the spot of application. This concentration is less than t he concentration of 63 mg a.i. / l required for 95% spore inhibition At the standard aerial treatment rate of 450 mg a.i. / l (0.084 kg a.i. / 76 l) for metsulfuron on OWCF in nat ural areas, approximately 78 mg a.i. / l of metsulfuron would remain in the upper foliage w here the majority of fertile leaflets are present. For standard ground treatm ents of OWCF using 90 mg a. i. / l (0.084 kg a.i. / 379 l) of metsulfuron, approximat ely 15.5 mg a.i. would remain at the spot of application which is less than the 63 mg a. i. / l required for 95% spore i nhibition. Theoretically, the standard amount of metsulfur on foliar applied to OWCF during aerial treatments in natural areas could inhibit 95% of OWCF spores, while the standard ground treatments are unlikely to affect OWCF spore germinati on. It is unlikely that standard herbicide application of the other herbicides tested in this study will inhibit germination of OWCF spores. Growth and development of seedlings following germination of seeds exposed to herbicides was inhibited for some pastu re and turf weeds (Egley and Williams 1978; Young et al. 1984). These results may ex plain was why lower OWCF cover was observed on tree islands treated with metsulfu ron compared to glyphosate (Chapter 4). The differential susceptibil ity of OWCF spores to different metsulfuron concentrations was not seen in the other five herbicides. The spore walls of ferns are composed of fatty acids and cellulose (Banks 1999; Pettitt 1979), and fern spores have a low affinity to water and hydrate slowly (B allesteros and Walters 2007). It is possible 61

PAGE 62

that metsul furon is capable of penetrati ng the spore wall of OWCF spores and preventing germination, while the other herbicides tested do not penetrate the spore wall, even after 24 hours ex posure. Germination rates of OWCF spores with metsulfuron were < 0.5% at 0.1 g a.i. / l and 0.0% at higher concentrations, while germination rates of OWCF spores exposed to imazapyr, also an ALS inhibitor, were 19% with highest rate (24 g a.i. / l) tested. Different responses of tall morningglory germination rates with herbicides of similar chemical st ructures have been reported (Cole and Coats 1973). These differences between the germination rates of OWCF with two ALS inhibiting herbicides may also be due to differences in the inert ingredients associated with the metsulfuron-containi ng product used (Escort XP) that allow metsulfuron to permeate the cell wall. Experiment 3 OWCF Spore Germination on Metsulfuron / Agar Medium OWCF spores grown on agar medium containing metsulfuron had lower germination and growth rates than spores grown on untreated checks. These results suggest that the residual activity of metsulfuron impacts gametophyte development, growth and survival. Under greenhouse co nditions, the growth rate of new OWCF sporophytes that developed under OWCF foliar tr eated with metsulfuron at 0.8 g a.i. / l exhibited slower growth rates from 15 to 35 weeks post-treat ment compared to sporophyte growth rates under OWCF clim bing fern treated with 50 g a.i. / l of glyphosate (Hutchinson and Langeland 2008). Sulfometuron, an ALS inhibiting herbicide, was reported to inhibit spore germination of Dennstaedtia punctilobula when applied to soils inoculated with spores at co ncentrations ranging from 30 to 210 g a.i. (Groninger and McCormick 1991). 62

PAGE 63

The average half-life of metsulfuron is 30 days in soil (Vencill, 2002). Based on the concentration of 0.4 g a.i. / l metsulfuron incorporat ed into the agar medium, the amount of herbicide present in the agar would be 0.025 g a.i. / l at 120 days. This concentration is slightly lower than the I95 value for gametophyte and sporophyte survival in experiment 1, but 2.5-fold lower than the I95 value for spore germination in experiment 2. Control of new OWCF growth from spore germination could be achieved for 120 days under ideal conditions in which the half-life of metsulfuron is 30 days. At germination, a single cell rhizoid is formed which acts to anchor and supply nutrients to developing gametophytes (Banks 1999). The initial de velopment of the single cell rhizoid could absor b small amounts of metsulfu ron and result in reduced growth or eventual mortality to gametophytes. Metsulfuron at concentrations of 26.4 mg a.i. / l inhibited OWCF gametophyte survival by 95%. OWCF spores present in the soil are unlikely to be inhibited from germination by metsulfuron, but recently germinated spores and young sporophytes could be inhibi ted by the presence of small amounts of herbicide present in the soil. Metsul furon could also be absorbed by expanding rhizomes and roots as OWCF sporophyte s mature, further affecting growth. Invasion of OWCF occurs primarily on hydric soils in Flori da with environmental conditions of high temperature, high humidity, and soil moisture during the summer months. The persistence of metsulfuron ba sed on its half-life in the soil decreases under these conditions (Vencill 2002). Based on a hypothesized half-life of 15 days for metsulfuron under high temperatures and moist soils, the half-life of the herbicide would be 25 mg a.i. / l 60 days post-treatmen t, an amount capable inhibiting gametophyte survival up to 95%. The persistence of mets ulfuron in the soil in Florida may be longer 63

PAGE 64

during the winter months under conditions of lower soil moisture, temperatures and humidity. This may explain why OWCF trea ted with metsulfuron results in les s new growth for several months following treatment compared to glyphosate. Glyphosate is rapidly adsorbed to soil particles and has no soil residual activity (Vencill 2002). It is unclear how OWCF spores in the mesic and hydric soils of Florida are affected by aging, climate, water level, and microbes. In experim ent 1, OWCF spores maintained under ambient temperatures in open plastic bags in the lab exhibited reduced germination for spores > 200 weeks of age. Conway (1953) found that spores of Pteridium aquilinum buried under soil for soil for 14 months and then sown on agar had germination rates of 19.7% while spores kept dry for 14 months and then sown on agar had germination rates of 65.5%. OW CF spores exposed to environmental conditions may exhibit a decrease in viabi lity over time, and possibly become more susceptible to lower concentrations of residual metsulfuron in the soil as they germinate. Thus, the concentration of metsulfuron used should influence the impact of the treatment, and possibly pre-em ergent and post-emergent effects on OWCF spores, gametophytes and sporophytes. The results of this study suggest that the most effective control will be achieved using metsulfuron at 432 mg a.i./ l and may prolong resistance of OWCF to metsulfuron. 64

PAGE 65

65 Table 3-1. Site, county, managing agency and geographical pre-positionin g system (GPS) coordinates of locations where OWCF spores were collected. Managing GPS Site County Agency 1 Coordinates Allapattah Flats Martin SFWMD N27 08.6, W080 32.0 Balm Scrub Hillsboroug h HCPRCD N27 44.8, W082 17.0 Blue Cypress Conservation Area Indian River SJRWMD N27 40.9, W080 44.2 DuPuis Reserve Martin SFWMD N26 53.5, W080 29.8 East Orlando Water Facility Orange Orange County N28 30.0, W081 11.2 Fisheating Creek W.M.A. Glades FFW CC N27 01.7, W081 26.1 Highlands Hammock State Park Highlands Florida Pa rk Service N27 26.4, W081 31.5 Immokalee (Pacific Tomato Growers) Collier Private Business N26 17.5, W080 25.3 Indian River Community College St. Lucie State of Florida N27 19.8, W080 24.8 Jonathan Dickinson State Park Martin Florida Park Service N27 00.3, W080 07.6 Lakeland Waste Water Management Facility Polk Polk County N27 55.2, W081 57.9 Loxahatchee National Wildlife Re fuge Palm Beach USFWS N26 31.0, W080 16.7 1 SFWMD (South Florida Water Managemen t District), HCPRCD (Hil lsborough County Parks, Recreation, and Conservation Department), SJRWMD (St. Johns River Wate r Management District, FFWCC (F lorida Fish and Wildlife Conservation Commission), USFWS (United States Fish and Wildlife Service).

PAGE 66

Table 3-2. Collection site and age of OWCF spores in weeks for each experiment. Age of OWCF Spores (weeks) Collection site Experiment 1 Experiment 2 Experiment 3 Allapattah Flats 1 & 72 N/A 25 Balm Scrub 37 N/A 12 Blue Cypress Conservation Area N/A N/A 170 DuPuis Reserve 169 & 380 N/A 26 & 432 East Orlando Water Facility N/A N/A 40 Fisheating Creek W. M.A. N/A N/A 78 Highlands Hammock State Park N/A N/A 13 & 74 Immokalee (Pacific Tomato Growers) 6, 16 & 31 N/A 16 & 40 Indian River Community College 11 & 36 29 21 & 46 Jonathan Dickinson State Park 4, 10, 12, 141 & 178 20 36, 45 & 194 Lakeland Waste Water Management Facility 7 & 35 N/A 17 Loxahatchee National Wildlife Re fuge 9, 12, 19 & 83 27 4 & 20 Table 3-3. Herbicide concentration s used to evaluate effects on Old Worl d climbing fern spore germination. Herbicide Concentrati ons (g a.i. / l) Imazapyr 1.5, 3.0, 6.0, 12.0, and 24.0 Glyphosate 3.2, 6.4, 12.8, 25.6, and 51.2 Metsulfuron 0.1, 0.2, 0.4, 0.8, and 1.6 Fluroxypyr 1.3, 2.6, 5.2, 10.4, and 20.8 Asulam 2.1, 4.2, 8.4, 16.8, and 33.6 Triclopyr 2.5, 5.0, 10.0, 20.0, and 40.0 Surfactant 0.05% (v/v) Untreated checks Distilled water 66

PAGE 67

67 Table 3-4. Effects of various concentrations of metsulfu ron on germination of Old World climbing fern spores and sporophyte survival. Metsulfuron Gametophyte Sporophyte Concentration % Germination % Survival % Survival Dry Weight (g) (mg a.i. / l) Pre-treatment 30 days post-trt 120 days post-trt 120 days post-trt 27 34.9 (4.9)1 1.4 (0.3) 0.014 (0.007) 0.061 (0.019) 54 39.7 (4.3) 1.2 (0.4) 0.009 (0.002) 0.003 (0.002) 108 39.8 (5.0) 0.8 (0.3) 0.010 (0.005) 0.002 (0.001) 216 37.6 (4.5) 0.9 (0.3) 0.002 (0.001) 0.001 (< 0.001) 432 38.9 (4.7) 0.4 (0.2) 0.000 (0.000) 0.000 (0.000) 864 34.0 (4.9) 0.3 (0.1) 0.000 (0.000) 0.000 (0.000) Surfactant 40.1 (4.5) 1.0 (0.3) 0.048 (0.019) 0.141 (0.034) Untreated checks 32.0 (5.1) 36.6 (5.4) 0.421 (0.187) 0.697 (0.094) LSD (P < 0.05) NS 4.8 0.185 0.106 1 Means of 32 replications followed by standard error.

PAGE 68

68 Table 3-5. Inhibition rates (I50 and I95) for germination of Old Wo rld climbing fern spores following exposure to six different her bicides based on non-linear regression. Old World Climbing Fern Spore Inhibition Herbicide Rate (g a.i. / l) 95% Confidence Interval Imazapyr I50 16.12 13.22 21.12 I95 69.67 52.25 103.81 Glyphosate I50 56.35 47.89 70.43 I95 243.56 165.62 435.97 Metsulfuron I50 0.014 0.011 0.017 I95 0.063 0.057 0.069 Fluroxypyr I50 22.88 18.76 29.06 I95 98.86 71.18 158.18 Asulam I50 66.01 52.15 89.11 I95 285.31 228.25 370.90 Triclopyr I50 17.96 15.80 21.01 I95 77.61 61.31 104.77

PAGE 69

11) Fisheatin g Creek W.M.A. 12) Immokalee (PTG) 10) Loxahatchee N.W.R. 9) DuPuis Reserve 8) Jonathan Dickinson State Park 7) Allapattah Flats 6) Highlands Hammock State Park 5) Indian River Community College 4) Balm Scrub 3) Blue Cypress Conservation Area 2) Lakeland Waste Water Mngt. Facility 1) East Orlando Water Facility Collection Sites: 10 11 12 8 5 3 7 9 6 4 2 1 Figure 3-1. Location of collection sites of Old World Climbing Fern spores in Florida. 69

PAGE 70

70 0 5 10 15 20 25 30 35 40 45 50 < 1011-2021-4070-100101-200>300 Age (weeks)% Germinationa a b a a a Figure 3-2. Germination rates of Old World climbing fern spores of different a ge groups (LSD = 15.0). Lines represent standard error.

PAGE 71

0 10 20 30 40 50 60 ImmokaleeLox NWRJDSPAllapattahIRCCLakelandBalm Location (South to North)% Germinationa a a a ab bc c Figure 3-3. Germination rates (%) of Old World climbing fern spores collected from different location in Florida (LSD = 14.1). Means of 28 replications with standard error bars. 71

PAGE 72

0 5 10 15 20 25 30 35 40 02754108216432864 Metsulfuron Concentrations (mg a.i. / l)Gametophyte Survival (%) y = 31.99 e -(0.1135x)R2 = 0.98 Figure 3-4. Survival of Old World climbing fern gametophytes 30 days after (60 total days in agar + nutrient medium) exposure to metsulfuron in agar medium. Means of 32 replicat ions with standard error bars. 72

PAGE 73

0 0.1 0.2 0.3 0.4 0.5 0.6 02754108216432864 Metsulfuron Concentration (mg a.i. / l)Sporophyte Survival (%) y = 0.42 e -(0.1245x)R2 = 0.98 Figure 3-5. Survival of Old World climbing fern sporophy tes 120 days (60 days in agar + nutrient medium and 60 days in potted soil) after exposure to me tsulfuron. Means of 32 replic ations with standard error bars. 73

PAGE 74

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 02754108216432864 Metsulfuron Concentrations (g a.i. / l)Sporophyte Dry Weight (g) y = 0.697 e -(0.0905x)R2 = 0.99 Figure 3-6. Old World climbing fern sporophyte dry weight (grams) at 120 days (60 days in agar + nutrient medium and 60 days in potted soil) after exposure to metsulfuron on agar medium and transplanted to potting soil. Means of 32 replications with standard error bars. 74

PAGE 75

0 1 2 3 4 5 6 7 8 9 0.00.10.20.40.81.6 Metsulfuron Concentration (g a.i. / l)Spore Germination / Survival (square-root %) y = sqrt 47.9 e -(47.85x)R2 = 0.99 Figure 3-7. Effects of metsulfuron on Ol d World climbing fern spore germination 30 days post exposure. Data are shown as percent germination with a ll exposure times combined. Means of si x replications per concentration with standard error bars. 75

PAGE 76

0 10 20 30 40 50 60 0.03.26.412.825.651.2 Glyphosate Concentration (g a.i. / l)Spore Germination / Survival (%) y = 47.51 e -(0.0123x)R2 = 0.94 Figure 3-8. Effects of gly phosate on Old World climbing fern spore germina tion 30 days post exposure. Data are shown as percent germination with a ll exposure times combined. Means of si x replications per concentration with standard error bars. 76

PAGE 77

0 10 20 30 40 50 60 0.01.53.06.012.024.0 Imazapyr Concentration (g a.i. / l)Spore Germination / Survival (%) y = 35.06 e -(0.0384x)R2 = 0.47 Figure 3-9. Effects of imazapyr on Old World climbing fern spore germination 30 days post exposure. Data are shown as percent germination with all exposure times combined. Means of six replications per concentration with standard error bars. 77

PAGE 78

0 10 20 30 40 50 60 0.02.55.010.020.040.0 Triclopyr Comcentration (g a.i. / l)Spore Germination / Survival (%) y = 44.02 e -(0.0385x)R2 = 0.95 Figure 3-10. Effects of triclopyr on Old World climbing fern spore germination 30 days post exposure. Data are shown as percent germination with all exposure times combined. Means of six replications per concentration with standard error bars. 78

PAGE 79

0 10 20 30 40 50 60 0.01.32.65.210.420.8 Fluroxypyr Concentration (g a.i. / l)Spore Germination / Survival (%) y = 43.75 e -(0.0301x)R2 = 0.81 Figure 3-11. Effects of fluroxypyr on Old World climbing fern spore germination 30 days post exposure. Data are shown as percent germination with a ll exposure times combined. Means of si x replications per concentration with standard error bars. 79

PAGE 80

0 10 20 30 40 50 60 0.02.14.28.416.833.6 Asulam Concentration (g a.i. / l)Spore Germiantion / Survival (%) y = 46.44 e -(0.0105x)R2 = 0.87 Figure 3-12. Effects of asulam on Old Wo rld climbing fern spore germination 30 da ys post exposure. Data are shown as percent germination with all exposure times combined. Means of six replications per concentration with standard error bars. 80

PAGE 81

0 5 10 15 20 25 30 35 40 45 50 Initial germination (15 days)G ametophyte survival (45 days)Germination / Survival (%) Metsulfuron Untreated check a a a b Figure 3-13. Germination rates of Old World climbing fern spores placed on an agar mixed with metsulfuron (0.4 g a.i. / l) or agar alone (untreated checks) 15 and 45 days post-treatment from spores collected at 12 locations. Different letters indicate significantly differ ent (P < 0.05) treatment means within times (days) based on paired t-tests. Means of 16 replications with standard error bars. 81

PAGE 82

0 5 10 15 20 25 30 35 40 45 50 IRCC (46 weeks) JDSP (36 weeks)Germination / Survival (%) Metsulfuron Untreated check a b b a Figure 3-14. Germination of Old Worl d climbing fern spores collected Indi an River Community College (IRCC) and Jonathan Dickinson State Park (JDSP) placed on an agar m edia mixed with metsulfuron (0.4 g a.i. / l) or agar alone (untreated checks) 45 days post-treat ment. Different letters represent significantly different (P < 0.05) treatment means based on paired t-te sts. Means of 16 replicatio ns with standard error bars. 82

PAGE 83

83 CHAPTER 4 REPEATED HERBICIDE APPLICATION FOR CONTROL OF OLD WORLD CLIMBING FERN AND THE EFFECTS ON NON-TARG ET VEGETATION ON EVERGLADES TREE ISLANDS Introduction Lygodium microphyllum (Cav.) R. Br. (Old World climbing fern; OWCF) is a nonnative vine that transforms tree islands in A.R.M. Loxahatchee National Wildlife Refuge (LNWR) from an open community of herbaceous ground cover of native ferns and a shrub layer to one with a nearly impenetrable thicket of broken limbs and a closed canopy of OWCF rachis. OWCF is known to decrease native vegetation cover, evenness and richness on tree islands (Brandt and Black 2001). Ugarte et al. (2006) suggested that OWCF, because of its twining growth into the upper portions of the canopy, acts in synergy with wind events such as hurricanes to alter the structure and composition of tree islands often causing tota l collapse of trees. Recruitment of OWCF is correlated with these disturbances (Lynch et al. 2009). In the northern Everglades wetland eco system of LNWR, OWCF invasion occurs in woody vegetation (i.e., tree islands) 89% of the time compared to sawgrass (2%) and wet prairies (9%) (Wu et al. 2006). The nor thern section of the refuge contains higher densities of OWCF due to the presence of more tree islands and prevailing southwest winds that distribute more spores into t he northern section (Wu et al. 2006). Based on models, OWCF is estimated to exceed the cover of the five most invasive non-native species ( Melaleuca quinquenervia Schinus terebinthifolius, Casuarina equisetifolia Colubrina asiatica and Ardisia elliptica ) combined in south Florida by 2014 if left unmanaged (Volin et al. 2004).

