BIOLOGY, HOST SPECIFICITY AND IMPACT OF ISCHNODEMUS VARIEGATUS, AN HERBIVORE OF HYMENACHNE AMPLEXICAULIS
By
RODRIGO DIAZ
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
2008
1
© 2008 Rodrigo Diaz
2
To Veronica Manrique, my best friend and companion
3
ACKNOWLEDGMENTS
I extend my deepest appreciation to Dr. Bill Overholt for his support and friendship. Dr.
Overholt provided a constructive scientific environment for the development of my dissertation.
His role as mentor has been paramount due the constant conversations and critical discussions about different subjects such as science, politics, religion, sports, economy and foreign affairs. I cannot ask for a more nutritious intellectual environment than Dr. Overholt’s friendship.
I thank Drs. Jim P. Cuda, Paul Pratt and Alison Fox for providing useful insights in
biological control and weed ecology. Don Schmitz from Florida Department of Environmental
Protection facilitated the funding for my research assistantship at the University of Florida.
Catherine Corbet from the Charlotte Harbor Estuary program provided funding for several field
experiments. I greatly appreciate their support.
I am also grateful to Dr. Overholt’s lab members for their continuous support and
friendship. I thank Diana Cordeau, Yordana Valenzuela, Eric Morgan, Brianne Schobert, Freddy
Soza, Douglas Gonzalez, Veronica Manrique, Jackie Markle, Brittany Evans, Ana Samayoa,
Fabian Diaz and Larry Markle. Paul Benshoff and Diana Donaghy from Myakka River State
Park provided support and ideas during the development of this project. I thank also the members
of the volunteer program from the Myakka River State Park.
The realization of this project could not have been possible without the great help of the
staff of the Indian River Research & Education Center at Fort Pierce, Department of Entomology
and Nematology in Gainesville and the quarantine officers of the Biological Control Research
and Containment Laboratory. Finally, I am gratefull to my parents, Rogelio and Maggie, for their
unconditional support in my life and to my brother, Daniel, for the good conversations about
happiness and family.
4
TABLE OF CONTENTS
page
ACKNOWLEDGMENTS ...... 4
LIST OF TABLES...... 9
LIST OF FIGURES ...... 11
ABSTRACT...... 14
CHAPTER
1 LITERATURE REVIEW ...... 16
Introduction...... 16 Why Do Exotic Plants Become Invasive?...... 17 Enemy Release Hypothesis ...... 17 Evolution of Increased Competitive Ability Hypothesis (EICA)...... 18 Increased Resource Availability Hypothesis...... 18 Resource-Enemy Release Hypothesis ...... 19 Biology and Ecology of Hymenachne amplexicaulis...... 19 Hymenachne amplexicaulis Invasion in Australia...... 21 Hymenachne amplexicaulis in Florida...... 22 Classification of Hymenachne amplexicaulis...... 23 Biology of Ischnodemus variegatus ...... 24 Goals and Hypotheses...... 25
2 TEMPERATURE-DEPENDENT DEVELOPMENT, SURVIVAL AND POTENTIAL DISTRIBUTION OF ISCHNODEMUS VARIEGATUS (HEMIPTERA: BLISSIDAE), AN HERBIVORE OF WEST INDIAN MARSH GRASS (HYMENACHNE AMPLEXICAULIS)...... 35
Introduction...... 35 Materials and Methods ...... 37 Source of Ischnodemus variegatus and Hymenachne amplexicaulis...... 37 Laboratory Studies...... 37 Development Time and Survival...... 38 Developmental Rate and Degree Day Requirement...... 39 Linear Model ...... 39 Nonlinear Model...... 39 Weather Data from Florida...... 40 Calculation of Degree-days and Number of Generations for GIS Analysis ...... 40 Generation of GIS Map for Prediction of I. variegatus Generations in Florida...... 40 Results...... 41 Size and Behavior of Stages...... 41 Immature Survival and Developmental Time ...... 43
5
GIS Mapping of I. variegatus Generations in Florida...... 44 Discussion...... 45
3 HOST SPECIFICITY OF ISCHNODEMUS VARIEGATUS, AN HERBIVORE OF WEST INDIAN MARSH GRASS (HYMENACHNE AMPLEXICAULIS)...... 57
Introduction...... 57 Materials and Methods ...... 60 Origin and Maintenance of Organisms...... 60 Test Plants ...... 60 No-choice Nymphal Development...... 61 No-choice and Choice Oviposition ...... 61 Field Colonization of Potted Plants...... 63 Spill-over to Co-occurring Species ...... 63 Statistical Analysis ...... 64 Results...... 65 No-choice Nymphal Development...... 65 No-choice and Choice Oviposition ...... 65 Field Colonization of Potted Plants...... 66 Spill-over to Co-occurring Species ...... 66 Discussion...... 67
4 LIFE HISTORY PARAMETERS OF ISCHNODEMUS VARIEGATUS (SIGNORET) (HEMITERA: BLISSIDAE) REARED ON TWO CLOSELY RELATED GRASSES...... 82
Introduction...... 82 Materials and Methods ...... 86 Origin and Maintenance of Organisms...... 86 Development of First Instars on Cuttings...... 87 Development of Third Instars on Growing Tips and Cuttings...... 87 Adult Longevity and Oviposition Test in No-choice Conditions...... 88 Oviposition Choice Test...... 88 Population Growth Under No-choice Conditions ...... 89 Statistical Analysis ...... 89 Results...... 90 Development of First Instars on Cuttings...... 90 Development of Third Instars on Growing Tips and Cuttings...... 90 Adult Longevity and Oviposition Trial in No-choice Conditions...... 90 Oviposition Choice Test...... 91 Population Growth Under No-choice Conditions ...... 91 Discussion...... 92
6
5 POPULATION DYNAMICS OF ISCHNODEMUS VARIEGATUS, AN ADVENTIVE HERBIVORE OF WEST INDIAN MARSH GRASS (HYMENACHNE AMPLEXICAULIS) IN FLORIDA ...... 100