PAGE 84

Models indicate that OWCF may infest 38% (1910 ha) of all tree islands within LNWR by 2012 if left untreated (Wu et al. 2006). Woodmansee et al. (2005) similarly estimated that the cover of OWC F coul d impact 25,200 ha in LNWR, which represents 42% of the total area within t he refuge. Attempts to tr ack the spread of OWCF in southern Florida with aerial flights have proven difficult. Based on aerial surveys from fixed-wing aircraft from 1993-2003, changes in OWCF in southern Florida cover ranged from a 14% increase in 1997, a 255 % increase in 1999, a 64% decrease in 2001, and a 181% increase in 2003 (Ferriter and Pernas 2006). These large differences in changes in area coverage of OWCF from aerial su rveys can be attributed to surveys flown following periods of freezing weather that to p-killed OWCF (Ferriter and Pernas 2006). Regardless, it is clear that OWCF has become a major invader of forested wetlands in Floridas natural areas. Infestations of OWCF often occur in natural areas t hat are remote and inaccessible to vehicles and personnel, making tr eatment difficult. This is especially true in the Florida Everglades. Spore production can occur throughout the year in south Florida and each fertile leaflet is capable of producing 28,600 spores (Volin et al. 2004), indicating that annual treatment s will be required to control existing populations as well as re-sprouts and emergent sporophytes. The rapid spread of OWCF across the landscape of southern Florida can be attributed to its sma ll (<65 m) spores which move with normal wind events and are moved fu rther away from wind events such as tropical storms and hurricanes. In addition, the four life stages (spore, gametophyte, sporeling and sporophyte) of OWCF make m anagement of the fern extremely difficult without continual monitoring and re-treatment. 84

PAGE 85

The greatest area of OWCF infestations occur in Marti n and Palm Beach Counties of southern Florida, the general location of the first documented record from a naturalized specimen in Florida (Beckner 1968; Pemberton and Ferriter 1998). Within LNWR, OWCF was first noted in the interi or in 1989, and was documented from a few tree islands in 1992 and 1993 (Thomas 2006 ). During 1995, OWCF covered an estimated 7,284 ha of LNWR. The cover in creased to 19,433 ha in 2003 based on aerial transects conducted by the SFWMD (Thomas 2006). As stated above, the most recent estimate of the coverage of OWCF in LN WR was 25,200 ha (Woodmansee et al. 2005). In 2007, extensive aerial and ground herbicide treatments were initiated throughout the refuge with follow-up treatm ents scheduled in 2009-2010. Management efforts throughout souther n Florida using glyphosate and metsulfuron have not been successful for long-term control of OWCF; vigilant annual retreatments are requi red to keep the fern in check (H utchinson et al. 2006). Brandt (2004) found that glyphosate was more successful than metsulfuron in controlling OWCF, while metsulfuron resulted in less non-target damage up to 1 year post treatment in LNWR. Glyphosate is a non-selective herbicide that damages or kills many types of plants (i.e., grasses, herbs, fern s, shrubs, trees). Graminoids such as Panicum hemitomon and Cladium jamaicense have shown tolerance to metsulfuron applied to control OWCF (Langeland and Link 2006). Natural areas present a complicated situat ion for invasive plant treatment while protecting native, non-target plant species. Selective herbicides are traditionally developed to control multiple weeds in agr icultural areas with no effect on the crop (Hobbs and Humphries 1995). However, in nat ural areas, the goal of management is to 85

PAGE 86

control a single weed while protecting or limit ing damage to many nontarget species. Evidenc e suggests that aerial herbicide tr eatment of OWCF can be selective with minimal impacts to non-target deciduous tr ees during winter application in Florida (Hutchinson et al. 2006). However, evergr een non-target ground vegetation is often greatly reduced (Hutchinson et al. 2006). This is also true during re-treatment of OWCF because of its twining and climbing nature as it grows over and around non-target plants. The mission of the United States Fish and Wildlife Service (USFWS) refuge system is to manage lands and water for the conservation of fish, wildlife, and plant resources and their habitats (NWRS 2010); thus finding herbicides that can effectively control OWCF while limiting non-target damage would be in compliance with the mission of USFWS. Tree islands in this study were in the la te stages of invasion by OWCF. However, in 2004, Hurricanes Frances and Jeanne (Nati onal Hurricane Center 2009) defoliated OWCF on most tree islands (Ugarte et al. 2006). Prior to the hurricanes, tree islands selected for this study were visually obs erved from a helicopter to have OWCF coverage >90%. At the start of the pr oject in December 2005, OWCF cover had returned to approximately 60% cover indicati ng recovery can be rapid from established populations of OWCF follo wing hurricane damage. Due to the isolation of tree islands (som e in LNWR take >1 hour to reach by airboat) and the large number of islands (sever al thousand in LNWR), aerial application of herbicide is the most effective and cost ef ficient method for initial control of OWCF. The objectives of this study were to eval uate the effectiveness of consecutive annual herbicide applications of glyphosate and metsulfuron using aerial and ground 86

PAGE 87

treatments for control of OW CF and to examine the uni ntended im pacts to native ground and canopy vegetation on northern Everglades tree islands in LNWR. Materials and Methods Study Site Studies were conducted on tree islands at Arthur R. Ma rshall Loxahatchee National Wildlife Refuge (26 N, 80 W; LNWR) in western Palm Beach County, Florida from 2006-2009. LNWR encompasses 59,464 ha and represents the last remnants of the northern Everglades. The Everglades ecosystem is listed as an International Biosphere Reserve, World Heritage Site, and a Wetland of International Importance (NPS 2010). It is a peat-bas ed wetland that is comprised of sawgrass strands, wet prairies, sloughs, and tree islands (Lovele ss 1959). Tree islands number in the thousands and are a unique upland feature of the Everglades wetland ecosystem that provide habitat for wildlife (Brandt et al 2000, 2003). The largest tree islands are 125 ha but 85% of the tree island s are <0.13 ha (Brandt et al 2000). Elevation on tree islands range from 0.2-1.0 m above the surr ounding marsh (Wetzel et al. 2005). Tree islands selected for this study were located in the north portion of LNWR (26 N, 80 W) and <0.13 ha in size. With th e exception of one nat ive neotropical tree species, Rapanea punctata all understory and canopy pl ants in the northern end of LNWR were temperate species. Neotropica l plant species are more common in the southern end of LNWR and the southern Ever glades. The three most common invasive plants in the LNWR are OWCF, Melaleuca quinquenervia and Schinus terebinthifolius which are tropical in origin and o ccur in the northern part of LNWR. The locations of tree islands heavily infested with OWCF were mapped from a helicopter with a Garmin 76S global positioning system (GPS). Tree islands with >50% 87

PAGE 88

cover of OWCF based on visual observation and the presence of OWCF in the canopy were then selected from an airboat. Select ed tree islands were marked at the south end with a 3 m PVC pipe. All PVC markers were paint ed with a specific color per treatment for visual reference and numbered with a permanent tag. GPS coordinates were taken at each tree island using a Ga rmin 76S. The use of GPS coordinates and color coded PVC markers allowed for the spec ific herbicide treatment s to be applied by helicopter to the exact island designated fo r herbicide treatment with glyphosate or metsulfuron. A researcher involved with the project flew with the pilot during herbicide application to verify the designated herbicide treatment was applied to correct tree islands. Evaluation Periods Pre-treatment evaluation per iods were conducted Dece mber 2005 January 2006 (year 1). Post-treatment evaluation periods were conducted in December 2006 January 2007 (year 2), December 2007 January 2008 (year 3), and December 2008 January 2009 (year 4). For each evaluation period, cover of the vegetation in two strata (soil level to 1.5 m and >1.5 m above the soil), and status (live or dead) of dominant shrubs and trees ( Ilex cassine Persea palustris Myrica cerifera and Rapanea punctata ) were recorded. Herbicide Treatment One of the eight treatment schedules listed in Table 4-1 were randomly applied to tree islands (all with >50% OWCF cover). Each treatment was replicated on five individual islands. An additional five tree is lands were randomly selected as untreated checks (total n = 45 islands). For each treatm ent, five tree islands were clustered within 400-500 m of each other to facilit ate aerial treatment. Aerial applications were applied 88

PAGE 89

by Helic opter Applications, Inc. (Gettysbur g, PA) from a Bell 20 6 L4 helicopter with a Thru Valve Boom (TVB, Waldrum Spec ialties Inc., Southampton PA). Ground treatments were applied on a spray to wet basis using 1.5 l hand-held pressurized spray bottles. The first ground treatment was applied by LNWR staff, and the second ground application was applied by c ontractors (Applied Aquatics Management, Inc., Bartow, Florida). Aerial applications contained 0.5% v/v DLZ adjuvant and ground applications contained 0.5% v/v Sunwet adj uvant. Ground applications we re directed toward OWCF only to minimize contact with native vegetation. Ground Cover Cover of all ground vegetation 1.5 m was estimated along three randomly placed transect lines the entire width of each tree island (Canfield 1941) (Figure 4-1). The beginning and end of each line were marked with PVC pipe. During each evaluation period, a meter tape was extende d from the beginning to end of each transect line. All ground cover estimates were recorded along the transect line individually by plant species. Ground cover was calculated for all pl ants by species along the transect line by dividing the total distance of each species overlapping the line by the length of the transect line. The three lines on each tr ee island were averaged to obtain the mean ground cover of each species per tree isl and. Changes in plant composition and structure on each tree island were evaluat ed pre-treatment and post-treatment three times. Canopy Cover Percent canopy cover was estimated at t he beginning, middle, and end of every transect using a concave densiometer (Forest Densiometer, Bartlesville, OK) for all transect lines <10m in length. For transect lines >10m in length, canopy cover points 89

PAGE 90

were taken at the beginning, end, and at 25% and 75% of the leng th of the line. Densitometer readings were taken in each cardinal direction at each point. Each reading was multiplied by 1.04, and the four reading per point were averaged for canopy cover (Lemmon 1957). Percentage canopy cover was taken for live and dead vegetation >1.5 m above the ground. Canopy cover along all three lines was averaged to obtain the mean canopy cover per tree island. Plant Species Species lists for both native and exotic plants were compiled at heights 1.5 m along each line transect to determine com positional changes over the evaluation periods. Botanical nomenclature followed that of Wunderlin and H ansen (2003). Plants that could not be identified in the field were submitted to the University of Florida herbarium for identification. This information was used as a site condition assessment to determine the effects on compositional c hanges to plants from repeated herbicide application. No tree islands used in this study had been previously treated with herbicide. Plants were classifi ed as late succession (species typical of tree islands with canopy intact), ruderal (species not typically present on tree islands with canopy intact) and exotic (non-native species not indigen ous to the area). Late succession species were based on the 16 most common native plant species documented during pretreatment evaluation and those reco rded by Brandt et al. (2003). Species richness patterns were determi ned by counting the number of plant species that intersected each transect line. The three lines on each tree island were averaged to obtain the mean species richne ss per tree island. Species evenness patterns were determined using the methods of William (1964a). The three lines on each tree island were averaged to obtain t he mean evenness per tree island. Evenness 90

PAGE 91

patterns were calculat ed as E = 1 / DS, where D is Simpsons Index [ (n / N )2] and S is species richness (Williams 1964a). Native Shrubs and Trees For each treatment, Myrica cerifera (n = 60), Ilex cassine (n = 60), Persea palustris (n = 60) and Rapanea punctata (n = 20) individuals were randomly selected and tagged within 5-10 m on either side of the trans ect line among the five tree islands per treatment. Randomly sele cted shrubs and trees were marked with a numbered aluminum tag to examine the effects of aer ial herbicide application on non-target native trees over the duration of the project. The location of each tree along the line was recorded. All tagged shrubs and trees were visually evaluated by 1) height ( 3m or >3m), 2) d.b.h. ( 8 cm or >8 cm) and 3) percent live foliage ( 75% or >75%). These values were subjectively categorized, but allowed for rapid assessment of the size (small or large) and condition (partial or full folia ge) of the plants in the field. All shrubs and trees were documented as a live if any live foliage was pr esent or there were basal re-sprouts during the final evaluation at year 4. Salix caroliniana (Carolina willow) and Cephalanthus occidentalis (buttonbush) were omitted because they had entered dormancy. Both species are common along the edges of tree islands but were not included in analysis because no or only partial foliage was pr esent during the evaluation periods. Statistical Analysis OWCF Data were analyzed to determine the most effective aerial treatment (year 2), follow-up ground treatment (year 3), and consec utive annual treatment s (year 2-4) for control of OWCF. 91

PAGE 92

A linear mixed-effect analysis of vari ance was used to assess differences in OWCF cover following aerial herbicide treatm ents (year 2). Proc Mixed procedure was used with the initial cover at y ear 1 as a covariate of year 2 and random effects of tree islands (SAS 9.2). Means were separated using Tukeys multiple comparison test (P<0.05). Data were transformed using ar csine square root transformation to improve homogeneity of variance assumptions (Zar 1999) by visual inspection of the plotted residuals. A linear mixed-effect model was used to compare a follow up ground treatment (year 3) following aer ial treatment (year 2) with fa ctors of treatment and time in Proc Mixed (SAS 9.2). Year two data were used as a covariate of year 3 with random effects of tree islands. Means were separat ed using Tukeys multiple comparison test (P<0.05). Data were transformed using ar csine square root transformation to improve homogeneity of variance assumptions (Zar 1999) by visual inspection of the plotted residuals. Repeated measures analysis of covari ance (ANCOVA) was used with the pretreatment conditions at year 1 as covariat es for three consecutive annual treatments (year 2-4) for control of OWCF. Data were analyzed using Proc Mixed repeated measures ANCOVA (SAS 9.2) with time as the repeated measure, treatment as a fixed factor, and transect lines and tree islands as random factors. Variables were analyzed by treatment, time, an d the treatment by time interact ion with pre-treatment conditions as covariates. Means were separated using Tu keys multiple comparis on test (P<0.05). Data were transformed using arcsine s quare root transformation to improve 92

PAGE 93

homogeneity of variance assumptions (Zar 1999) by visual inspection of the plotted residuals. Cover, Richness, an d Evenness Separate analyses were conducted to evaluate the responses of native species ground cover, canopy cover, total species ri chness, late succession species richness, ruderal species richness, other non-native s pecies richness (not including OWCF), and species evenness to experimental treatment s. For ground cover, separate analyses were conducted for the 11 late successi on species that occurred on tree islands throughout the three year study. All ruderal and other non-native species occurred too infrequently to be analyzed individually. Data were analyzed using Proc Mixed ANCOVA (SAS 9.2) with time as the repeated measure, treatment as a fixed factor, and transect lines and tree islands as random factors. Variables were analyzed by treatment, time, and the treatment by time interaction with pre-treatment conditions as covariates. Data were transformed using arcsine square root transformation to im prove homogeneity of variance assumptions (Zar 1999) by visual inspecti on of the plotted residuals. Shrub and Tree Survival Analysis of tree survival pattern by spec ies were assessed by multiple logistic regression to model the dichotomous variable (0=dead, 1=live) as a function of the nominal variable treatment and ordinal variab les height, d.b.h., foliage and interactions. Proc Logistics (SAS 9.2) was used for all variables, and model significance was based on chi-square values with P<0.05. Pairwise contrasts were used to compare treatment responses that were significant at P<0.05. 93

PAGE 94

Results OWCF Ground Cover All aerial herbicide treatments reduced OWCF coverage to less than 9% one year after application, or an 88% to 99% decr ease compared to untreated checks (Table 42). Metsulfuron reduced coverage to a gr eater extent than glyp hosate but the impact was independent of conc entration (P<0.05). Two years after aerial treat ments and one year following si ngle ground treatments OWCF cover was less than 20%, translating to a 67% to 97% decrease compared to untreated checks (Table 4-3). As observed a fter a single aerial application, islands treated with a single metsulfuron ground applic ation had lower OWCF coverage than those treated with glyphosate but no rate ef fect was observed for either herbicide (P<0.05). A second ground treatment, following aeria l treatment and t he initial ground application, again resulted in much reduc ed OWCF cover compared to untreated checks and, again, metsulfuron treated islands had lower cover than those treated with glyphosate (P<0.05; Table 4-4) Cover the year following aerial application and a single ground application was somewhat higher on all islands, treat ed and untreated, than the year following aerial treatment only or the year following the second ground application. Numerous new sporophytes <1.5 cm in height and some re-growth of OWCF were observed on all treated islands during the fi nal evaluation at year 4. OWCF was not completely eliminated by any combination of treatments during this study. OWCF cover continuously increased with a 41.3% increase in ground cover by the end of the 4-year study on untreated islands. 94

PAGE 95

Native Plant Ground Cover Both B lechnum serrulatum (P<0.05; Table 4-5) and Smilax laurifolia (P<0.05; Table 4-6) cover were up to 80% and 90% lo wer, respectively, on all herbicide treated islands three years following aerial appl ication and two annua l ground treatments compared to the untreated island controls. I observed no effect of herbicide or concentration in cover of these species. The greatest reduction for both species occurred the year following aerial application. Osmunda cinnamomea cover was reduced by both metsulfuron treatments compared to untreated islands (P<0.05; Table 4-7), but showed no response to glyphosate. Pre-treatment cover of Woodwardia virginica was only 0.6% to 1.5% and changed little during the study. Th ree years after the aeria l application and two ground treatments, cover on tree islands treated wit h metsulfuron was not different than untreated islands, while those islands treated with glyphosate had higher cover than untreated islands (P<0.05; Table 4-8). Vitas rotundifolia cover increased overall from pretreatment levels. This species became the most common native plant based on cover at three years post-treatment. Cover on metsulfuron treated islands was not different than untreated islands and was higher on those islands treated with glyphosate (P<0.05; Table 4-9). Pre-treatment cover of Osmunda regalis was low, decreased overall following the initial aerial herbicide applic ations, and remained low throughout the study. There were differences in cover among metsulfuron treated islands and glyphosate and untreated islands (P<0.05; Table 4-10), and it was co mpletely absent from those islands treated with metsulfuron. 95

PAGE 96

Pre-treatment cover of Cladium jamaicense which was recorded almost exclusively around edges of the tree island s, was less than 1% and remained rare during the study. No changes in cover were observed during the study and no differences were observed among herbicide tr eated or untreated isl ands (P=0.18; Table 4-11). It was, however, completely absent from glyphosate treated islands. Shrub and Tree Ground Cover Myrica cerifera cover, following all three her bicide applications, was less than 1.0% on glyphosate treated islands compared to over 6.0% on untreated islands while cover on metsulfuron treated islands was not different than untreated islands (P<0.05; Table 4-12). Persea palustris cover was lower on all herbicide treated islands compared to untreated islands, e xcept for the higher rate of metsulfuron, which was not different than untreated islands (P<0.05; T able 4-13). Islands treated with the lower rates of glyphosate had the least Persea palustris cover than any treated or untreated islands. Ilex cassine cover was less on all herbici de treated islands compared to untreated islands but there were no differ ences among herbicide treatments (P<0.05; Table 4-14). Rapanea punctata cover was very low pre-treatment and no changes were observed during the study (P =0.43; Table 4-15), however no Rapanea punctata ground cover was observed on metsulfuron treated islands after pretreatment sampling. Canopy Cover One year following the three herbicide app lications, canopy cover was lower on all herbicide treated islands compared to unt reated islands and lower on glyphosate treated islands than on those treated with metsulfuron (P<0.05; Table 4-16). Canopy cover was reduced 88% to 71% on glyphosat e treated islands compared to untreated islands but only 31% to 48% on metsulfuron treated islands. 96

PAGE 97

Shrub and Tree Survival Ilex Cassine There were no differences (P=0.0963; Table 4-17) in the survival of Ilex cassine between herbicide treated island s and untreated islands indicating this species, though damaged, will survive those treatment rates of metsulfuron and glyphosate tested. There were differences (P=0.0090; Table 4-18) in height of surviving Ilex cassine with plants 3 m in height more susceptible than pl ants >3 m. No other variables or interactions were significant for Ilex cassine On control islands, the survival rate of tagged Ilex cassine was 100%. Persea Palustris Persea palustris was highly susceptible to gl yphosate, but exhibited a high tolerance to metsulfuron (P<0.0001 Table 4-17) There were interactions (P<0.0344) for herbicide*foliage for Persea palustris (Table 4-19). Plants treated with glyphosate with 75% foliage were less likely to survive compared to glyphosate treated plants with >75% foliage or any treatment with metsulfuron. No other variables or interactions were significant for Persea palustris On control islands, the survival rate of tagged Persea palustris was 100%. Myrica Cerifera Myrica cerifera was highly susceptible to glyphosate, but exhibited a high tolerance to metsulfuron (P<0.0001; Table 4-17). There were differences (P=0.0217) for percent foliage of Myrica cerifera (Table 4-20). Plants with 75% foliage were more susceptible to herbicide treatment compar ed to plants >75% foliage. Significant interactions were detected for herbicide *height (P<0.0001), herbicide*dbh (P<0.0001), and herbicide*foliage (P<0.0001) for percent survival of Myrica cerifera For all 97

PAGE 98

signific ant interactions, plants treated with gl yphosate <3 m in height, <8 cm in dbh and having 75% foliage exhibited lower survival rates compared to plants treated with metsulfuron. On control islands the survival rate of tagged Myrica cerifera was 93%. Rapanea Punctata Rapanea punctata was highly susceptible to metsul furon, but exhibited tolerance to glyphosate, especially at the lower rate (P <0.05; Table 4-17). No other variables or interactions were significant (Table 421). The highest survival percentage for Rapanea punctata was 88% for the low rate of glyphosate. On control islands, the survival rate of tagged Rapanea punctata was 100%. Plant Richness Dynamics General Trends Sixteen climax species, 34 ruderal specie s, and five exotic (non-native) plant species were documented on tree islands over the three year duration of this project (Appendix A). General trends fr om pre-treatment (year 1) to three years post-treatment (year 4) indicated no decline in climax spec ies for any treatment, large increases in ruderal species for all treatments, and sm all increases in other non-natives for glyphosate treated tree islands (Table 4-22). Total Species Richness Differences in total species richness, in cluding both native and non-native species, were observed at three years post-treatment (Table 4-23). Total species richness was higher on tree islands treated with the high ra te of glyphosate compared to all other herbicide treated and untreated islands (P<0.05). 98

PAGE 99

Late Succession Sp ecies Richness Lower late succession species richness was observed on tree islands treated with the low rate of metsulfuron compared to all other treatment s and untreated islands (P<0.05; Table 4-24). There was no diffe rence in late succession species richness among metsulfuron (high rate), glyphosate (low or high rate) and untreated tree islands. Late succession species not recorded at year 4 we re rare at the start of the project and were typically edge or gap species. Ruderal Species Richness Ruderal species richness was two to th ree times higher on those islands treated with glyphosate compared to islands treat ed with metsulfuron and untreated islands (P<0.05; Table 4-25). The largest increase in ruderal species was found on tree islands treated with the high rate of glyphosate. Other Non-Native Species Richness The mean number of other nonnative species never accounted for >1 species at any time during the study. The number of non-native species was higher on islands treated with the highest rate of glyphosate compared to other treatments and untreated islands (P<0.05; Table 4-26). At th ree years post-treatment, seedlings of Melaleuca quinquenervia and Schinus terebinthifolius were documented. Seedlings of Melaleuca quinquenervia were observed at three years posttreatment on tree islands treated with the low rate of metsulfuron, and both the low and high rates of glyphosate. Seedlings of Schinus terebinthifolius were observed on tree islands treated with both rates of glyphosate at the three years post-treatment. 99

PAGE 100

Species Evenness Species ev enness was lower on glyphosate treated islands compared to metsulfuron treated and untreated tree islands (P<0.05; Table 4-27). The low evenness values for the glyphosate treated islands are attributed to large increases in Vitis rotundifolia cover at three y ears post-treatment. Discussion Ground Cover OWCF Management Consistent with previous findings (Chapters 2,3), these results indicate that metsulfuron provided significantly better co ntrol of OWCF than glyphosate for aerial, ground follow-up and three cons ecutive herbicide treatment s. However, hundreds of new OWCF sporophytes were observed on tree islands for all glyphosate and metsulfuron treated islands during the final ev aluation. Limited information is available on the viability of OWCF spores under lab or natural conditions. Spores of OWCF maintained under dry, ambient conditions have been reported to germinate after four years in agar (Michael Lott, Florida Atlant ic University, personal communication) and spores in experiment 2 (Chapter 3) exhibited 1.5% germi nation after seven years and three months after being stored under the same conditions. The mean germination rates of OWCF spores was 84-97% on natural soils from LNWR based on lab tests (Call et al. 2007). This indicates that even with to tal eradication of mature OWCF, spores in the soil and ground surface debris as well as those blown in by wind currents will require long term managemen t. Four biannual ground treatments over two years using glyphosate, metsulfuron, triclopyr, and imazapic alone or in combination did not totally eradicate OWCF at seven sites in central or southern Florida but most OWCF observed 100

PAGE 101

was new growth from germinated spores (Jeffrey T. Hutchins on, unpublished data). In contrast, S tocker et al. (2008) continued to find OWCF re-growth from rhizomes (<1%) cover two years post treatment with tr iclopyr following bi monthly and biannual treatments but reporte d no new growth from spores. This suggests that unless infestations of OWCF are onl y a few square meters in size, re-treatment will be required for some re-growth and new growth from spores. In this study, aerial treatment of OWCF with metsulfuron reduced cover to <1.5%, which is similar to the results observed for aerial treatment of bracken fern ( Pteridium aquilinum ) with asulam (Pakeman et al. 2005). Both species are rhizomatous and can re-sprout from surviving stems and fronds. For control of br acken, multiple treatments were more effective than 1-2 treat ments (Stewart et al. 2007). Without follow-up treatments, OWCF will eventually reco ver on the tree islands to pre-treatment levels. Based on the results of this study, it is possible that OWCF could return to pre-treatment levels on tree island s in 5-6 years post treat ment if no follow-up treatments are pref ormed. In this study, OWCF incr eased in cover 1.4 to 17.5-fold on tree islands receiving two treatments compar ed to three treatments. The increase in OWCF cover at year 3 following the first ground treatment at ye ar 2 may be due to applicators over-looking young s porophytes. By the second gr ound application at year 3, these sporophytes may have become mo re conspicuous to the applicators and resulted in lower cover at year 4. This in dicates the difficulty in finding re-growth and new sporophytes during re-treatment, especially in thick OWCF rachis mats. Stocker et al. (2008) work indicates that re-treatments <6 months apart may be more beneficial to 101

PAGE 102

native spec ies as OWCF can be spot-treated with herbicide before it has the chance to climb onto and twine ar ound native vegetation. An initial objective of this study was to evaluate the effects of three annual aerial herbicide treatments with glyphosate and metsulfuron for control OWCF and the impacts to native plants due to limited access and the high cost of ground treatments. Twelve months post-treatment, the cover of OWCF was <8% for all treatments and occurred in small clusters scattered throughout the tree islands, making an additional aerial treatment unreasonable. An additional aerial herbicide treatment would have resulted in additional losses of native ground and canopy cover, further changing the structure and composition of the tree islands. Tree islands are important for providing habitat heterogeneity in the wetl and landscape of the norther n Everglades and provide food and refuge for multiple taxa of wildlife including migratory birds (Brandt and Black 2001; Brandt et al. 2003). Large scale aerial application of glyphosate over hundreds or thousands of tree islands would greatly alter t he structure of the tree islands and is not recommended. During the year two evaluations following the initial aerial tr eatment, I observed that most of the live OWCF cover occurr ed <1-2 m from the edges of tree islands indicating it was not treated. These low den sities of OWCF may not have been evident to the applicator from the ai r. Regardless, I recommend t hat the edges of the islands be treated with metsulfuron during aerial app lication. Metsulfuron has no impact on Cladium jamaicense (Langeland and Link 2006), which is common along the edges, or the native shrubs and trees that occur along the edges. 102

PAGE 103

Other evaluations of ground tr eatments of OWCF in LNWR revealed that most of the live OWCF documented on tree islands occurred along the edges at three and four and half years post-treatment (Thomas and Br andt 2003; Barrett et al. 2006). In this study, invasions of new OWCF spor ophytes were common along the edges and throughout the interior at year two and thr ee following aerial treat ments which reduced both ground and canopy creating more openings. It appears that the combined loss of ground and canopy cover following aerial ap plication creates more suitable, open conditions for OWCF spores to germinate in the interior portions of tree islands compared to ground treatments. With ground treatments, there is minimal tree mortality and loss of canopy cover. Increased spor e germination and gametophyte development of ferns can be attributed to increased light intensity and soil disturbance (Watkins et al. 2007). Gaps created from disturbance and loss of canopy can result in increased native and non-native vine cover in hardwood forest of south Florida (Horvitz et al. 1998). On Everglades tree islands, a significant number of OWCF sporophytes were found at the base of treefalls with reduced canopy cover following hurricanes (Lynch et al. 2009). Plant Diversity Dynamics The increase in species richness observed in this study following herbicide treatment is not reflective of the vegetati on typical of tree islands (Brandt and Black 2001; Brandt et al. 2003), but more reflecti ve of a habitat modified by OWCF and subsequent herbicide treatment. The lower number of ruderal species documented on tree islands treated with metsulfuron may be due to limited canopy cover damage following treatment. The total number of sp ecies recorded on tree islands prior to treatments ranged from 1521. This number of species is similar to number of species (16-19) reported on tree islands with low in festations of OWCF (Brandt and Black 2001; 103

PAGE 104

Brandt et al. 2003). However, by year 4 of the study, the number of species recorded increased to 38-42 species on glyphosate treated tree islands and 28-31 spec ies on metsulfuron treated tree islands. Tree islands treated with glyphosate exhibite d a composition shift from OWCF and late succession species (primarily native ferns) to ruderal species dominated by Vitis rotundifolia While Vitis rotundifolia is considered a late succession species typical of tree islands, it was only a small component of cover at year 1, but became the dominant native plant at year 4. Aerial treatments of bracken fern with asulam have also resulted in a shift of vegetation changes from bra cken and Ericaceae species, though in this case to one dominated by early successional species such as mosses and grasses (Pakeman et al. 2005). As suggested by Gordon (1998), this increase in species richness is due to an increase in ruderal species on tree islands. The mean number of late succession native species decreased by 1-2 species on metsulfuron treated tree islands from year 1 to year 4, but remained the same or increased slightly on tree islands treated with glyphosate. On tree islands treated with metsulfuron, Osmunda regalis O cinnamomea and Smilax laurifolia became uncommon at 4 years. For ruderal species, there was an increase of 4-6 species from year 1 to year 4 on tree islands treated with glyphosate. On metsulfuron treated tree islands, the increase in ruderal species was 2-3. Evenness values for ground cover were 0.49 at all evaluation periods from year 1 to year 4 in this study, but lower for gly phosate treated islands compared to metsulfuron treated islands. Brandt et al. (2003) reported the mean evenness value for ground cover was 0.58 on tree islands with minimal OWCF cover. Species richness is relatively 104