Materials and Methods ...... 102 Study Sites...... 102 Myakka River State Park Marsh Description...... 103 Environmental Variables...... 103 Myakka River State Park Insect and Plant Sampling Methodology ...... 104 Myakka River State Park Plant Damage Methodology...... 104 Fisheating Creek Insect and Plant Sampling Methodology ...... 105 Between-tiller Distribution of Life Stages ...... 105 Intraspecific Association Between Adults and Nymphs ...... 106 Prediction map of H. amplexicaulis in Florida...... 106 Statistical Analysis ...... 107 Results...... 107 Hymenachne amplexicaulis and Ischnodemus variegatus Population Dynamics at Myakka River State Park and Fisheating Creek ...... 107 Egg Dynamics and Parasitism at Myakka River State Park...... 108 Presence of Beauveria bassiana at Myakka River State Park...... 108 Between-plant Distribution and Intraspecific Association of Adults and Nymphs ...... 109 Seasonal Dynamics of Tiller and Water Height and Insect Density at Myakka River State Park During the Outbreak in 2004...... 109 Comparison of Hymenachne amplexicaulis Plant Parameters, Damage and Ischnodemus variegatus Density During October 2006 ...... 110 Discussion...... 110 Seasonality of Ischnodemus variegatus Populations...... 110 Occurrence of Aggregation in Ischnodemus variegatus ...... 113 Feeding Damage and Plant Performance ...... 115
6 DAMAGE OF ISCHNODEMUS VARIEGATUS (HEMIPTERA: BLISSIDAE) TO HYMENACHNE AMPLEXICAULIS UNDER THE INFLUENCE OF VARIOUS FERTILIZER AND SOIL MOISTURE LEVELS...... 131
Introduction...... 131 Materials and Methods ...... 132 Insect and Plant Cultures...... 132 Effect of Fertilizer and I. variegatus on H. amplexicaulis Performance...... 133 Effect of Soil Moisture and I. variegatus Density on H. amplexicaulis Performance ..134 Effect of Soil Moisture, I. variegatus Density and Fertilizer on H. amplexicaulis Performance ...... 135 Statistical Analysis ...... 135 Results...... 136 Effect of Fertilizer and I. variegatus on H. amplexicaulis Performance...... 136 Effect of Soil Moisture and I. variegatus Density on H. amplexicaulis Performance ..136 Effect of Soil Moisture, I. variegatus Density and Fertilizer on H. amplexicaulis Performance ...... 137
7
Discussion...... 137
7 GENERAL CONCLUSIONS AND PERSPECTIVES ...... 149
APPENDIX
A IMPACTS OF HYMENACHNE AMPLEXICAULIS ON PLANT AND ARTHROPOD COMMUNITIES...... 156
Introduction...... 156 Material and Methods...... 157 Study Area...... 157 Sampling...... 158 Statistical Analysis ...... 159 Results...... 159 Marsh Characteristics ...... 159 Plant Community and Biomass Accumulation...... 160 Macroinvertebrates Summer ...... 160 Macroinvertebrates Fall...... 161 Discussion...... 162
LIST OF REFERENCES...... 174
BIOGRAPHICAL SKETCH ...... 187
8
LIST OF TABLES
Table page
1-1. Wetland plant diversity at risk by the invasion of Hymenachne amplexicaulis in the Myakka River State Park (Paula Benshoff unpublished data)...... 29
1-2. Host association of Ischnodemus spp. in Florida ...... 30
2-1. Mean developmental time in days (mean ± SE) and surviving individuals of immature I. variegatus stages at 10 constant temperatures...... 50
2-2. Linear regression parameter estimates describing the relationship between temperature and developmental rates (1/D) of I. variegatus stages...... 51
1 2-3. Parameter estimates (a, To, TL) for the Brière-1 nonlinear model describing the relationship between temperature and developmental rate (1/D) of I. variegatus stages...... 52
3-1. No-choice nymphal survival of I. variegatus on test plants...... 72
3-2. Mean number of I. variegatus per stem and plant parameters collected from spill-over studies in Florida during 2006 ...... 75
4-1. Development and reproductive performance parameters of I. variegatus exposed to H. amplexicaulis and H. acutigluma...... 96
5-1. Plant parameters, damage and I. variegatus numbers in different wetlands at Myakka River State Park during October 2006...... 117
6-1. Hymenachne amplexicaulis variables measured 30 days after the inoculation of I. variegatus...... 141
6-3. Effect of Ischnodemus variegatus density and soil moisture on different Hymenachne amplexicaulis parameters. Second Experiment after 16 days...... 142
6-2. Effect of Ischnodemus variegatus density and soil moisture on different Hymenachne amplexicaulis parameters after 14 days ...... 142
6-4. Factorial analyses of the effect of soil moisture, fertilizer dose and insect density on Hymenachne amplexicaulis seedling biomass, height and damage...... 143
A-1. Water quality data collected in native (N) and Hymenachne (H) sites in three marshes at Myakka River State Park ...... 167
A-2. Average emergent stems per unit of area and cover in native (N) and Hymenachne (H) sites1 in three marshes at Myakka River State Park...... 167
A-3. Macroinvertebrates collected from aquatic and aerial samples during Summer 2006...... 168
9
A-4. Total count of arthropods and diversity indexes. Summer 2006...... 170
A-5. Macroinvertebrates collected from aquatic and aerial samples during Fall 2006 ...... 171
A-6. Total count of arthropods and diversity indexes. Fall 2006...... 173
10
LIST OF FIGURES
Figure page
1-1 Current distribution of Hymenachne amplexicaulis in Florida...... 31
1-2. Current distribution of Hymenachne amplexicaulis in the neotropics. Source: TROPICOS - Catalogue of New World Grasses. 2008 ...... 31
1-3. Hymenachne amplexicaulis infestation in Queensland, Australia...... 32
1-4. Monoculture of Hymenachne amplexicaulis in a marsh periodically flooded by the Myakka River. Myakka River State Park, Sarasota Co. August 2003...... 33
1-5. Native distribution of H. donacifolia (A), H. amplexicaulis (B), H. grumosa (C) and H. pernambucensis (D)...... 34