PAGE 105

low on tree islands in LNWR where OWCF i s not present averaging eight native species with Blechnum serrulatum being the most abundant ground cover (Brandt and Black 2001). The increase in the number of ruderal species, especia lly with the high rate of glyphosate, likely did not affect evenness val ues as the cover of most ruderal species was typically <5%. This indicates that as OWCF cover decreased following treatment, there was a concomitant increase in Vitis rotundifolia especially on glyphosate treated islands, and decrease in native ferns. Native Ground Cover Both glyphosate and metsulfuron im pacted the native ground cover on tree islands. Aerial application of glyphosate resu lted in more ruderal species at three years post-treatment. Metsulfuron impacted native fe rns more than glyphosa te. Native ferns, such as Blechnum serrulatum and Osmunda cinnamomea have been shown to be sensitive to metsulfuron at rates as low as 0.01 kg a.i./187 l/ha (Hutchinson and Langeland 2008). With the exception of ruderal species, few recruits of late succession native species were observed on treated islands At three years post-treatment, some small native ferns were observed growing on tussocks and at the base of dead trees, but new growth of native ferns were far out-numbered by new OWCF sporophytes. Community-level response to herbicide treat ment is difficult to analyze as native species cover has already been reduced by t he invasive plant (Marrs 1985; Laufenberg et al. 2005). The presence of OWCF on tree islands prior to treatment had likely already reduced native ground cover and limited propagule production. On tree islands, the combined impacts of OWCF (limited growth and propagule production of natives) prior to treatment along with herbicide treatments (further reductions in native plant 105

PAGE 106

abunda nce) and thick rachis mats will greatly inhibit and limit the re-colonization of ground cover. In this study, the only native species t hat was abundant at the end of the project was the climbing vine, Vitis rotundifolia. In hurricane damaged tropical hardwood forests of south Florida, management using cutting and her bicide of non-native vines resulted in an increased recruitment of gener alist native herbs, shrubs, and trees, but not vines (Horvitz and Koop 2001). Native fern s were still present on all tree islands at year 4, but the dominant native fern, Blechnum serrulatum was reduced by 55-77% on treated islands compared to pre-treatm ent conditions. The increase in Vitis rotundifolia cover may have prevented further invasi on of OWCF, other non-native plants such Melaleuca quinquenervia and Schinus terebinthifolius as well as limiting re-colonization of native vegetation at three years posttreatment. The largest increases in Vitis rotundifolia cover were most common over dead rach is mats of OWCF. Russell et al. (1998) found that Dicranopteris linearis a climbing fern native to the Old World tropics and Polynesia, influenced forest floor light patterns, direct ed flora development patterns and possibly prevents invasion of non-natives in Hawaiian rainfore sts. However, Vitis rotundifolia is tardily deciduous which may create suitable conditions for germination of OWCF spores in the winter. This aspect of Vitis rotundifolia may have also limited its exposure to herbicides during wi nter treatments; which in turn, allowed it to thrive and become the dominant plant at three years post-treatment due to increased sunlight. In addition to the increased cover of Vitis rotundifolia, the rachis mat of dead OWCF may have prevented the re-establishm ent of native plants. Rachis mats of OWCF can be >1 m thick (Pemberton and Ferriter 1998). On tr ee islands, many of the dead rachis mats 106

PAGE 107

climbing 6-8 m into the canopy had a d.b.h. (diameter at 1.3 m) estimated to be >400 cm that were interconnected with other OWCF rachis and native shrubs and trees forming a amorphous thicket. At the conclu s ion of the study, few native plants were observed in areas where dead OWCF rachis mats and Vitis rotundifolia were present. Prior to treatment, 16 native species charac teristic of late succession tree islands were documented. This number of late succ ession species is similar to the number reported by on tree islands with little or no OWCF cover in LNWR (Brandt et al. 2003). On tree islands heavily infested with OW CF, the abundance of native plants was significantly lower (Brandt and Black 2001). Afte r the initial aerial herbicide treatment of OWCF, there was a decrease in ground cover of native late succession species. As suggested by MacDougall and Turkington ( 2005), there are shifts from species dominants (i.e. late succession species) to annual and perennial forbs, and graminoid species that are functionally distinct (i.e ruderal) in disturbed ecosystems. Annual plants respond to the loss of canopy cover and increased sunlight, while perennial plants take longer to recover from dist urbance due to their interdependence on the previous years growth (Lindgren and Sullivan 2001). All the native late succession species on tree islands are perennials, but exhibited little recovery by the end of the study. It appears likely that native late succession species such as ferns will require a long period of time to recover. At year 4, native ferns were at approximately 30% of pre-treatment level, wh ich may indicate a recovery time of 9-10 years. Wit hout follow-up treatment of OW CF, it will eventually spread horizontally and vertically on the tree islands resulting in further suppression of native vegetation. Spot treatment by ground applicators targeti ng OWCF re-sprouts and new 107

PAGE 108

growth could limit non-target damage. Bimonthly and biannual gr ound treatments of OWCF with triclopyr resulted in no impacts to native spec ies cover, richness, evenness or diversity (Stocker et al. 2008). Long te rm research greater is needed to determine the recovery rate of native plant gr ound cover and OWCF in response to repeated herbicide treatments. Tree Survival and Canopy Cover Limited tree mortality and preservati on of the canopy was achieved with metsulfuron treatments. The dominant shrub and tree species Myrica cerifera Ilex cassine, and Persea palustris all exhibited a high tolera nce to metsulfuron. In Everglades National Park, similar result s were reported for aerial treatment with metsulfuron with limited damage to Myrica cerifera and Persea palustris but high mortality to native ferns (Taylor 2006). The dramatic reduction in shrubs and trees on islands aerially treated with glyphosate comple tely altered the structure and composition of the islands. No tropical storms or hurricanes occurred during this study, so the majority of the decrease in canopy cover on glyphosate treated islands was due to aerial herbicide application. Native shrub and tree species represent ed 22% of the ground cover on tree islands in this study prior to treatment. This seems to indicate that the growth habit of OWCF altered the growth form of shrubs and trees due to the twining nature of its rachis over the limbs and boles of shrubs and trees. Most of the shrubs and trees documented as ground cover exhibited b ent or bowed limbs and boles due to near complete entanglement with OWCF rachis. As suggested by Ugarte et al. (2006) for large trees, the thick rachis mats of OWCF can alter growth patte rns and increase the chance of snapping and up-rooti ng. On tree islands with established OWCF, the growth 108

PAGE 109

habit of OWCF with dense rachis mats climbing into the canopy likely caus es a cascade effect in which multiple large trees are toppled over. Numerous fallen trees were observed on every tree island in this study. The lack of invasion by non-native tree seedlings such as Melaleuca quinquenervia and Schinus terebinthifolius at year 4 was surprising considering both species are numerous in the ref uge and surrounding areas. However, Melaleuca quinquenervia and Schinus terebinthifolius seedlings were observed on several tree islands and may become problematic over time, especially on tree islands in which canopy was reduced to <20% cover. Hutc hinson and Vankat ( 1997) reported that canopy disturbance increased the potential for invasion of non-native plants. Disturbance and loss of canopy trees from hurricane damage on LNWR tree islands also increased the potential for OWCF inva sion (Ugarte et al. 2006). The increased availability of light may promote seed germina tion of both these invasive species. On tree islands in LNWR that were aerially treated for Melaleuca quinquenervia with imazapyr + glyphosate, high native tree mortality was observed and OWCF formed near complete ground cover within 3-4 years post-tr eatment (Gayle Martin, LNWR, personnel communication). Without persistent monitori ng and follow-up treatments, it appears that tree islands that have been treated for OWCF will again become dominated by invasive plants. The dynamics of the tree is lands following herbicide tr eatment are complex. The combined impacts of OWCF, hurricanes, and herbicide treatments greatly alter the structure and composition of tree islands. Multiple dist urbances have been suggested to act in synergy to alter habitats (H obbs and Huenneke 1992) which appears to be the 109

PAGE 110

case with OWCF on tree islands. In additi on, the size, frequency, and intensity of disturbances may limit native plant re cruitment and establishment (Hobbs and Huenneke 1992). Based on the results of this study, a conc eptual model is proposed to describe the effects of OWCF invasion and herbicide tr eatment on tree islands and emphasize the difficulties in managing this invasive plant (F igure 4-2). Spores of OWCF are blown in by wind currents, land in a suitable site and germinate. The twining rachis mat of OWCF growing over native plants inhibits the growth of native vegetation suppressing or limiting propagule production. Over time, t he impacts of multiple layers of OWCF fronds combined with thick inter-twining rach is mats greatly reduce native plant cover and abundance. As spores are released, th ey fall nearby or could be carried in wind currents to invade other sites. When herbicide treatment occurs, OWCF cove r is greatly reduced as well as much of the remaining native ground cover. Furthe r reductions in canopy cover result in increased solar radiation and increased soil mois ture. These conditions create suitable sites for ruderal species, OWCF, and other non-native species to invade. Early successional natives become the domi nant plant cover, but numerous OWCF sporelings occur throughout the tree island on tussocks, elevated soil mounds, and at the base of dead trees. Managem ent of OWCF is dependent on re-treatment which is limited by access, funding, and management priori ties. Providing that re-treatment of OWCF does occur on a regular basis, it will like ly take >10 years for a tree island to return to its original state provided the c anopy cover has remained relatively intact. If there is substantial tree mortality, then recovery may take many decades. 110

PAGE 111

In summary, the use of herbicides can e ffectively control OWCF on tree islands, but results in substantial non-target damage to ground cover. Aerial applications of metsulfuron limited shrub and tree mortalit y and reductions in canopy cover compared to glyphosa te treatments. Monitoring and follow-up treatments will be required for control of OWCF on an annual basis. Wi thout herbicide management, OWCF will continue to form monocultures on tree isl ands and other habitats resulting in lower native species richness, evenness, and canopy cover on tree islands. Follow-up treatments will reduce OWCF growth and spore production, limiting the ability of OWCF to infest new sites from wind blown spores. Funding limitations for long-term monitoring and treatment along with limited accessibility make management of OWCF very difficult in LNWR. The introduction of effective bi ocontrol agents is the best hope for long term management of OWCF and maintenance of t he structure and composition of tree islands. Management Implications The isolated nature of OWCF infestat ions at LNWR make ground treatments expensive, and with thousands of tree islands infested with OWCF, adequate funding may not be available for follow-up treatm ents. The benefits of using metsulfuron for aerial treatments of OWCF on Everglades tree islands ar e limited tree mortality and preservation of canopy cover. Conversely, aerial treatments with glyphosate result in high tree mortality and reductions in canopy cover. Both metsulfuron and glyphosate resulted in substantial declines in ground cover. Recovery of trees and canopy cover will require a much longer time, possibly dec ades, due to the constraints of secondary growth, while ferns should recover faster due to a lack of secondary growth. 111

PAGE 112

It is likely that OWCF cover can r eturn to pre-treatment levels <6 years following a single herbicide treatment. Biocontrol agents may be the best long term management strategy, but may take decades for effectiv e populations to become established. Until biocontrol agents are discovered that significantly im pact OWCF populations on a landscape scale, the best management strategy for OWCF on tree islands is an initial aerial application with metsulfuron over large infestations and annual monitoring and follow-up ground treatments wit h herbicides of differ ent modes of action. It is not recommended that the same herbicide be used repeatedly because OWCF exhibits characteristi cs of weeds that can rapidly develop herbicide resistance (Hutchinson et al. 2007). This is especially true for the acetolactate synthase (ALS) inhibitor herbicides, such as metsulfuron, where resistance in some weeds can occur after several applications to the target weed (Lovell et al. 1996; Heap 1997). The use of metsulfuron, glyphosate, and triclopyr should be used in rotation or in combinations for control of OWCF to limit t he potential for development of re sistance. In addition, other herbicides such as imazamox, imazapyr, imazapic, fluroxypyr, and aminopyralid should be incorporated into management plans for follow-up treatments. These herbicides have shown efficacy for control of OWCF under greenhouse conditions (Jeffrey T. Hutchinson, unpublished data). Gl yphosate, triclopyr, imazapy r, and imazamox are also approved for use over standing water. Metsulfuron, fluroxypyr, and aminopyralid can only be used for control of OWCF when no standing water is present. These seven herbicides represent three modes of action t hat can be utilized to treat OWCF under dry conditions or when standing water is present. 112

PAGE 113

113 Table 4-1. Herbicide treatments and treatment frequency for aerial and ground herbicide application on tree islands in the A.R.M. Loxahatchee National Wildlife Refuge. Treatment rate (kg a.i./187 l/ha) (g a.i./l) Aerial Ground 1 Herbicide February 2006 (Year 1) February-March 2007 (Year 2) February-March 2008 (Year 3) Metsulfuron 0.08 0.03 No treatment Metsulfuron 0.08 0.03 0.03 Metsulfuron 0.16 0.06 No treatment Metsulfuron 0.16 0.06 0.06 Glyphosate 2.80 5.20 No treatment Glyphosate 2.80 5.20 5.20 Glyphosate 5.60 10.40 No treatment Glyphosate 5.60 10.40 10.40 Untreated check No treatment No treatment No treatment 1 Spray to wet application Table 4-2. Effects of aerial herbicide applications on Old World climbing fern ground cover one year after application (mixed linear model ANOVA, F = 17.2, df = 4, P < 0.0001) with pre-tr eatment cover as the covariate. Pre-treatment 1 year post-treatment 1 Treatment (kg a.i./ha) % Ground Cover (SE) % Ground Cover (SE) Metsulfuron 0.08 67.3 (4.4) 2 0.4 (0.1) a 0.16 62.0 (3.2) 1.7 (0.7) a Glyphosate 2.80 66.5 (3.6) 7.0 (1.6) b 5.60 69.7 (2.4) 8.1 (1.3) b Untreated check 46.0 (5.1) 56.9 (4.8) c 1 Different letters in last column indicate significant differences among treatments at P < 0.05 for aerial herbicide treatment based on Tukeys adjusted least square means. 2 Means from 10 tree islands followed by standard error.

PAGE 114

Table 4-3. Effects of aerial and a si ngle ground treatment of herbicide on Old World climbing fern ground cover at one year following ground treatment and two years following aerial application (mixed linear model ANOVA, F = 155.0, df = 4, P < 0.0001) with one year post treatment cover as the covariate. 1 year post-treatment 2 years post-treatment 1 Treatment OWCF % Ground Cover (SE) OWCF % Ground Cover (SE) Metsulfuron 0.08 kg a.i./ha aerial & 0.03 g a.i./l ground 0.4 (0.1) 2 4.2 (0.6) a 0.16 kg a.i./ha aerial & 0.06 g a.i./l ground 1.7 (0.7) 3.8 (0.9) a Glyphosate 2.80 kg a.i./ha aerial & 5.20 g a.i./l ground 7.0 (1.6) 19.9 (2.6) b 5.60 kg a.i./ha aerial & 10.40 g a.i./l ground 8.1 (1.3) 14.3 (2.6) b Untreated check 56.9 (4.8) 59.9 (3.0) c 1 Different letters in last column indicate significant differences among treatments at P < 0.05 for aerial herbicide treatment based on Tukeys adjusted least square means. 2 Means of 10 tree islands followed by standard error. Table 4-4. Effects of aerial and two annual ground treatments of herbicide on Old World clim bing fern ground cover at year 4 (repeated-measures ANCOVA, F = 100.2, df = 4, P < 0.0 001) with year 1 cover as the covariate and time (F = 303.3, df = 3, P < 0.0001). OWCF% ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 59.8 (3.8) 2 0.6 (0.2) 3.6 (1.2) 0.4 (0.1) a Metsulfuron (0.16 & 0.06) 54.3 (3.9) 0.4 (0.2) 3.5 (0.9) 1.2 (0.9) a Glyphosate (2.80 & 5.20) 66.3 (3.5) 9.0 (3.0) 16.6 (2.8) 5.7 (1.2) b Glyphosate (5.60 & 10.40) 72.5 (2.1) 9.0 (1.7) 17.5 (4.6) 9.2 (2.1) b Untreated check 46.0 (5.1) 57.0 (4.8) 60.0 (3.0) 65.0 (4.2) c Time 1 a c b c 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. 114

PAGE 115

Table 4-5. Effects of aerial and two annual gr ound treatments of herbicide on Blechnum serrulatum (swamp fern) ground cover at year 4 (repeated-measures A NCOVA, F = 9.4, df = 4, P = 0.0002) with year 1 cover as the covariate and time (F = 63.5, df = 3, P < 0.0001). Blechnum serrulatum % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 17.9 (3.7) 2 2.7 (1.1) 9. 1 (3.5) 8.0 (6.1) a Metsulfuron (0.16 & 0.06) 19.5 (4.9) 0.9 (0.3) 7.0 (1.7) 7.0 (1.6) a Glyphosate (2.80 & 5.20) 33.4 (7.7) 7.6 (1.3) 9.3 (2.8) 8.2 (2.0) a Glyphosate (5.60 & 10.40) 24.5 (4.2) 1.8 (0.6) 3.0 (1.5) 5.4 (3.5) a Untreated check 33.8 (5.5) 33.3 (6.2) 33.3 (7.2) 32.2 (6.5) b Time 1 a b c b 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adjusted least square means. 2 Means of 5 tree islands followed by standard error. Table 4-6. Effects of aerial and two annual ground treatments of herbicide on Smilax laurifolia (green brier) ground cover at year 4 (repeated-measures ANCOVA, F = 12.5, df = 4, P < 0.0001) with year 1 cover as the covariate and time (F = 149.1, df = 3, P < 0.0001). Smilax laurifolia % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 8.4 (3.5) 2 0.7 (0.2) 0.1 (0.1) 0.4 (0.2) a Metsulfuron (0.16 & 0.06) 11.8 (3.2) 0.4 (0.2) 0.3 (0.2) 0.1 (0.1) a Glyphosate (2.80 & 5.20) 17.2 (2.9) 0.3 (0.2) 0.2 (0.1) 1.8 (0.9) a Glyphosate (5.60 & 10.40) 18.3 (2.6) 0.4 (0.1) 0.1 (0.1) 0.3 (0.1) a Untreated check 12.0 (1.2) 6.8 (1.3) 5.3 (1.1) 8.2 (1.0) b Time 1 a b c b 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adjusted least square means. 2 Means of 5 tree islands followed by standard error. 115

PAGE 116

Table 4-7. Effects of aerial and two annual gr ound treatments of herbicide on Osmunda cinnamomea (cinnamon fern) ground cover at year 4 (repeated-measures ANCOVA F = 6. 2, df = 4, P = 0.0020) with year 1 cover as the covariate and time (F = 12. 0, df = 3, P < 0.0001). Osmunda cinnamomea % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 1.0 (0.6) 2 0.2 (0.1) 1.2 (0.5) 0.4 (0.3) a Metsulfuron (0.16 & 0.06) 0.3 (0.1) 0.2 (0.1) 1.2 (0.3) 1.0 (0.7) a Glyphosate (2.80 & 5.20) 1.0 (0.4) 2.2 (0.7) 3.3 (0.7) 3.9 (1.0) b Glyphosate (5.60 & 10.40) 1.6 (0.5) 5.0 (1.2) 4.3 (1.2) 5.5 (1.2) b Untreated check 1.3 (0.7) 7.5 (2.1) 5.5 (2.4) 5.6 (1.8) b Time 1 a b b b 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adjusted least square means. 2 Means of 5 tree islands followed by standard error. Table 4-8. Effects of aerial and two annual ground treatments of herbicide on Woodwardia virginica (Virginia chain fern) ground cover at year 4 (repeated-measures ANCOVA, F = 4. 0, df = 4, P = 0.0159) with year 1 cover as the covariate and time (F = 0.9, df = 3, P = 0.4168). Woodwardia virginica % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 0.6 (0.2) 2 0.1 (0.1) 0.5 (0.3) 0.2 (0.1) a Metsulfuron (0.16 & 0.06) 0.8 (0.2) 0.3 (0.1) 1.1 (0.3) 0.8 (0.2) a, b Glyphosate (2.80 & 5.20) 0.2 (0.1) 2.4 (0.8) 1.1 (0.5) 2.3 (1.2) b Glyphosate (5.60 & 10.40) 1.2 (0.4) 2.4 (0.7) 0.7 (0.3) 2.4 (0.9) b Untreated check 1.5 (0.6) 0.6 (0.2) 0.3 (0.2) 0.5 (0.2) a Time 1 NS NS NS NS 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. 116

PAGE 117

Table 4-9. Effects of aerial and two annual gr ound treatments of herbicide on Vitis rotundifolia (muscadine) ground cover at year 4 (repeated-measures ANCOVA, F = 6.0, df = 4, P = 0.0024) with year 1 as the covariate and time (F = 41.2, df = 3, P < 0.0001). Vitis rotundifolia % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 3.4 (0.8) 2 25.2 (5.9) 8.7 (1.9) 13.0 (4.0) a Metsulfuron (0.16 & 0.06) 8.4 (2 .6) 29.0 (6.6) 12.7 (3.8) 29.4 (8.7) a, b Glyphosate (2.80 & 5.20) 10.2 (2.9) 16.9 (4.7) 42.7 (8.3) 59.4 (4.1) b Glyphosate (5.60 & 10.40) 10.1 (2.0 ) 19.3 (5.5) 37.5 (8.2) 45.8 (7.2) b Untreated check 7.2 (2.9) 2. 1 (0.8) 7.2 (2 .4) 10.6 (3.5) a Time 1 a b b c 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. Table 4-10. Effects of aerial and two annual ground treatments of herbicide on Osmunda regalis (royal fern) ground cover at year 4 (repeated-measures ANCOVA, F = 3.8, df = 4, P = 0.0195) with year 1 as the covariate and time (F = 9.5, df = 3, P < 0.0001). Osmunda regalis % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 0.6 (0.4) 2 0.0 (0) 0.1 (0.1) 0.0 (0) a Metsulfuron (0.16 & 0.06) 1.1 (0.7) 0.0 (0) 0.0 (0) 0.0 (0) a Glyphosate (2.80 & 5.20) 1.2 (0.6) 0.7 (0.2) 0.4 (0.2) 0.2 (0.1) b Glyphosate (5.60 & 10.40) 2.4 (1.1 ) 0.4 (0.2) 0.2 (0.1) 0.1 (0.1) b Untreated check 0.4 (0.1) 0.7 (0.3) 0.5 (0.2) 0.5 (0.2) b Time 1 a b b b 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. 117

PAGE 118

Table 4-11. Effects of aerial and two annual ground treatments of herbicide on Cladium jamaicense (sawgrass) ground cover at year 4 (repeated-measures A NCOVA, F = 1.8, df = 4, P = 0.1749) with year 1 cover as the covariate and time (F = 2.2, df = 3, P = 0.0943). Cladium jamaicense % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 0.1 (0.1) 2 0.1 (0.1) 0.4 (0.2) 0.4 (0.3) Metsulfuron (0.16 & 0.06) 0.9 (0.4) 1.7 (0.9) 2.4 (1.3) 3.4 (2.1) Glyphosate (2.80 & 5.20) 0.3 (0.1) 0.0 (0) 0.1 (0.1) 0.0 (0) Glyphosate (5.60 & 10.40) 0.9 (0.4) 0.0 (0) 0.0 (0) 0.0 (0) Untreated check 0.9 (0.5) 0.8 (0.5) 1.2 (0.8) 1.1 (0.7) Time 1 NS NS NS NS 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adjusted least square means. 2 Means of 5 tree islands followed by standard error. Table 4-12. Effects of aerial and two annual ground treatments of herbicide on Myrica cerifera (wax myrtle) ground cover at year 4 (repeated-measures ANCOVA, F = 10.5, df = 4, P < 0.0001) with year 1 cover as the covariate and time (F = 10.5, df = 3, P < 0.0001). Myrica cerifera % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 3.2 (0.6) 2 3.9 (1.5) 4.8 (1.6) 2.8 (0.3) b Metsulfuron (0.16 & 0.06) 3.4 (1.0) 1.9 (0.9) 3.8 (1.0) 6.2 (0.8) b Glyphosate (2.80 & 5.20) 2.6 (0.6) 0.1 (0.1) 0.3 (0.1) 0.9 (0.3) a Glyphosate (5.60 & 10.40) 2.2 (0.6) 0.2 (0.1) 0.4 (0.2) 0.7 (0.2) a Untreated check 3.4 (1.4) 3.9 (1.3) 2.9 (0.6) 6.4 (1.8) b Time 1 a b b a 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adjusted least square means. 2 Means of 5 tree islands followed by standard error. 118

PAGE 119

Table 4-13. Effects of aerial and two annual ground treatments of herbicide on Persea palustris (swamp bay) ground cover at year 4 (repeated-measures A NCOVA, F = 8.1, df = 4, P = 0.0005) with year 1 cover as the covariate and time (F = 15.4, df = 3, P < 0.0001). Persea palustris % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 6.6 (1.7) 2 9.0 (2.7) 9.2 (2.4) 3.4 (1.7) b Metsulfuron (0.16 & 0.06) 11.2 (2.3) 9.4 (1.3) 11.4 (1.9) 10.1 (2.2) c Glyphosate (2.80 & 5.20) 8.7 (0.7) 1.2 (0.7) 0.7 (0.5) 0.5 (0.2) a Glyphosate (5.60 & 10.40) 11.4 (4.6) 0. 9 (0.5) 1.6 (0.4) 3.3 (1.1) b Untreated check 10.8 (3.4) 8.6 (1.4 ) 8.8 (2.4) 6.8 (1.0) c Time 1 a b b b 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. Table 4-14. Effects of aerial and two annual ground treatments of herbicide on Ilex cassine (Dahoon holly) ground cover at year 4 (repeated-measures ANCOVA, F = 6.4, df = 4, P = 0.0018) with year 1 cover as the covariate and time (F = 28.2, df = 3, P < 0.0001). Ilex cassine % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 6.0 (1.9) 2 2.1 (0.9) 1.0 (0.5) 1.0 (0.6) a Metsulfuron (0.16 & 0.06) 10.2 (2.0) 2.5 (0.8) 2.6 (0.5) 3.5 (1.9) a,b Glyphosate (2.80 & 5.20) 11.3 (4.0) 1.2 (0.5) 0.8 (0.2) 1.8 (0.5) a,b Glyphosate (5.60 & 10.40) 3.3 (1 .9) 0.2 (0.1) 0.3 (0.2) 0.2 (0.1) a Untreated check 6.5 (1.8) 5.6 (1.4) 3.3 (0.8) 6.3 (1.4) c Time 1 a b b b 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. 119

PAGE 120

Table 4-15. Effects of aerial and two annual ground treatments of herbicide on Rapanea punctata (myrsine) ground cover at year 4 (repeated-measures ANCOVA, F = 1.0, df = 4, P = 0.4311) with year 1 cover as the covariate and time (F = 4.2, df = 3, P = 0.0088). Rapanea punctata % ground cover (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 0.8 (0.6) 2 0.0 (0) 0.0 (0) 0.0 (0) Metsulfuron (0.16 & 0.06) 0.2 (0.1) 0.0 (0) 0.0 (0) 0.0 (0) Glyphosate (2.80 & 5.20) 1.0 (0.5) 0.5 (0.2) 0.4 (0.3) 1.4 (1.1) Glyphosate (5.60 & 10.40) 1.6 (1.1 ) 0.1 (0.1) 0.9 (0.5) 1.1 (0.7) Untreated check 1.2 (0.9) 0.8 (0.5) 0.9 (0.5) 0.8 (0.4) Time 1 a b b b 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. Table 4-16. Effects of aerial and two annual ground treatments of he rbicide on tree canopy cover at year 4 (repeatedmeasures ANCOVA, F = 7.5, df = 4, P < 0.0001) with year canopy cover as t he covariate and time (F = 13.8; df = 3, P < 0.0001). % Canopy (SE) Treatment Evaluation Period (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 1 Metsulfuron (0.08 & 0.03) 51 (9) 2 51 (6) 44 (11) 31 (6) b Metsulfuron (0.16 & 0.06) 44 (8) 52 (8) 43 (8) 40 (6) b Glyphosate (2.80 & 5.20) 55 (5) 44 (4) 21 (5) 7 (3) a Glyphosate (5.60 & 10.40) 42 (4) 33 (6) 13 (3) 16 (4) a Untreated check 69 (5) 71 (4) 65 (8) 58 (7) c Time 1 a a b c 1 Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. 120

PAGE 121

Table 4-17. Survival rates (%) of the four most common shrub and tree species ( Ilex cassine Persea palustris Myrica cerifera and Rapanea punctata ) at three years post aerial herbicide treatment. Herb icide rates represent aerial treatment. % Tree survival (SE) Three years post-trt Herbicide (kg a.i.) Ilex cassine Persea palustris Myrica cerifera Rapanea punctata Metsulfuron 0.08 80 (7) 2 77 (6) b 65 (7) b 12 (8) a Metsulfuron 0.16 89 (4) 93 (3) b, c 81 (4) a,b 18 (8) a Glyphosate 2.80 75 (6) 6 (3) a 18 (6) c 88 (13) b,c Glyphosate 5.60 70 (7) 20 (6) a 17 (5) c 58 (10) b Untreated check 100 (0) 100 (0) c 93 (5) a 100 (0) c 1 Different letters in columns for tree s pecies indicates significant difference at P < 0.05 for the thre e year evaluation period based on logistic regression analysis. 2 Means of 50 Ilex cassine 60 Persea palustris 60 Myrica cerifera and 20 Rapanea punctata replications per treatment followed by standard error. Table 4-18. Summary statis tics for logistic regression analysis of tree variables as model predictors of Ilex cassine (Dahoon holly) survival following herbicide treatment. Parameters DF Wald Chi-Square P-value 1 Overall model 7 33.18 < 0.0001 Herbicide 4 7.87 0.0963 Height 1 6.83 0.0090 D.B.H. 1 0.03 0.8563 Foliage 1 0.09 0.7655 Herbicide*Height 4 3.56 0.4685 Herbicide*DBH 4 3.96 0.4112 Herbicide*Foliage 4 6.93 0.1395 1 Significance level of P < 0.05 for test s of variables and their interaction. 121

PAGE 122

Table 4-19. Summary statis tics for logistic regression analysis of tree variables as model predictors of Persea palustris (swamp bay) survival following herbicide treatment. Parameters DF Wald Chi-Square P-value 1 Overall model 7 81.97 < 0.0001 Herbicide 4 75.31 < 0.0001 Height 1 0.01 0.9410 D.B.H. 1 0.72 0.3955 Foliage 1 0.28 0.5971 Herbicide*Height 4 5.67 0.2255 Herbicide*DBH 4 5.26 0.2613 Herbicide*Foliage 4 10.39 0.0344 1 Significance level of P < 0.05 for test s of variables and their interaction. Table 4-20. Summary statis tics for logistic regression analysis of tree variables as model predictors of Myrica cerifera (wax myrtle) survival follo wing herbicide treatment. Parameters DF Wald Chi-Square P-value 1 Overall model 7 75.91 < 0.0001 Herbicide 4 61.25 < 0.0001 Height 1 0.73 0.3919 D.B.H. 1 0.09 0.7596 Foliage 1 5.27 0.0217 Herbicide*Height 4 25.09 < 0.0001 Herbicide*DBH 4 24.82 < 0.0001 Herbicide*Foliage 4 30.70 < 0.0001 1 Significance level of P < 0.05 for test s of variables and their interaction. 122

PAGE 123

123 Table 4-21. Summary statis tics for logistic regression analysis of tree variables as model predictors of Rapanea punctata (myrsine) survival following herbicide treatment. Parameters DF Wald Chi-Square P-value 1 Overall model 7 16.98 0.0175 Herbicide 4 12.85 0.0120 Height 1 2.20 0.1377 D.B.H. 1 0.38 0.5402 Foliage 1 0.12 0.7322 Herbicide*Height 4 7.61 0.1069 Herbicide*DBH 4 8.34 0.0798 Herbicide*Foliage 4 4.46 0.3469 1 Significance level of P < 0.05 for test s of variables and their interaction.

PAGE 124

124 Table 4-22. Total number of late successi on, ruderal, non-native (including OWCF) and all plant species recorded pre-treatment (y ear 0) to year three post treatment for metsulfuron and glyphosate treated tree islands and untreated checks. Herbicide rates in kg represent aerial treatment (year 0) and g represent ground treatments (year 1 and 2). Total number of Species Late Herbicide Year Succession Ruderal Exotic Total Metsulfuron (0.08 kg & 0.03 g a.i.) 1 15 2 3 20 2 13 4 3 20 3 14 13 3 30 4 12 13 3 28 Metsulfuron (0.16 kg & 0.06 g a.i.) 1 14 4 2 20 2 13 6 1 20 3 15 13 2 30 4 13 16 2 31 Glyphosate (2.80 kg & 5.20 g a.i.) 1 14 2 1 17 2 12 3 1 16 3 14 14 3 31 4 14 20 4 38 Glyphosate (5.60 kg & 10.40 g a.i.) 1 14 6 1 21 2 14 7 2 23 3 13 15 3 31 4 14 24 4 42 Untreated check 1 14 0 1 15 2 12 2 1 15 3 14 4 1 19 4 13 2 1 16

PAGE 125

125 Table 4-23. Effects of aerial and two annual ground treatments of he rbicide on total species richness (including nonnative species) at year 4 (repeated-m easures ANCOVA, F = 8.2, df = 4, P < 0.0001) with year 1 as the covariate and time (F = 24. 6, df = 3, P < 0.0001). Treatment Total species richness (SE) (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 Metsulfuron (0.08 & 0.03) 8.6 (0.4) 2 6.5 (0.4) 8.1 (0.6) 8.1 (0.6) a Metsulfuron (0.16 & 0.06) 8.9 (0.3) 5. 9 (0.4) 9.5 (0.7) 9.2 (0.4) a,b Glyphosate (2.80 & 5.20) 8.3 (0.4) 6. 9 (0.5) 9.2 (0.9) 11.9 (0.7) b Glyphosate (5.60 & 10.40) 8.9 (0.6) 8.1 (0.6) 10.3 (0.9) 14.0 (1.3) c Untreated check 8.8 (0.4) 8.4 (0.3 ) 9.9 (0.3) 9.4 (0.6) a,b Time 1 a b a c 1 0.05 for the three year eval2 Table 4-24. Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < uation period bas ed on Tukeys adj usted least square means. Means of 5 tree islands followed by standard error. Effects of aerial and two annua l ground treatments of herbicide on late succession species richness at year 4 (repeated-measures ANCOVA, F = 6.9, df = 4, P < 0.0001) with year 1 as the covariate and time (F = 10.6, df = 3, P < 0.0001). Treatment Late succession species richness (SE) (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 Metsulfuron (0.08 & 0.03) 7.4 (0.3) 2 5.7 (0.3) 6.0 (0.4) 5.6 (0.3) a Metsulfuron (0.16 & 0.06) 7.6 (0.2) 5.1 (0.4) 7.1 (0.4) 6.7 (0.3) b Glyphosate (2.80 & 5.20) 7.3 (0.4) 5.7 (0.4) 6.0 (0.7) 7.1 (0.4) b,c Glyphosate (5.60 & 10.40) 7.7 (0.6) 5.7 (0.6) 5.7 (0.7) 7.0 (0.7) b,c Untreated check 7.8 (0.4) 7.1 (0.5) 8.4 (0.3) 8.1 (0.6) c Time 1 a b c d 1 0.05 for the three year eval2Different letters in last column and ro w indicate significant differences among treatments and time, respectively, at P < uation period based on Tukeys adj usted least square means. Means of 5 tree islands followed by standard error.