2-1. Life stages of I. variegatus and length in mm (mean ± SD): (A) Egg mass on H. amplexicaulis culm, 2.97 ± 0.13 (n=25) ; (B) first instar, 1.45 ± 0.28 (n=23) ; (C) second instar, 2.70 ± 0.39 (n=47); (D) third instar, 3.06 ± 0.31 (n=42); (E) fourth instar, 3.95 ± 0.32 (n=53); (F) fifth instar, 5.45 ± 0.43 (n=46); (G) female, 7.23 ± 0.56 (n=28); male, 6.05 ± 0.22 (n=49), (H) female sclerites at ventral tip of abdomen; (I) male sclerites at ventral tip of abdomen; (J) scent glands in thorax of adult...... 53
2-2. Proportion survival of I. variegatus stages at constant temperatures (0C)...... 54
2-3. Developmental rates (1/D) of I. variegatus at different temperatures (0C). Linear regression of eggs to adult stages and observed and predicted values by Brière-1 nonlinear model...... 55
2-4. Geographical information system map showing the predicted number of generations of I. variegatus in Florida...... 56
3-1. Laboratory set-up for I. variegatus host range tests. (A) no-choice development, (B) no- choice oviposition, (C) multiple-choice oviposition, and (D) female laying eggs...... 76
3-2. Mean (± SE) adult longevity and number of eggs laid under non-choice conditions. Plant species with different letters within a variable are significantly different (P<0.05)...... 77
3-3. Mean (± SE) number of eggs laid at three different adult densities...... 78
3-4. Open field test results at site 1. Adult and nymphs: (F = 13.52 ; d.f. = 3, 19; P <0.001); eggs (F = 5.28; d.f.= 3,19; P<0.001)...... 79
3-5. Open field test results at site 2. Adult and nymphs: (F = 14.0; d.f. = 3, 26; P <0.001); eggs:(F= 8.99, d.f.= 3,26; P <0.001)...... 80
11
3-6. Age distribution as mean (± SE) of all the stages of I. variegatus and I. sallei on their hosts found at marsh 4 ...... 81
4-1. Cumulative number of I. variegatus adults emerging from third instar nymphs per sampling date on H. amplexicaulis and H. acutigluma ...... 97
4-2. Cumulative number of I. variegatus eggs collected under no-choice conditions ...... 98
4-3. Number of live I. variegatus present in H. amplexicaulis and H. acutigluma after 60 days...... 99
5-1. Location of study sites in Central Florida. A) Myakka River State Park (Sarasota Co.), B) Fisheating Creek (Glades Co.)...... 118
5-2. Study sites at Fisheating Creek (Glades Co.) during 2003 to 2006...... 119
5-3. Study sites at Myakka River State Park (Sarasota Co.) during 2002 to 2006...... 120
5-4. Environmental variables at MRSP. Dashed line in gage height graph represent flooding conditions...... 121
5-5. Number of I. variegatus adults and nymphs present per tiller at three different marshes at MRSP (Sarasota Co.) ...... 122
5-6. Number of I. variegatus eggs and proportion of parasitism in three different marshes at Myakka River State Park (Sarasota Co.) ...... 123
5-7. Proportion of tillers with I. variegatus infested with B. bassiana at Myakka River State Park...... 124
5-8. Regressions of the log variance on the log mean for I. variegatus adults, nymphs and eggs...... 125
5-9. Relationship of proportion of tiller infested to the mean number of I. variegatus adults, nymphs and eggs...... 126
5-10. Seasonal dynamics of tiller and gage height and I. variegatus density during 2004 ...... 127
5-11. Regression of accumulative number of adults and nymphs and leaf color score during 2004...... 128
5-12. Regression of number of adults and nymphs with leaf red score during Summer and Fall of 2004 at Myakka River State Park...... 129
5-13. Model prediction of climate suitability for Hymenachne amplexicaulis using herbarium specimens from New York and Missouri Botanical Garden...... 130
12
6-1. Schematic representation of soil moisture levels used for the factorial experimens: a) dry: no water, b) saturation: water maintained 4 cm from the base of the pots, and c) flood: water level maintained at 1 cm below the top of the pot...... 143
6-2. Changes in Hymenachne amplexicaulis parameters recorded from plants with fertilizer (filled circles) or control (empty circles)...... 144
6-3. Hymenachne amplexicaulis leaf damage with fertilizer (a) and control (b)...... 145
6-4. Changes in H. amplexicaulis parameters under control (no bugs), medium (2 fourth instars) and high (4 fourth instars) I. variegatus densities...... 146
6-5. Changes in H. amplexicaulis parameters under control (no bugs), medium (4 fourth instars) and high (6 fourth instars) I. variegatus densities...... 147
6-6. Effect of soil moisture, fertilizer and I. variegatus density on Hymenachne amplexicaulis weight (a), height (b) measured 21 days after insect inoculation; and, interactions of soil moisture and fertilizer level (c) and insect density with fertilizer level (d) on leaf damage...... 148
A-1. Location of marshes at Myakka River State Park (27016’N, 82016’W)...... 165
A-2. Mean abundance of arthropods in native and Hymenachne sites. Summer 2006...... 166
A-3. Mean abundance of arthropods in native and Hymenachne sites. Fall 2006...... 166
13
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
BIOLOGY, HOST SPECIFICITY AND IMPACT OF ISCHNODEMUS VARIEGATUS, AN HERBIVORE OF HYMENACHNE AMPLEXICAULIS
By
Rodrigo Diaz
May 2008
Chair: William A. Overholt Major: Entomology and Nematology
West Indian marsh grass Hymenachne amplexicaulis (Rudge) Nees (Poaceae) is a major wetland weed in central and south Florida. In 2000, the sap-feeding bug Ischnodemus variegatus
(Signoret) (Hemiptera: Blissidae) was discovered feeding on H. amplexicaulis in Sarasota Co.
The objectives of this study were to determine the temperature-dependent developmental time, host range, impact on H. amplexicaulis and seasonal dynamics of I. variegatus in Florida.
Complete egg and nymphal mortality occurred at temperatures ≤20.50C and at 380C. The lower
thresholds for complete development (egg to adult) estimated with a linear and nonlinear model
were 14.6 and 17.40C, respectively. The total degree-days required to complete development
estimated by the linear model were 588. Based on these predictions, the insect can complete 3-5
generations per year in areas where it is currently present in Florida. Developmental host range
was examined on 57 plant species in seven plant families. Complete development was obtained
on H. amplexicaulis (23.4% survivorship), Panicum anceps (2.2%), Paspalum repens (0.4%),
and Thalia geniculata (0.3%). Oviposition on suboptimal host species was positively correlated
with I. variegatus density under multiple choice conditions. Field experiments indicated that H.
amplexicaulis had higher densities of I. variegatus than other species. Spill-over to suboptimal
14
hosts occurred in an area where H. amplexicaulis was growing under poor conditions and there was a high density of I. variegatus. Thus, laboratory and field studies demonstrated that I. variegatus performed better on H. amplexicaulis compared to other hosts, and that suboptimal hosts could be colonized temporarily. Seasonal dynamics of I. variegatus in Central Florida during the sampling period revealed a biannual cycle in deep water open marshes. High infestations of I. variegatus were present during the late fall of 2002, 2004 and 2006 at Myakka
River State Park. Feeding damage during these periods was recognized by an accumulation of anthocyanins in the leaves. However, this loss of photosynthetic area was not associated with reduction in seed number or plant vigor. Greenhouse experiments demonstrated that H. amplexicaulis seedling performance was negatively affected by the feeding damage of I. variegatus nymphs, low soil moisture and poor fertilization.
15
CHAPTER I LITERATURE REVIEW
Introduction
Humans have been intentionally and unintentionally spreading exotic species for hundreds of years. However, the rate of arrival of exotics has increased dramatically due to modern trade, travel, and technology in such a way that globalization facilitates and intensifies the spread of exotic species (Meyerson and Mooney 2007). Exotic plants in particular cause severe problems to native ecosystems worldwide because they alter ecosystem processes, compete with native species and cause economic losses (Vitousek et al. 1996, DiTomaso 2000).
Pimentel et al. (2005) estimated that the economic impact of invasive species in USA is
$125 billion per year with invasive plants accounting for $34 billion. Invasive plants occupy over
40 million hectares in the United States and are estimated to be spreading at a rate of 1.2 million hectares per year (National Invasive Species Council 2001). The Ecological Society of America outlined a series of recommendations for federal, state and local governments, which included increasing measures for prevention and detection of new invasions, early control and reduced spread of new invasions, and provision of a national center to coordinate activities (Lodge et al.
2006).
The rate of arrival of exotic plants in Florida is probably higher compared to other states.
Wunderlin and Hansen (2003) reported 1,316 exotic plants species as naturalized in Florida, with
138 species being serious threats to natural areas (FLEPPC 2007). Of those, 67 are considered highly invasive (Category I) because they are disruptive to native plant communities. Several factors contribute the exotic plant problem in Florida. For instance, the constant inflow of people and goods is due to the presence of major airports, tourist attractions and cargo ports that are considered entry points for invasive species. Commercial nurseries also are a major source of
16
exotic plants because importation laws are too permissive. Pemberton (2000b) showed that plants sold by nurseries over extended periods were more likely to become naturalized. Florida’s environment also may provide adequate conditions for the establishment and colonization of exotic plants. Temperature conditions in Central and South Florida allowed the establishment and colonization of several subtropical species such as Schinus terebinthifolius Raddi, Melaleuca quinquenervia (Cav.) S.T.Blake, Casuarina equisetifolia L. and Eichhornia crassipes (Mart.)
Solms. Florida’s watersheds are particularly vulnerable to exotic plants because of their abundance, potential nutrient enrichment, interconnectivity and boat traffic. The annual expenditure to prevent and control invasive species in Florida during the fiscal year 2003-2004 by state agencies was $103.8 million (DEP 2008).
Why Do Exotic Plants Become Invasive?
Of all the plants that arrive in a new habitat, only a small percentage become invasive.