PAGE 126

Table 4-25. Effects of aerial and two annual ground treatments of he rbicide on ruderal species richness at year 4 (repeated-measures ANCOVA, F = 19.9, df = 4, P < 0.0001) wit h year 1 as the covariate and time (F = 70.5, df = 3, P < 0.0001). Treatment Ruderal species richness (SE) (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 Metsulfuron (0.08 & 0.03) 0.1 (0.1) 2 0.3 (0.1) 1.1 (0.2) 1.8 (0.2) a Metsulfuron (0.16 & 0.06) 0.1 (0.1) 0.4 (0.1) 1.3 (0.3) 1.8 (0.3) a Glyphosate (2.80 & 5.20) 0.0 (0.0) 0.3 (0.1) 2.1 (0.4) 3.7 (0.6) a,b Glyphosate (5.60 & 10.40) 0.3 (0.1) 1.3 (0.4) 3.2 (0.9) 5.6 (0.9) b Untreated check 0.0 (0.0) 0.2 (0.1) 0.6 (0.3) 0.3 (0.1) c Time 1 a a b c 1 Different letters in last column and row indicate significant differences among tr eatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. Table 4-26. Effects of aerial and two annua l ground treatments of herbicide on nonnative species (other than Old World climbing fern richness at year 4 (repeated-measures ANCO VA, F = 5.4, df = 4, P = 0.0007) with year 1 as the covariate and time (F = 3.9, df = 3, P = 0.0107). Treatment Non-native s pecies richness (SE) (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 Metsulfuron (0.08 & 0.03) 0.1 (0.1) 2 0.1 (0.1) 0.0 (0.0) 0.1 (0.1) a Metsulfuron (0.16 & 0.06) 0.3 (0.1) 0.0 (0.0) 0.0 (0.0) 0.1 (0.1) a Glyphosate (2.80 & 5.20) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.2 (0.1) a Glyphosate (5.60 & 10.40) 0.0 (0.0 ) 0.1 (0.1) 0.4 (0.1) 0.4 (0.1) b Untreated check 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) a Time 1 a a a b 1 Different letters in last column and rows indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. 126

PAGE 127

Table 4-27. Effects of aerial and two annual ground treatments of herbicide on species evenness at year 4 (repeatedmeasures ANCOVA, F = 5.0, df = 4, P = 0.0012) with year 1 as the covariat e and time (F = 7.7, df = 3, P < 0.0001). Treatment Species evenness (SE) (aerial kg a.i./ha & ground g a.i./l) Year 1 Year 2 Year 3 Year 4 Metsulfuron (0.08 & 0.03) 0.34 (0.03) 2 0.44 (0.05) 0.49 (0.04) 0.44 (0.04) b Metsulfuron (0.16 & 0.06) 0.40 (0.03) 0. 44 (0.01) 0.49 (0.04) 0.39 (0.03) b Glyphosate (2.80 & 5.20) 0.42 (0.03) 0. 49 (0.05) 0.32 (0.05) 0.21 (0.02) a Glyphosate (5.60 & 10.40) 0.38 (0.02) 0. 45 (0.05) 0.30 (0.03) 0.27 (0.03) a Untreated check 0.41 (0.03) 0.36 (0. 01) 0.29 (0.02) 0.37 (0.04) b Time 1 a b a c 1 Different letters in last column and rows indicate significant differences among treatments and time, respectively, at P < 0.05 for the three year eval uation period based on Tukeys adj usted least square means. 2 Means of 5 tree islands followed by standard error. 127

PAGE 128

^ ^ ^ ^ ^ ^ Figure 4-1. Sampling design used to evaluate responses of ground cover ( 1.5 m), canopy cover (> 1.5 m), and tagged trees on northern Everglades tree is lands treated with herbicides to control Old World climbing fern. * ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^Legend: ---= Ground cover = Canopy cover ^ = Tagged trees 128

PAGE 129

Figure 4-2. Conceptual model on the effects of OWCF on tree islands and sequence of events following herbic ide treatment. Formation of thick OWCF rachis mat Decrease in native p lant cover / abundance Loss of cano py cover from tree fall Decrease in solar radiation Decrease in soil moisture Decrease in native p lant p ro p a g ules Limited recruitment of native p lants Increased chance of fire Herbicide applied for control of OWCF Altered structure and composition of tree island Loss of g round and cano py cover Reduction of OWCF Reduction in native p lant abundance Increase in OWCF s p ore g ermination Increase in solar radiation Increase in soil moisture Increase in ruderal s p ecies Other non-native invade Limited recruitment of native p lants Recovery > 3 years Access Funding Re-treatment Mngt. priority Tree Island OWCF spores blown in Increase in OWCF cove r T i OWCF inva de other sites m e Additional Management p roblems 129

PAGE 130

130 CHAPTER 5 SUMMARY The invasive potential of OWCF may redef ine how invasive potential is defined and may have resulted in a perpetual management problem in natural areas of Florida. The spread of most invasive plants of Fl oridas natural areas can be linked to a long history of various activities such as ornamental use, forage production or erosion control. There is no clear documentation on how OWCF was introduced into Florida (Nauman and Austin 1978; Pemberton and Fe rriter 1998). The earliest documented occurrences of live OWCF in Florida are from two herbarium reports from 1958 indicating the species was present in nurseri es. It maybe that OWCFs presence in Florida is due to just a few unintentional introductions. After 1958, it was not noted again for a decade until it was discovered invading natural areas. Since that time, the speed of invasion into natural wetlands has made OWCF one of the top two or three most invasive plants in natural areas of Florida. The management of OWCF is difficult because of the numerous spores it produces, it s ability to colonize and thrive in remote sites, and its extensive range in central and southern Florida. This study examined various aspects of OWCF management during all phases of its life cycle using herbicides. Translocation studies using 14C radiolabeled herbicides indicated that limited movement of glyphosate, mets ulfuron, and triclopyr occurred in the rhizomes of OWCF. In the 14C study, herbicides moved basipitally in the rachis, but movement of the herbicides stopped at the rhizome and did move vertically through the rhizomes. The key to successful control of OWCF is t he elimination of the rhizomes because new fronds emerge at irregular intervals on the rhizome. In established populations of

PAGE 131

OWCF, the number of new fronds that ar ise from rhizomes may exceed severa l hundred per plant. Complete foliar coverage with herbicide is required for successful control of OWCF. This may be difficult in dense infestations of OWCF during the initial treatment due to over-lapping fronds of OWCF that are inter-twined with native plants. A follow-up treatment will be required wit hin six months after t he initial treatment for adequate control. Occurrences of OWCF, less than a few square meters in size, may be eliminated with a single herbi cide treatment, but larger infestations will require multiple treatments. The spores of OWCF were highly susc eptible to metsulfuron, but exhibited tolerance to imazapyr, glyphosate, fluroxyp yr, asulam, and triclopyr exposed in bathing solution. It is unclear if metsulfuron applied to mature OWCF sporophytes will inhibit spore development in the sori. Results of this study also indicated that OWCF gametophytes are highly susceptible to metsul furon. However, gametophytes exhibited tolerance to metsulfuron at concentrati ons up to 0.16 kg a.i. and developed into sporophytes at concentrations 0.04 kg a.i. These results indicate that OWCF exhibits the potential for resistance to metsulfuron especially at lower concentrations. Metsulfuron also exhibited residual activi ty on spore germination and growth in agar, indicating it may provide longer control of OWCF. Aerial and ground application of metsulfuron over Ever glades tree islands was more effective in reducing OWCF cover t han aerial application of glyphosate. Both metsulfuron and glyphosate tr eatments resulted in signif icant loss of native ground cover. Evergreen trees and shrubs exhibi ted a high tolerance to metsulfuron, and minimal impacts to the canopy cover were observed on tree islands treated with 131

PAGE 132

metsulfuron. Aerial treatment with glyphosat e on tree islands result ed in significant tree mortality and loss of canopy cover. T he dominant deciduous trees on tree islands, Cephalanthus occidentalis and Salix caroliniana suffered no damage to winter application of either metsul furon or glyphosate. It is recommended that aerial application of metsulfuron be appli ed initially for control of OWCF on tree islands. After the initial aerial treatment, additional aerial treatments are not recommended for at least three years to prevent furt her non-target impa cts. Follow-up gr ound treatments will be required for OWCF re-growth, un-treated ar eas of OWCF missed during the initial treatment, and numerous OWCF sporophytes that develop from spores following treatment. Re-treatment of OWCF on tree islands by ground crews is recommended at one year post aerial treatment and in future years. Management Recommendations 1. Aerial treatment of OW CF on Everglades tree islands should be carried out using metsulfuron. 2. Ground treatment of OWCF should be performed by rotating metsulfuron, glyphosate, triclopyr and other herbicides to limit the potential for populations becoming resistant. 3. Complete spray coverage of OWCF is essential to maximize herbicide translocation into the rhizom es and prevent re-sprouting. 4. Re-treatment and long term monitori ng will be required for adequate control of OWCF. 5. Monitoring and re-treatment should be carried out one year following the initial treatment, but preferab ly at six months. 6. Aerial treatment of OWCF with metsulfuron should be considered for any infestations of OWCF on Ev erglades tree islands that can be observed from the air. 7. Sites where threatened and endangered plants and animals are known to exist will require ground treatments. These sites should receive high funding priority. 132

PAGE 133

8. Primary funding for control of OW CF should be consol idated to geographical sections of Florida to limit the spread of s pores in the areas where it occurs at the highest densities. Current ly, the Treasure Coast regi on in southeast Florida should receive the highest funding for OWCF control. This geographical area includes Martin, Palm Beach, and St. Lucie Counties. As OWCF coverage is reduced in these counties, funding should be directed to the next geographical area with the largest infestations of OWCF. 9. Secondary funding for control of OWCF should be committed for expanding populations at the sout hern and northern ranges of its di stribution in Florida. Dade and Monroe counties at the southern range of OWCFs distribution should receive priority funding for OWCF control since more endangered habitat and listed species occur in these counties relative to northern counties. 10. Funding should be earmarked for monitori ng and re-treatment of all OWCF sites that receive init ial treatments. Additional Research 1. Evaluate the translocation patterns for t he maximum herbicide rates of glyphosate, metsulfuron, and triclopyr in OWCF. 2. Evaluate the efficacy of treatments using rotations of herbicides with different modes of action in field trials. 3. Examine the viability of OWCF spores fr om sporophytes with mature fertile leaflets treated with metsulfuron and other herbicides. 4. Evaluate the residual activity of mets ulfuron and other herbicides on OWCF spore germination and gametophyte survival using native soils. 5. Compare new sporophyte development at sites treated with metsulfuron and other herbicides. 6. Compare the efficacy of herbicide s on varying plant sizes of OWCF. 7. Evaluate the efficacy of aer ial application of 0.04 kg a.i. /187 l/ha of metsulfuron for control of OWCF on tree islands and it s effect on non-target vegetation. 8. Determine the time frame needed between the initial aerial application and a second aerial application when metsulfur on can be applied over tree islands without further reductions in ground and can opy cover and structural changes to tree islands. 9. Compare the efficacy of aer ially applied glyphosate and metsulfuron for control of OWCF in deciduous cypress and bayhead swamp habitat. 133

PAGE 134

10. Evaluate the effectiveness of aerial application for spot -treatment of OWCF on Everglades tree islands and other habitats. 11. Evaluate the potential of pl anting native plants in selected instances to reestablish native plant diversity and competition with OWCF. 12. Evaluate triclopyr for ae rial treatment of OWCF. 13. Continue testing novel or untested herbicides for efficacy on OWCF. 134

PAGE 135

135 APPENDIX PLANT SPECIES DOCUMENTED ON TREE ISLANDS AT A.R.M. LOXAHATCHEE N.W.R. TREAETED WITH METSULFRUON (0.08 AND 0.16 KG A.I.) AND GLYPHOSATE (2.80 AND 5.60 KG A.I.) AT PRE-TREATMENT (YEAR 1) TO THREE YEARS POST-TREATMENT. PRESENCE OF PLANTS INDICATED BY ASTERISK (*), CLIMAX SPECIES (+), AND NON-NA TIVE PLANTS (^). ALL OTHER PLANTS ARE CLASSIFIED AS RUDERAL SPECIES.

PAGE 136

136 Appendix Metsulfuron Glyphosate Control Plant Species Year 0.06 kg 0. 12 kg 3.0 kg 6.0 kg Untreated Andropogon virginicus L. 1 2 3 4 Asclepias incarnate L. 1 2 3 4 Bidens laevis (L.) 1 2 3 4 + Blechnum serrulatum Rich. 1 * 2 * 3 * 4 * Boehmeria cylindrical (L.) Sw. 1 2 3 * 4 + Cephalanthus occidentalis L. 1 * 2 * 3 * 4 * Ceratopteris pteridoides (Hook.) Hieron. 1 2 3 4

PAGE 137

Appendix c ont. Metsulfuron Glyphosate Control Plant Species Year 0.06 kg 0. 12 kg 3.0 kg 6.0 kg Untreated+ Cladium jamaicense Crantz 1 * 2 3 * 4 Crinum americanum L. 1 2 3 4 Cyperus haspan L. 1 2 3 * 4 * Drosera brevifolia Pursh 1 2 3 4 ^ Emilia fosbergii Nicolson 1 2 3 4 Erechtites hieracifolia (L.) Raf. Ex DC. 1 2 3 * 4 Eriocaulon compressum Lam. 1 2 3 4 137

PAGE 138

Appendix c ont. Metsulfuron Glyphosate Control Plant Species Year 0.06 kg 0. 12 kg 3.0 kg 6.0 kg Untreated Eupatorium capillifolium (Lam.) Small 1 2 3 4 Galium tinctorium L. 1 2 3 4 Habenaria floribunda Lindl. 1 2 3 4 Hypericum fasciculatum Lam. 1 2 3 * 4 + Ilex cassine L. 1 * 2 * 3 * 4 * Lachnanthes caroliana (Lam.) Dandy 1 2 * 3 * 4 * ^ Ludwigia peruviana (L.) H. Hara 1 2 3 4 * 138

PAGE 139

Appendix c ont. Metsulfuron Glyphosate Control Plant Species Year 0.06 kg 0. 12 kg 3.0 kg 6.0 kg Untreated Ludwigia repens J.R. Frost 1 * 2 * 3 * 4 * ^ Lygodium microphyllum (Cav.) R. Br. 1 * 2 * 3 * 4 * ^ Melaleuca quinquenervia (Cav.) S.T. Blake 1 2 3 4 Mikania scandens (L.) Willd. 1 2 3 4 + Myrica cerifera L. 1 * 2 * 3 * 4 * + Nephrolepis biserrata (Sw.) Schott 1 2 3 4 + Osmunda cinnamomea L. 1 * 2 * 3 * 4 * 139

PAGE 140

Appendix c ont. Metsulfuron Glyphosate Control Plant Species Year 0.06 kg 0. 12 kg 3.0 kg 6.0 kg Untreated+ Osmunda regalis L. 1 * 2 * 3 * 4 * Panicum hemitomon Schult. 1 2 3 4 Parthenocissus quinquefolia (L.) Planch. 1 2 3 4 + Peltandra virginica (L.) Schott & Endl. 1 * 2 * 3 * 4 * + Persea palustris (Raf.) Sarg. 1 * 2 * 3 * 4 * Phytolacca americana L. 1 2 3 4 Pluchea rosea R.K. Godfrey 1 2 3 * 4 140

PAGE 141

Appendix c ont. Metsulfuron Glyphosate Control Plant Species Year 0.06 kg 0. 12 kg 3.0 kg 6.0 kg Untreated Polygonum densiflorum Meisn. 1 2 3 4 Pontedaria cordata L. 1 2 3 4 * + Rapanea punctata (Lam.) Lundell 1 * 2 3 * 4 * Rhus copallinum L. 1 2 3 4 Rhynchospora inundata (Oakes) Fernald 1 2 3 4 * Rhynchospora microcarpa Baldwin ex A. Gray 1 2 3 4 Rhynchospora tracyi Britton 1 2 3 4 * 141