Researchers have formulated several hypotheses to explain the invasiveness of certain exotic plants:
Enemy Release Hypothesis
The enemy release hypothesis attributes the invasiveness of a plant in a new habitat to the release from herbivores and diseases that kept it under control in the native range (Williams 1954, Keane and Crawley 2002). Most studies testing this hypothesis rely on the difference in richness and abundance of natural enemies between invaded and native ranges (Wolfe 2002; DeWalt et al.
2004). However, selective pressures to which an exotic plant is exposed in the new habitat could vary tremendously in intensity and, consequently, it is difficult to point to a lack of specialist herbivores as having major predictive variable for invasion.
17
Evolution of Increased Competitive Ability Hypothesis (EICA)
This hypothesis predicts that invasive plants in the absence of natural enemies should
allocate more energy to growth and reproduction and less energy to defense. This is particularly
relevant for biological control of weeds since specialist herbivores collected in the native range
might perform better in the adventive range due the poor defenses of the invader plant. Blossey
and Notzold (1995) evaluated the differences in growth and reproduction of Lythrum salicaria L.
in a reciprocal transplant experiment using seeds from France and New York. The authors also
assessed the performance of two specialist biological control agents fed foliage of these two
populations. They found that L. salicaria grew bigger and produced more seeds and it was more
heavily attacked in New York (adventive range) than in the native range.
Studies have focused on the difference between native versus invaded populations using
the following variables: a) content of qualitative defensive compounds (b) changes in
morphological features, c) changes in reproductive output, d) susceptibility to herbivore attack, e) tolerance-resistance studies and f) herbivory pressure (Cipollini et al. 2005, Franks et al.
2008). Most of the studies only show differences in phenotypic variability and not changes at the genome level. Some published studies either support (Agrawal et al. 2005), or partially support
EICA (reviewed by Colautti et al. 2004) whereas others provide no evidence for this hypothesis
(Franks et al. 2008, Joshi and Vrieling 2005).
Increased Resource Availability Hypothesis
The resource availability hypothesis suggests that plant colonization is facilitated by high resource availability (e.g. light, water, soil, nutrients) due to poor resource uptake by competing species or high resource supply (Davis et al. 2000).
18
Resource-Enemy Release Hypothesis
The resource-enemy release hypothesis predicts that enemy release and increased resource availability may act in concert to cause invasion (Blumental 2005). The argument for this interaction is that fast growing, high-resource requiring species also tend to be highly susceptible to enemies. When introduced to a new range, these species are likely to benefit from both high resource availability and enemy release (Blumental 2005). Predictions of this hypothesis are a) enemy release and resource availability may often act in concert to facilitate invasion, b) exotic species will have a greater advantage over native species in high- than low- resource environments, c) successful exotic species will have high-resource traits such as fast growth, low metabolic cost of tissue production and little investment in defense, relative to coexisting native species, d) evolution of increased competitive ability will be most important for well-defended species adapted to low resource habitats (Blumental 2006). This hypothesis has not yet been tested.
Biology and Ecology of Hymenachne amplexicaulis
Large infestations of exotic grasses can reduce the biodiversity in aquatic ecosystems.
Recent studies in wetlands demonstrate that exotic grasses are capable of simplifying the plant diversity and reducing or changing the arthropod community (Herrera and Dudley 2003,
Houston and Duivenvoorden 2002, Tadley and Levin 2001, Posey 1988). These changes can alter trophic structure, affecting habitat usage by birds, fish and other vertebrates.
Hymenachne amplexicaulis is native to South America and the West Indies and has spread to most countries of the neo-tropics. Figure 1-1 shows the current distribution of H. amplexicaulis in the neotropics.
The pathway and timing of the introduction of this grass into Florida is uncertain; however, the first herbarium record was from a ponded pasture in Palm Beach County in 1957
19
(University of Florida Herbarium). This suggests that the grass could have been intentionally
introduced as cattle forage. The next record was from a wet pasture in Collier county in 1977
(University of Florida Herbarium). Current records confirm that H. amplexicaulis is present in
wetlands and rivers in 16 counties in Florida (Fig. 1-2).
Hymenachne amplexicaulis is well adapted to wetland environments, especially changing
in water levels. In the Brazilian Pantanal, its native range, H. amplexicaulis grows within four
plant formations: marsh ponds, waterlogged basins, tall grasslands and forest edges (Pinder and
Ross 1998). Observations of marshes in Myakka River State Park suggest that when subject to
inundation, H. amplexicaulis is capable of fast stem elongation, increase in foliage volume and rapid nodal adventitious root production (R. Diaz observations). In Venezuela, Tejos (1978) found a positive relationship between H. amplexicaulis growth and depth of flooding and that biomass production ranged from 5,911 - 18,162 t/ha/yr during the flood period and from 5,553 -
7,836 t/ha/yr during the dry season. Tejos also found that H. amplexicaulis grew 8.22 cm/day during the flooding period.
Invasion of H. amplexicaulis is favored by aggressive mechanisms of reproduction and dispersal. In Australia, a single inflorescence can produce more than 4000 seeds (Tropical Weeds
Research Centre 2003) with approximately 98% viability (Lyons 1996). Another method of reproduction is through vegetative growth (stolons). For example, Hymenachne amplexicaulis experimental colonies at the Indian River Research & Education Center in Fort Pierce are easily propagated by planting small pieces of stems containing at least one node. Moreover, ponded pastures in Australia were easily established by simply casting pieces of the grass from boats
(Lukacs 1996). Summer floodwaters in Florida can transport H. amplexicaulis seeds and stolons great distances through watersheds complicating management programs. A factor that may
20
facilitate the establishment and dominance of H. amplexicaulis in these rivers is the high nutrient
enrichment of surface water (especially with nitrogen and phosphorous) due to runoffs from
agricultural fields and geological deposits of phosphate (Charlotte Harbor Environmental Center,
Inc 2002).
Hymenachne amplexicaulis Invasion in Australia
Hymenachne amplexicaulis was imported into Australia from Venezuela in the late 1970s
to evaluate its potential as a forage for cattle (CSIRO 1973), but its invasiveness was noted in the
mid 1990s (Csurhes et al. 1999). In 1980, the grass was already causing problems for sugarcane
farmers, and park managers confirmed its presence in natural wetlands. Based on its native
distribution and climatic adaptations, H. amplexicaulis has the potential to invade all seasonally
flooded wetlands in Australia (Charleston 2006). Large infestations can be found in cane fields,
water storages facilities, irrigation/drainage channels, roadside ditches and natural lagoons (Fig.
1-3). It exhibits prolific growth in wetlands that are subjected to high nutrient and sediment
influx from upstream agricultural land. Australia has a congeneric species H. acutigluma
(Steudel) Guilliland that shares the same ecological niche as H. amplexicaulis. Both species are morphologically similar but can be distinguished by the base of the leaf which clasps the culm in
H. amplexicaulis, but not in H. acutigluma. The use of H. acutigluma as a forage grass in ponded pastures in Australia has been suggested since 1990. However, in flooded conditions, H. acutigluma has a lower photosynthetic rate and reduced photosynthetic leaf area compared to H. amplexicaulis (Kibbler and Bannisch 1999b), and this could explain the invasiveness of the latter in Queensland wetlands.
In Australia’s natural wetlands, H. amplexicaulis is threatening the habitat of magpie geese (Anseranas semipalma Latham) by displacing native plants such as Oryza spp., Eleocharis spp. and Ischaemum spp. The Queensland Fish Industry reports that this grass is expected to
21
reduce total run-off volumes, and either diminish or stop natural water flows, and consequently
reduce fish recruitment success. In a recent study, Houston and Duivenvoorden (2002) found that
wetlands invaded with H. amplexicaulis had lower plant species richness, greater relative
abundance of exotic fish and altered macroinvertebrate community structure. Currently, H.
amplexicaulis covers more than 50,000 ha of wetlands, and the Australian government has
named this plant a “Weed of National Significance” (Csurches et al. 1999).
Hymenachne amplexicaulis in Florida
Invasion of exotic species poses a serious threat to Florida’s sensitive ecosystems.
International trade, tourism, agricultural and urban disturbance have increased the probability of
establishment of exotic plants. Exotic plant species in Florida wetlands spread rapidly due to
floods, large interconnected waterway systems, and increased use of commercial and recreational
boats. Fertilizer and sediment runoff from agricultural lands and waste water from beef and dairy
cattle operations contribute to successful establishment of aquatic exotic plants.
The Myakka River flows through 45 square miles of Myakka River State Park which is
one of the natural areas in Florida most affected by the invasion of H. amplexicaulis. The dominance of H. amplexicaulis is particularly severe during summer and early fall when the water levels are high (Fig. 1-4). Previous plant surveys in Myakka River State Park demonstrated that wetlands supported a great diversity of species (Table 1-1) (Paula Benshoff unpublished data), which was being displaced by H. amplexicaulis, especially in marshes located next to the
Myakka River.
Simplification of wetland ecosystems due to the invasion of H. amplexicaulis could have severe impacts on the native fauna. Diverse aquatic habitats are places for feeding, resting, refuge and reproduction for wading birds, wood stork, snail kite, killifishes, live bearers, juvenile
22
sunfishes, southern leopard frog, pig frog, green tree frog, American alligator and American
crocodile among others.
Methods for controlling H. amplexicaulis are primarily limited to the use of registered herbicides. However, according to invasive plant managers from the South Florida Water
Management District, herbicides offer only short term control of H. amplexicaulis, as there is
substantial regrowth from stolons and seeds after herbicide treatment (M. Bodle, person. comm.).
Therefore, combinations of strategies may provide more effective control of this invasive grass.
Classification of Hymenachne amplexicaulis
Based on the Catalogue of New World Grasses (2008) the current higher classification of
H. amplexicaulis is as follows:
Kingdom Plantae
Division Magnoliophyta
Class Monocotyledonae (Liliopsida)
Subclass Commelinidae
Order Poales
Family Poaceae
Subfamily Panicoideae
Tribe Paniceae
Subtribe Paspalinae
Genus Hymenachne P. Beauv.,
Species Hymenachne amplexicaulis (Rudge) Nees 1829
The genus Hymenachne belongs to the tribe Paniceae and contains seven species
worldwide which inhabit mostly tropical wetlands. There are two native species in Asia, one in
Australia, and four in America; several species are considered valuable as forage. Native species
23
from the neotropics include H. donacifolia (Raddi) Chase, H. grumosa (Nees) Zuloaga and H. pernambucencis (Spreng.) Zuloaga. All Hymenachne spp. are perennial, decumbent aquatics, bisexual and have a C3 type photosynthetic pathway (Watson and Dallwitz 1992). Collections of the four Hymenachne spp. in South America reveal a sympatric distribution across wetland habitats (Fig. 1-5; Catalogue of New World Grasses 2008). There is no information about the phylogenetic relationships among species in this genus.