PAGE 142

Appendix c ont. Metsulfuron Glyphosate Control Plant Species Year 0.06 kg 0. 12 kg 3.0 kg 6.0 kg Untreated Rhynchospora chalarocephala Fernald & Gale 1 2 3 4 Saccharum giganteum (Walter) Pers. 1 2 3 4 + Salix caroliniana Michx. 1 2 3 4 Sarcostemma clausum (Jacq.) Roem. & Schult. 1 2 3 4 ^ Schinus terebinthifolius Raddi 1 2 3 4 Scoparia dulcis L. 1 2 3 4 + Smilax laurifolia L. 1 * 2 * 3 * 4 * 142

PAGE 143

Appendix c ont. Metsulfuron Glyphosate Control Plant Species Year 0.06 kg 0. 12 kg 3.0 kg 6.0 kg Untreated+ Thelypteris interrupta (Willd.) K. Iwats 1 2 3 * 4 * Triadenum virginicum (L.) Raf. 1 2 3 4 Utricularia spp. 1 2 3 4 + Vitis rotundifolia Michx. 1 * 2 * 3 * 4 * + Woodwardia virginica (L.) Sm. 1 * 2 * 3 * 4 * Xyris ambigua Beyr. ex Kunth 1 2 3 4 * 143

PAGE 144

144 LIST OF REFERENCES Anonymous. 2007. Accelerated invasive sp ecies control plan for A.R.M. Loxahatchee National Wildlife Refuge: 2007-2008. Unpublished report, South Florida Water Management District, West Palm Beach, Florida Ballesteros, D. and C. Walter s. 2007. Water properties in fern spores: sorption characteristics relating to water affini ty, glassy states, and storage stability. Journal of Experimental Botany 58:1185-1196. Banks, J. A. 1999. Gametophyte development in ferns. Annual Review of Plant Physiology and Plant Molecular Biology 50:163-186. Barrett, M. A., L. A. Brandt, and B. Thomas Jr. 2006. Monitori ng ground treatments of Old World climbing fern (Lygodium mi crophyllum) on the Arthur R. Marshall Loxahatchee National Wildlife Refuge: a follow-up report Wildland Weeds 10:47. Beckie, H. J., I. M. Heap, R. J. Smeda, and L. M. Hall. 2000. Screening forherbicide resistance in weeds. Weed Technology 14:428-445. Beckner, J. 1968. Lygodium microphyllum, another fern escaped in Florida. American Fern Journal 58:93-94. Biehl, T. 2004. Lygodium infestations in the Lake Wales Ridge. Wildland Weeds 7:4-5. Boughton, A. J. and R. W. Pemberton. 2009. Establishment of an imported natural enemy, Neomusotima conspurcatalis (Lepidoptera; Cram bidae) against an invasive weed, Old World climbing fern Lygodium microphyllum, in Florida. Biocontrol Science and Technology 19: 769-772. Brandt, L. A. 2004. Effectiv eness of different aerial spra y for control of Lygodium microphyllum on tree islands in the A. R.M. Loxahatchee NWR 24 month monitoring period. Unpublished report, USFWS, A.R.M. Loxahatchee NWR, Boynton Beach, Florida. Brandt, L. A. and D. W. Bl ack. 2001. Impacts of the introduced fern, Lygodium microphyllum, on the native veget ation of tree islands in the Arthur R. Marshall Loxahatchee National Wildlif e Refuge. Florida Scientist 64:191-196. Brandt, L. A., K. M. Portier and W. M. Kitchens. 2000. Patterns of change in tree islands in the Arthur R. Marshall Loxahat chee National Wildlife Refuge from 1950 to 1991. Wetlands 20:1-14. Brandt, L. A., D. Ecker, I. G. Rivera, A. Traut and F. J. Mazzotti. 2003. Wildlife and vegetation of bayhead islands in t he A.R.M. Loxahatchee National Wildlife Refuge. Southeastern Naturalist 2:179-194.

PAGE 145

Brown, V. M. 1984. A biosystematic study of the fern genus Lygodium in eastern North America. M. S. Thesis. Biology Depar tment, University of Central Florida. Orlando, Florida. Call, E. M., L. A. Brandt and D. L. Deangelis 2007. Old Wo rld climbing fern (Lygodium microphyllum) spore germination on natural substrates. Florida Scientist 70:5561. Canfield, R. H. 1941. Application of the line interc eption method in sampling range vegetation. Journal of Forestry 39:388-394. Carroll, E. W., O. J. Schwarz, and L. G. Hickok. 1988. Biochemical studies of paraquat-tolerant mutants of the fern Ceratopteris richardii. Plant Physiology 87:651-654. Chachalis, D. and K. N. Reddy. 2004. Pelargonic acid and rainfall effects on glyphosate activity in trumpetcreeper (C ampsis radicans). Weed Technology 18:66-72. Clark, D. W. 2002. The effect of Old Wo rld climbing fern [Lygodium microphyllum (Cav.) R. Brown] on south Florida cyp ress [Taxodium distichum (L.) Rich.] swamp plant and insect community stru cture. M. S. Thesis. Agronomy Department, University of Flori da. Gainesville, Florida. Clark, H. M. 1936. The morphology and anatomy of Lygodium japonicum. American Journal of Botany 23:405-413. Cole, A. W., and G. E. Coats. 1973. Ta ll morningglory germination response to herbicide and temperature. Weed Science 21:443-446. Conway, E. 1953. Spore and sporeling survival in bracken (Pteridium aquilinum (L.) Kuhn). Journal of Ecology 41: 289-294. Crone, E. E., M. Marler, and D. E. Pearson. 2009. Non-ta rget effects of broadleaf herbicide on a native perennial forb: a demographic framework for assessing and minimizing impacts. Journal of Applied Ecology 46: 673-682. De la Cretaz, A. L. and M. J. Kelty. 1999. Establishment and c ontrol of hay-scented fern: a native invasive species. Biological Invasions 1:223-236. Doyle, P. and M. Stypa. 2004. Reduced herbicide rates a Canadian perspective. Weed Technology 18:1157-1165. DuPont Corporation. 2003. DuPont Escort XP Herbic ide Special Local Need 24(c) Label. Available online < http://www.cdms.net/LDat/ld5QT008.pdf >. Accessed 4 January 2010. 145

PAGE 146

Egley, G. H. and R. D. Willi ams. 1978. Glyphosate and paraquat effects onweed seed germination and seedling emergenc e. Weed Science 26: 249-251. Fairbrother, A. and L. A. Kapustka. 2001. Low-dose, high-potency herbicides: a historical perspective of environmental conc erns to frame the issues. Pp. 117 in S. A. Ferenc, ed., Impacts of Low-Dose, High-Potency Herbicides on Nontarget and Unintended Plant Species. SETAC Press, Pensacola, Florida. Fernandez, H. and M. A. Revilla. 2003. In vitro culture of ornamental ferns. Plant Cell, Tissue, and Organ Culture 73:1-13. Ferriter, A. and T. Pernas. 2006. An explos ion in slow motion: tracking the spread of Lygodium microphyllum in Florida. Wildland Weeds 9:7-9. [FLAS] University of Florida Herbarium. 2008. Gainesville, FL. Available online < http://www.flmnh.ufl.edu/herbarium/ >. Accessed 12 December 2009. Flinn, K. M. 2006. Reproductive biology of three fern species may contribute to differential colonization success in post-agr icultural forests. American Journal of Botany 93:1289-1294. [FSU] Florida State University Herbarium. 2008. Tallahassee, Florida. Available online < http://herbarium.bio.fsu.edu/index.php >. Accessed 12 December 2008. [FTG] Fairchild Tropic al Gardens Herbarium. 2008. Miami, Florida. Available online< http://www.virtualherba rium.org/default.html >. Accessed 12 December 2009. Gandiaga, S., J. C. Volin, E. L. Kruger, and K. Kitajima. 2009. Effects of hydrology on the growth and physiology of an invasive exotic, Lygodium microphyllum (Old World climbing fern). Weed Res earch 49:283-290. Gardner, S. C. and C. E. Grue. 1996. Effects of Rodeo and Garlon 3A on nontarget wetland species in Central Washington. Environmental Toxicology and Chemistry 15:441-451. Geiger, D. R. and H. D. Bestman. 1990. Selflimitation of herbicide mobility by phytotoxic action. Weed Science 38:324-329. Gillman, F. S. 2007. The ecological significance of the herbaceous layer in temperate forest ecosystems. Bioscience 57: 845-858. Goolsby, J. A. 2004. Pot ential distribution of the inva sive Old World climbing fern, Lygodium microphyllum in No rth and South America. Natural Areas Journal 24:351-353. Goolsby, J. A., A. D. Wright, and R. W. Pe mberton. 2003. Ex ploratory surveys in Australia and Asia for natural enemies of Old World climbing fern, Lygodium microphyllum: Lygodiaceae. Bi ological Control 28:33-46. 146

PAGE 147

Goolsby, J. A., P. J. De Ba rro, J. R. Makinson, R. W. Pe mberton, D M. Hartley and D. R. Frohlich. 2006. Matching the origin of an invasive weed for selection of a herbivore haplotype for a biological control programme. Molecular Ecology 15:287-297. Gordon, D.R. 1998. Effects of invasive, non-indigenous plant s pecies on ecosystem processes: lessons from Florida. Ecological Applications 8:975-989. Groninger, J. W. and L. H. McCormick. 1991. Effects of sulfometuron on hay-scented fern spore germination. Canadian J ournal of Forest Research 21:942-943. Harms, C. T. and J. J. DiMaio. 1991. Primisulfuron herb icide-resistant tobacco cell lines. Applications of fluctuation test des ign to in vitro mutant selection with plant cells. Journal of Plant Physiology 137:513-519. Haughn, G. and C. R. Somerville. 1987. Selection for herbicide resistance at the whole plant level. Pp. 98-107 in H. M. LeBar on, R. O. Mumma, R. C. Honeycutt, and J. H. Duesing, eds., Biotechnology in Agricu lture Chemistry. American Chemical Society, Washington, D.C. Heap, I. 1997. The occurrence of herbicide resistance weeds worldwide. Pesticide Science 51:235-243. Henry, J. A., K. M. Portier and J. Coyne. 1994. The Clima te and Weather of Florida. Pineapple Press, Inc., Sarasota, Florida. Hickok, L. G. and R. M. Kir iluk. 1984. Effects of aux ins on gametophyte development and sexual differentiation in the fern Ce ratopteris thalictroides (L.) Brongn. Botanical Gazette 145:37-42. Hickok, L. G., T. R. Warne, and R. S. Fr ibourg. 1995. The bi ology of the fern Ceratopteris and its use as a model syst em. International Journal of Plant Sciences 156:332-345. Hobbs, R. J. and S. E. Humphries. 1995. An integrated approach to the ecology and management of plant invasions. Conservation Biology 9:761-770. Hobbs, R. J. and L. F. Huenneke. 1992. Disturbance, diversity, and invasion: implications for conservation. Conservation Biology 6:324-337. Hofstra, D. E., P. D. Cham pion, and T. M. Dugdale. 2006. Herbicide trials for the control of parrotsfeather. Journal of Aquatic Plant M anagement 44:13-18. Horvitz, C. C. and A. Koop. 2001. Remo val of non-native vines and post-hurricane recruitment of tropical hardwood forests of Florida. Biot ropica 33:268-281. 147

PAGE 148

Horvitz, C. C., J. B. Pascarella, S. McMann, A. Freedman, and R. H. Hofstetter. 1998. Functional role of invasive non-indi genous plants in hurricane-affected subtropical hardwood f orests. Ecol ogical Applications 8:947-974. Hoss, N. E., K. Al-Khatib, D. E. Peters on, and T. M. Loughin. 2003. Efficacy of glyphosate, glufosinate, and imazet hapyr on selected weed species. Weed Science 51:110-117. Hutchinson, T. F. and J. L. Vankat. 1997. Invasion by Lonicera maackii (Amur honeysuckle) in southwestern Ohio forests: community invisibi lity and effects on tree seedlings and herbs. Cons ervation Biology 11:1117-1124. Hutchinson, J.T. and K.A. Langel and. 2006. Survey of cont rol measures on Old World climbing fern (Lygodium microphyllum) in southern Florida. Florida Scientist 69:217-223. Hutchinson, J. T. and K. A. Langeland. 200 8. Response of selected nontarget native Florida wetland plant species to metsulfu ron methyl. Journal of Aquatic Plant Management 46:72-76. Hutchinson, J.T., G. E. MacDonald and K. A. Langelan d. 2007. The potential for herbicide resistance in non-native plants in Floridas natural areas. Natural Areas Journal 27:258-263. Hutchinson, J., A. Ferriter, K. Serbeso ff-King, K. Langeland, and L. Rodgers (eds.). 2006. Old World Climbing Fern (Ly godium microphyllum) Management Plan for Florida. South Florida Water Management District. West Palm Beach, FL. Available online < http:// www.fleppc.org >. Accessed 10 December 2008. Kalnay, P. A. and S. Glenn. 2000. Translocation of nico sulfuron and dicamba in hemp dogbane (Apocynum cannabinum). Weed Technology 14:476-479. Keary, I. P., C. Thomas, and E. S heffield. 2000. The effects of the herbicide asulam on the gametophytes of Pteridium aquilinum, Cryptogramma cris pa, and Dryopteris filix-mas. Annals of Botany 85:47-51. Khan, A. A. 1981. Effect of leaf position and plant age on the translocation of 14Cassimilates in onion. Journal of Agriculture Science 96:451-455. Kirkwood, R. C., R. Hetherington, T. L. Reynolds, and G. Marshall. 2000. Absorption, localization, translocation, and activity of glyphosate in barnyardgrass [Echinochloa crus-galli (L) Beauv.]: influence of herbicide and surfactant concentration. Pest M anagement Science 56:359-367. Klekowski, E. J., Jr. 1969. R eproductive biology of the Pteri dophyta III. A study of the Blechnaceae. Botanical Journal of the Linnean Society 62:153-169. 148

PAGE 149

Koger, C. H. and K. N. R eddy. 2005. Glyphosate efficacy, absorption, and translocation in pitted morningglory (I pomoea lacunose). Weed Science 53:277283. Kohler, E. A., C. S. Thro ssell, and Z. J. Reicher. 2004. 2,4-D rate response, absorption, and translocation of two ground ivy (Glechoma hederacea) populations. Weed Technology 18:917-923. Krebs, C.J. 1989. Ecological Methodology. Ha rper Collins Publishers, New York, NY. Kurumatani, M., K. Yagi, T. Murata, M. Tezuka, L. N. Mander, M. Nishiyama, and H. Yamane. 2001. Isolation and identifica tion of antheridiogens in the ferns, Lygodium microphyllum and Lygodium reticula tum. Bioscience, Biotechnology, and Biochemistry 65:2311-2314. Lakela, O. and R. W. Long. 1976. Ferns of Florida. Banyan Books, Miami, FL. Langeland, K. A. and M. L. Link 2006. Evaluation of metsul furon methyl for selective control of Lygodium microphyllum gro wing in association with Panicum hemitomon and Cladium jamaicense. Florida Scientist 69:149-156. Langeland, K. A., H. M. Cherry, C. M. Mc Cormick and K. A. Craddock Burks. 2008. Identification and Biology of Nonnative Plants in Floridas Natural Areas. IFAS Communication Services. University of Florida. Gainesville, Florida. Laufenberg, S. M., R. L. Shel ey, J. S. Jacobs, and J. Borkowski. 2005. Herbicide effects on density and biomass of Ru ssian knapweed (Acroptilon repens) and associated plant species. Weed Technology 19:62-72. Le Duc, M. G., R. J. Pakem an, and R. H. Marrs. 2003. Changes in the rhizome system of bracken subjected to long-term exper imental treatment. Journal of Applied Ecology 40:508-522. Le Duc, M. G., R. J. Pakem an, P. D. Putwain, and R. H. Marrs. 2000. The variable responses of bracken fronds to control tr eatments in Great Britain. Annals of Botany 85 (Sup. B):17-29. Lemmon, P. E. 1957. A new instrument for measuring forest overstory density. Journal of Forestry 55:667-668. Lindgren, P. M. F. and T. P. Sullivan. 2001. Influence of alternative vegetation management treatments on conifer plantat ion attributes: abundance, species diversity, and structural diversity. Forest Ecology and Management 142:163-182. Lott, M. S., J. C. Volin, R. W. Pemberton, and D. F. Austin. 2003. The reproductive biology of the invasive ferns Ly godium microphyllum and L. japonicum (Schizaeaceae): implications for invasive potential. American Journal of Botany 90:1144-1152. 149