Biology of Ischnodemus variegatus
In 2000, the neotropical bug Ischnodemus variegatus (Signoret) (Hemiptera: Blissidae)
was discovered feeding and causing severe damage to H. amplexicaulis at Myakka River State
Park. Scientists from the Florida Department of Agriculture and Consumer Services (FDACS)
identified I. variegatus as a new record for the continental USA (Halbert 2000). The native
distribution of I. variegatus includes Central and South America and collection records indicate
H. amplexicaulis as the only host (Baranowski 1979, Slater 1987). Like other species in the
Blissidae family (Hemiptera classification follows Henry 1997), Ischnodemus feeds on the sap of
monocotyledonous plants (Slater 1976). Population outbreaks of this insect in central and south
Florida occur from August to November. Feeding effects of I. variegatus diminish carbon
dioxide assimilation, growth rate and biomass of H. amplexicaulis (Overholt et al. 2004). Despite
its potential as a fortuitous biological control agent of H. amplexicaulis, there are no studies that
address the basic biology of I. variegatus or its host range (Brambila and Santana 2004).
Host associations of species in the genus Ischnodemus have been described by Slater
(1976, 1987), Slater and Wilcox (1973) and Baranowski (1979). Ischnodemus spp. feed and
reproduce on monocotyledonous plants and some species show a remarkable degree of host
specificity, including several marsh grass specialists (Harrington 1972, Slater 1976, Slater and
Wilcox 1973). Slater (1976) reports that species in the genus Ischnodemus have been found
24
breeding on the grass subfamilies Festucoideae, Panicoideae, Eragrostoideae and Arundinoideae.
Table 1-2 shows the host associations of Ischnodemus spp. found in Florida. Slater and Wilcox
(1973) speculated that the majority of elongated, slender bodied species (i.e. Ischnodemus) tend
to be associated with grasses that occur in relatively moist habitats.
The genus Ischnodemus is the largest and more widely distributed of the family Blissidae
(Slater 1976). Members are characterized by elongate, parallel sided bodies, closed fore coxal cavities, terete (cylindrical) antennae, a straight apical corial margin, forewing membrane well differentiated in texture from the clause and corium (Slater and Wilcox 1969). The genus
Ischnodemus was classified as ‘Type I’ body shape by Slater (1976) which includes species with elongate, slender body shape that is usually slightly flattened dorsoventrally. Slater mentioned that a slender body is advantageous for insect living on the stems of grasses or related plants with elongate, narrow growth form. The neotropical species are mostly present in South America with greatest number concentrating in Argentina and southern Brazil (Slater and Wilcox 1969).
Natural enemies of I. variegatus in Florida include the egg parasitoid Eumicrosoma sp.
(Hymenoptera: Scelionidae) (Terry Nuhn person. communication), and the entomopathogen
Beauveria bassiana (Balsamo) Vuillemin (Deuteromycotina: Hyphomycetes) that have been reported also as common natural enemies of the blissid bug Blissus antillus Leonard (Krueger et al. 1992, Coracini and Samuels 2002).
Goals and Hypotheses
The overall goal of this study was to better understand the biology, host specificity, population dynamics and impact of I. variegatus on H. amplexicaulis. A series of studies were conducted under laboratory, greenhouse and field conditions to assess the following hypotheses:
Hypothesis 1: Temperature will allow multiple generations of I. variegatus per year in Florida with more generations in the south and fewer in the north
25
Objective 1: Determine the developmental rates of I. variegatus at different temperatures.
Objective 2: Generate a map predicting the number of generations/yr of I. variegatus across
Florida.
Exotic organisms face a myriad of limitations upon arrival to a new habitat. One of the most important factors is temperature. I hypothesized that I. variegatus will experience more generations in south versus central Florida. Among the outcomes of temperature-dependent development studies in arthropods is the calculation of the degree days required to complete one generation and the establishment of lower and upper temperature threshold for development.
This information should help to understand the potential areas for establishment of. I. variegatus and the insect’s activity throughout the growing season. The availability of Geographical
Information Systems (GIS) software with interpolation functions allowed the production of maps showing the number of generations I. variegatus may experience in Florida.
Hypothesis 2: Due to longer coevolutionary history of I. variegatus and H. amplexicaulis
in their native range, native plants in Florida will be unsuitable hosts for development
and oviposition.
Objective 1: Determine the fundamental host range of I. variegatus.
Objective 2: Assess the reproductive performance of I. variegatus under no-choice and choice conditions.
Objective 3: Examine the extent to which non-host plants are attacked by I. variegatus under field conditions.
Host range testing is an essential step for quantifying risks posed by exotic herbivores to native and economically important plant species. The arrival of I. variegatus to Florida wetlands triggered several questions regarding its specificity and potential impacts to the invasive grass H.
26
amplexicaulis as well as to native species within the wetland communities and to cultivated grasses. Therefore, I quantified the fundamental host range of I. variegatus under controlled conditions and, through field experiments, monitored spill-over to suboptimal hosts.
Hypothesis 3: A specialist grass herbivore will incur higher fitness costs on a novel congeneric host compared to its natural host.
Objective 1: Compare the life history parameters of I. variegatus when feeding solely on H. amplexicaulis and H. acutigluma.
Successful colonization and establishment of specialized exotic herbivores in new geographic areas is determined in part by the availability of their primary host. In the case of insects used for classical weed biological control, the occurrence of congeners of the target species creates serious concerns due to higher risk of non target damage (Paynter et al. 2008,
Pemberton 2000a). Local congeneric hosts often share morphological and chemical characteristics that could elicit feeding or oviposition responses in the agent. Host specificity testing performed under quarantine conditions evaluates the performance of the agent on the primary host and its closest relatives (Wapshere 1989). Comparisons of life history parameters of specialized herbivores reared on old association hosts and potential new associations provide baseline information about host suitability, and possibilities for non-target effects (van Klinken
2000). I hypothesized that feeding on a novel host H. acutigluma will incur higher fitness costs to I. variegatus.