PAGE 150

Loveless, C. M. 1959. A study of the v egetation in the Florida Everglades. Ecology 40:1-9. Lovell, S.T., L.M. Wax, M. J. Horak and D.E. Peterson. 1996. Imidazolinone and sulfonylurea resistance in biotypes of common waterhemp (Ama ranthus rudis). Weed Science 44:789-794. Lynch, R. L., H. Chen, L. A. Brandt and F. J. Mazzotti. 2 009. Old World climbing fern (Lygodium microphyllum) invasion in hurricane caused treefalls. Natural Areas Journal 29: 210-215. MacDougall, A.S. and R. Turki ngton. 2005. Are invasive species the drivers or passengers of change in degraded ecosystems? Ecology 86:42-55. Main, C. L., J. E. Beeler, D. K. Robi nson, and T. C. Mu eller. 2006. Growth, reproduction, and management of Chinese yam (Dioscorea oppositifolia).Weed Technology 20:773-777. Marrs, R.H. 1985. The effects of potent ial bracken and scrub control herbicides on lowland Calluna and grass heath communities in East Anglia, UK. Biological Conservation 32:13-32. Marrs, R.H., S. W. Johnson, and M. G. Le Duc. 1998. Control of bracken and restoration of heathland. VII. The response of bracken rhizomes to 18 years continuous bracken control or 6 years of control followed by recovery. Journal of Applied Ecology 35:748-757. McIntyre, G. I. 1962. Preliminary studies on the translocation of 14C-labelled herbicides in bracken (Pteridium aquilinum). Weed Research 2:51-59. Moldenke, H. N. 1944. A contribution to our knowledge of the wild and cultivated flora of Florida 1. American Midland Naturalist 32:529-590. Moran, R.C. 2004. A Natural History of Fe rns. Timber Press, Inc., Portland, Oregon. Mueller, R. J. 1982. Shoot morphology of the climbing fern Lygodium (Schizaeaceae): general organography, leaf in itiation, and branching. Botanical Gazette 143:319330. Mueller, R. J. 1983. Indeterminate growth and ramification of the climbing leaves of Lygodium japonicum. American Journal of Botany 70:682-690. National Hurricane Center. 2009. 2004 Atlant ic Hurricane Season. National Oceanic and Atmospheric Administration, National Weather Service, Miami, Florida. Available online < http://www.nhc.noaa.gov/ >. Accessed 2 October 2009. 150

PAGE 151

[NWRS] National Wildlife Refuge System. 2010. The United States Fish and Wildlife Services National Wildlife Refuges. Available online < http://www.fws.gov/refuges/ >. Accessed 15 June 2010. Nauman, C. E. and D. F. Austin. 1978. Spread of the exotic fern Lygodium microphyllum in Florida. American Fern Journal 68:65-66. Norsworthy, J. K., N. R Burgos, and L. R. Oliver. 2001. Differences in weed tolerance to glyphosate involve different me chanisms. Weed Technology 15:725-731. [NPS] National Park Service. 2010. Evergl ades: International Designations. Available online< http://www.nps.gov/ever/parknews/ internationa ldesignations.htm >. Accessed 9 March 2010. Page, C. N. 1979. The diversity of ferns an ecological perspective. Pp. 10-56, In: A. F. Dyer (ed.), The Experimental Biolog y of Ferns. Academic Press, New York. Pakeman, R. J. and R H. Ma rrs. 1992. The conservation va lue of bracken, Pteridium aquilinum (L.) Kuhn, dominated communiti es in the UK, and an assessment of the ecological impact of br acken expansion or its removal. Biological Conservation 62:101-114. Pakeman, R. J., M. G. LeDuc, and R. H. Marrs. 2000. Bracken dist ribution in Great Britain: strategies for its control and the sustainable management of marginal lands. Annals of Botany 85 (Supp B):37-46. Pakeman, R. J., J. L. Sma ll, M. G. Le Duc and R. H. Marrs 1998. An assessment of aerially applied asulam as a method of log-term bracken control. Journal of Environmental M anagement 53:255-262. Pakeman, R. J., R. H. Thwaites, M. G. Le Duc and R. H. Marrs. 2000. Vegetation reestablishment on land previously subject to control of Pteridium aquilinum by herbicide. Applied Veget ation Science 3:95-104. Pakeman, R. J., J. L. Sma ll, M. G. Le Duc and R. H. Ma rrs. 2005. Recovery of moorland vegetation after aerial spraying of bracken (Pteridium aquilinum (L.) Kuhn) with Asulam. Rest oration Ecology 13:718-724. Pemberton, R.W. 1998. T he potential of biological control to manage Old World climbing fern (Lygodium microphyllum), an invasive weed in Florida. American Fern Journal 88:176-182. Pemberton, R. W. 2003. Northward range extension in Florida of the invasive fern Lygodium microphyllum (Ly godiaceae). Sida 20:1761-1761. Pemberton, R.W. 2006. Taki ng down climbing fern biological control of Lygodium microphyllum. Wildla nd Weeds 9:10-11. 151

PAGE 152

Pemberton, R. W. and A. P. Ferriter. 1998. Old Worl d climbing fern (Lygodium microphyllum), a dangerous invas ive weed in Florida. American Fern Journal 88:165-175. Petrov, P. and R. H. Marrs. 2000. Follow-up methods for bracken control following an initial glyphosate application: the us e of weed wiping, cutting and reseeding. Annuals of Botany 85 (Sup. B):31-35. Pettitt, J. M. 1979. Ultra structure and cytochemistry of spore wall morphogenesis. Pp. 213-252 in A. F. Dyer, ed., The Experimental Biology of Fe rns. Academic Press, New York. Prather, T. S., J. M. DiTomaso, and J. S. Holt. 2000. Histo ry, mechanisms, and strategies for prevention and management of herbicide resistant weeds. Proceedings of the California Weed Science Society 52:155-163. Preston, C. and S. B. Powles. 2002. Evolution of herbi cide resistance in weeds: initial frequency of target site-based resistance to acetolactate synthase-inhibiting herbicides in Lolium rigidum. Heredity 88:8-13. Ranker, T. A. and H. A. Houst on. 2002. Is gametophyte sexuality in the laboratory a good predictor of sexuality in nature? American Fern Journal 92:112-118. Ray, T. B. 1986. Sulfonylurea herbicides as inhibitors of amino acid biosynthesis in plants. Trends in Biochem ical Sciences 11:180-183. Reddy, K. N. 2000. Factor s affecting toxicity, absorpt ion, and translocation of glyphosate in redvine (Brunnichia ov ata). Weed Techno logy 14:457-462. Rice, P. M., J. C. Toney, D. J. Bedunah, an d C. E. Carlson. 1997. Plant community diversity and growth form responses to herbicide applications for control of Centaurea maculosa. Journal of Applied Ecology 34: 1397-1412. Roberts, D. 1996. The monster of Hobe So und or Why does Lygodium microphyllum like it so well here. Pp. 29-30 in M. Bodle, ed., Proceedings from Climbing (Twining) Ferns and Invasive Exotic Vine Workshop. South Florida Water Management District, West Palm Beach, Florida. Roberts, D. 1998. Old World climbing fern research and mitigation at Jonathan Dickinson Park. Resource Management Notes 9:30-32. Roberts, R. E., R. O. Woodbury, and J. Popenoe. 2006. Vascular plants of Jonathan Dickinson State Park. Fl orida Scientist 69:288-327. Russell, A. E., J. W. Raich, and P. M. Vitous ek. 1998. The ecology of the climbing fern Dicranopteris linearis on the windward M auna Loa, Hawaii. Journal of Ecology 86:765-779. 152

PAGE 153

[SAS] Statistical Analy sis Software. 2002. SAS for Windows, Version 9.2. Cary, NC: SAS Institute. Schneller, J. J. 2008. Anther idiogens. Pp. 134-158 in T. A. Ranker and C. H. Haufler, eds., Biology and Evolution of Ferns and Lycophytes. Cambridge University Press, United Kingdom. Schneller, J. J., C. H. H aufler, and T. A. Ranker. 1990. Antherid iogen and natural gametophyte populations. American Fern Journal 80:143-152. Smith, A. R., K. M. Pryer, E. Schuettpelz, P. Korall, H. Schneider, and P. G. Wolf. 2006. A classification of extant ferns. Taxon 55:705-731. Smith, A. R., K. M. Pryer, E. Schuettpelz, P. Korall, H. Schn eider, and P. G. Wolf. 2008. Fern classification. Pp. 417-467 in T. A. Ranker and C. H. H aufler, eds., Biology and Evolution of Ferns and Lycophytes. Cambridge University Press, United Kingdom. Soltis, D. E. and P. S. Soltis. 1992. The distribution of selfing rates in homosporous ferns. American Journal of Botany 79:97-100. Stewart, G. B., A. S. Pullin and C. Tyler. 2007. The effectiveness of Asulam for bracken (Pteridium aquilinum) control in the United Kingdom: a meta-analysis. Environmental M anagement 40:747-760. Stocker, R. K., A. Ferriter, D. Thayer, M. Rock, and S. Smith. 1997. Old World climbing fern hitting south Florida below the belt. Wildland Weeds 1:6-10. Stocker, R. K., R. E. Miller Jr., D. W. Black, A. P. Fe rriter, and D. D. Thayer. 2008. Using fire and herbicide to control Ly godium microphyllum and effects on a pine flatwoods plant community in south Flor ida. Natural Areas Journal 28:144-145. Swami, P. and V. Raghavan. 1980. Control of morphogenesis in the gametophyte of a fern by light and growth hormones. C anadian Journal of Botany 58:1464-1473. Taylor, J. 2006. Managem ent of Old World climbing fern in Everglades National Park. Pp. 67-69 in Hutchinson, J., A. Ferriter, K. Serbesoff-King, K. Langeland, and L. Rodgers, eds., Old World Climbing Fern (Lygodium microphyllum) Management Plan for Florida. South Florida Water Management District, West Palm Beach, FL. Available online < http:// www.fleppc.org >. Accessed 10 December 2009. Thomas, B. 2006. A.R.M. Loxahatchee Nati onal Wildlife Refuge. Pp. 73-78 in Hutchinson, J., A. Ferriter, K. Ser besoff-King, K. Langeland, and L. Rodgers, eds., Old World Clim bing Fern (Lygodi um microphyllum) Management Plan for Florida. South Florida Water Management District, West Palm Beach, FL. Available online < http:// www.fleppc.org >. Accessed 10 December 2009. 153

PAGE 154

Thomas, B. Jr. and L. A. Brandt. 2003. Monitoring ground treatments of Old World climbing fern (Lygodium microphyllum) on the Arthur R. Marshall Loxahatchee NWR. Wildland Weeds 6:9-11. Tobe, J. D., K. C. Burks, R. W. Cantrell, M. A. Garland, M. E. Sweeley, D. W. Hall, P. Wallace, G. Anglin, G. Nelson, J. R. Cooper D. Bickner, K. Gilbert, N. Aymond, K. Greenwood, and N. Raymond. 1998. Florida Wetland Plants An Identification Manual. Florida Department of Environmental Protection. Tallahassee, Florida. Toth, L. A. 2009. Hydrological restor ation of wetlands on the Kissimmee River Floodplain: complementarity of an herbicide treatm ent of the exotic vine Lygodium microphyllum. Flor ida Scientist 72:103-120. Tranel, P. J., and T. R. Wright. 2002. Resistance of weeds to ALS-inhibiting herbicides: what have we lear ned? Weed Science 50:700-712. Tryon, R. M. and G. Vitale 1977. Evidence for ant heridiogen production and its mediation of a mating system in natural populations of fern gametophytes. Botanical Journal of t he Linnean Society 74:243-249. Twiss, E. M. 1910. The prothallia of Aneimia and Lygodi um. Botanical Gazette 49:168-181. Ugarte, C. A., L. A. Brandt, S. Melvin, F. J. Mazzotti and K. G. Rice. 2006. Hurricane impacts on tree islands in Arthur R. Marshall Loxahatchee National Wildlife Refuge, Florida. Southeastern Naturalist 5:737-746. Unland, D. R., K. Al-Khatib, and D. E. Peterson. 1999. Interactions between imazamox and diphenylethers. Weed Science 47:462-466 Veerasekaran, P., R. C. Kirkwood, and W. W. Fletcher. 1977. Studies on the mode of action of asulam in bracken (Pteridium aquilinum L. Kuhn) I. Absorption and translocation of (14C) asul am. Weed Research 17:33-39. Vencill, W. K. (ed.). 2002. Herbicide Handbook (8th ed.). Weed Science Society of America. Lawrence, Kansas. Voeller, B. 1971. Developm ental physiology of fern gam etophytes. Bioscience 21:266270. Vogelmann, T., A. Bassel, and J. Miller. 1981. Effects of microt ubule-inhibitors on nuclear migration and rhizoid differentia tion in germinating fern spores (Onoclea sensibilis). Protoplasma 109:295-316. Volin, J. C., M. S. Lott, J. D. Muss, and D. Owens. 2004. Predicting rapid invasion of the Florida Everglades by Old World c limbing fern (Lygodi um microphyllum). Diversity and Distributions 10:439-446. 154

PAGE 155

Walker, E. R. and L. R. O liver. 2008. Translocation and absorption of glyphosate in flowering sicklepod (Senna obtus ifo lia). Weed Science 56:338-343. Watkins, J. E., Jr., M. K. Mack, and S. S. Mulkey. 2007. Gametophyte ecology and demography of epiphytic and terrestrial tr opical ferns. American Journal of Botany 94:701-708. Wetzel, P.R., A.G. van der Valk, S. Newm an, D.E. Gawlik, T.T. Gann, C.A. CoronadoMolina, D.L. Childers, and F.H. Sklar. 2005. Maintaining tree islands in the Florida Everglades: nutrient redistribution is the key. Frontiers in Ecology and the Environment 3:370-376. Wikstrm, N., P. Kenrick, and J. C. Vogel. 2002. Schizaeaceae: a phylogenetic approach. Review of Palaeobotany and Palynology 119:35-50. Williams, C.B. 1964a. Patterns in the Balance of Nature. Academic Press, London, United Kingdom. Williams, R.D. 1964b. Assim ilation and translocation in perennial grasses. Annals of Botany 28:419-426. Wilson, K. A. 2002. Contin ued pteridophyte invasion of Hawaii. American Fern Journal 92:179-183. Wolf, S. 1993. Effect of leaf age on photosynthesis, carbon transport and carbon allocation in potato plants. Potato Research 36:253-262. Woodmansee, S. W. and J. L. Sadle. 2005. New occurrences of Schizaea pennula Sw. in Florida. American Fern Journal 95:84-87. Woodmansee, S. W., K. Bradley, and S. H odges. 2005. Systematic Reconnaissance Flights and Exotic Plant Species Mapping at Selected National Wildlife Refuges in Florida. Final Report. Submitted to U.S. Fish and Wildlif e Service, Sanibel, Florida. Wu, Y. K. Rutchey, N. Wang, an d J. Godin. 2006. The spatia l pattern and dispersion of Lygodium microphyllum in the Evergl ades wetland ecosystem. Biological Invasions 8:1483-1493. Wunderlin, R. P. and B. F. Hansen. 2003. Gui de to the Vascular Plants of Florida, 2nd edition. University Press of Florida, Gainesville, Florida. Wunderlin, R. P., and B. F. Hans en. 2008. Atlas of Florida Va scular Plants. Institute for Systematic Botany, Universi ty of South Florida, Tampa. [S. M. Landry and K. N. Campbell (application develop ment), Florida Center fo r Community Design and Research]. Available online < http://www.plantatlas.usf.edu/ >. Accessed: 12 December 2008. 155

PAGE 156

Young, F. L., D. R. Gealy, and L. A. Morrow 1984. Effect of her bicides on germination and growth of four grass weeds. Weed Science 32: 489-493. Zar, J. H. 1999. Biostatisi cal Analysis, 4th Edition. Prentice Hall International Inc., Englewood Cliffs, New Jersey. 156

PAGE 157

157 BIOGRAPHICAL SKETCH Jeffrey T. Hutchinson was born in Salisbu ry, North Carolina. He graduated from West Rowan High School in 1980 and attended N.C. State University from 1980-1982. In 1983, he attended Western Carolina University for two semesters. From 1984-1990 he served as an electrical technician in the United States Marine Corps and was stationed in South Carolina, T ennessee, Georgia, California, Hawaii, Japan and Korea. From 1990-1992, Jeff attended Central Florida Community College in Ocala and received an A.S. in biology in 1992. From 1992-1995, he attended the University of Florida and received a B.S. in wildlife ecology with a minor in forestry in 1995. From 1996-1998, he attended the University of Kentucky and received his M.S. in forestry in 1998 under the direction of Dr. Michael Lacki. His thesis focused on characterizing the roost sites and habitat characteristics of forest dwelling bats. Jeff worked as a district biologist for the Florida Park Service fr om 1998-2002 and as land manager at Archbold Biological Station fr om 2002-2004. In 2004, Jeff began his dissertation work on OWCF under the direction of Dr. K enneth Langeland at the University of Floridas Center for Aquatic and Invasive Plants. His career inte rests include all aspects of natural resource conservation.