Hypothesis 4: The seasonal dynamics of I. variegatus are controlled by abiotic and biotic factors.
Objective 1: Quantify the seasonal dynamics of I. variegatus in Central Florida.
Monitoring the population dynamics of exotic insect herbivores in the adventive range is a major step towards understanding their potential ecological impacts (Louda et al. 2003).
27
Information collected provides baseline data about damage inflicted to the host plant, the seasonality of the populations, presence of local natural enemies, and the influence of biotic and abiotic factors as sources of mortality, among others. I established transects at Myakka River
State Park and monitored the population of I. variegatus over the course of 5 years. Information obtained was critical to understanding the presence of outbreaks during summer and fall and its implication for spill-over to other grass species.
Hypothesis 5: Hebivory by I. variegatus, soil hydration, and nutrient availability will affect the performance of H. amplexicaulis.
Objective 1: Evaluate the effect of I. variegatus density, water and fertilization levels on the performance of H. amplexicaulis seedlings.
Hymenachne amplexicaulis performance is quite variable depending mostly on water fluctuations, access to sun, sediment nutrients and presence of herbivores. I studied the effect of
I. variegatus density, water and fertilization levels on the performance of H. amplexicaulis seedlings. This study allowed the recognition of important plant variables that could be affected under different scenarios.
28
Table 1-1. Wetland plant diversity at risk by the invasion of Hymenachne amplexicaulis in the Myakka River State Park (Paula Benshoff unpublished data).
Family Species Family Species ACANTHACEAE 1 LAMIACEAE 2 ADOXACEAE 1 LEMNACEAE 2 ALISMATACEAE 4 LENTIBULARIACEAE 7 AMARANTHACEAE 2 LENTIBULARIACEAE 1 AMARYLLIDACEAE 2 LYCOPODIACEAE 1 APOCYNACEAE 2 LYTHRACEAE 3 AQUIFOLIACEAE 1 MALVACEAE 2 ARACEAE 1 MARANTACEAE 1 ARALIACEAE 5 MELASTOMATACEAE 2 ASTERACEAE 18 MENYANTHACEAE 1 BLECHNACEAE 2 MYRICACEAE 1 BRASSICACEAE 1 NYMPHAEACEAE 3 CAMPANULACEAE 1 ONAGRACEAE 9 CERATOPHYLLACEAE 1 ORCHIDACEAE 1 CLUSIACEAE 6 OSMUNDACEAE 1 CONVOLVULACEAE 1 PARKERIACEAE 1 CORNACEAE 1 POACEAE 33 CUPRESSACEAE 1 POLYGALACEAE 8 CYPERACEAE 37 PONTEDERIACEAE 2 DROSERACEAE 1 POTAMOGETONACE 1 ERIOCAULACEAE 5 RUBIACEAE 2 FABACEAE 5 SALVINIACEAE 2 GENTIANACEAE 5 SCROPHULARIACEAE 6 HAEMODORACEAE 1 TURNERACEAE 1 HALORAGACEAE 4 TYPHACEAE 2 HYDROCHARITACEAE 4 URTICACEAE 1 IRIDACEAE 1 XYRIDACEAE 7 JUNCACEAE 5 TOTAL 222
29
Table 1-2. Host association of Ischnodemus spp. in Florida
Species Plant host (breeding) Geographic distribution Source Spartina bakeri, S. Slater and Baranowski I. badius Primary costal species alterniflora (1990). Slater and Baranowski I. brunipennis Panicum hemitomon Across peninsular Florida (1990). Slater and Baranowski I. conicus, Spartina bakeri North-West Florida (1990). Thalia geniculata, Canna Slater and Baranowski I. fulvipes Central and South Florida indica, Mussa spp. (1990). Slater and Baranowski I. lobatus Unknown South-West Florida (1990) Cyperus aggregatus, Slater and Baranowski I. praecultus Coastal South Florida Cyperus ligularis (1990). Echinochloa sp, Paspalidium Slater and Baranowski I. robustus Central and South Florida geminatum (1990). Slater and Baranowski I. rufipes Cyperus odoratus Central and South Florida (1990). Slater and Baranowski I. sallei Thalia geniculata Central and South Florida (1990). Sacciolepis striata, Panicum Slater and Baranowski I. slossoane Coastal Florida agrostoides (1990). I. variegatus Hymenachne amplexicaulis Central and South Florida Baranowski 1979.
30
Counties infested with Hymenachne amplexicaulis
Figure 1-1 Current distribution of Hymenachne amplexicaulis in Florida reported from Wunderlin and Hansen 2008, voucher specimens and R. Diaz (person. observ.).
Figure 1-2. Current distribution of Hymenachne amplexicaulis in the neotropics. Source: TROPICOS - Catalogue of New World Grasses. 2008
31
Babinda, Queensland
Figure 1-3. Hymenachne amplexicaulis infestation in Queensland, Australia (Photo credit: Csurhes et al. 1999)
32
Figure 1-4. Monoculture of Hymenachne amplexicaulis in a marsh periodically flooded by the Myakka River. Myakka River State Park, Sarasota Co. August 2003.
33
Figure 1-5. Native distribution of H. donacifolia (A), H. amplexicaulis (B), H. grumosa (C) and H. pernambucensis (D). Source: TROPICOS - Catalogue of New World Grasses. 2008
34
CHAPTER 2 TEMPERATURE-DEPENDENT DEVELOPMENT, SURVIVAL AND POTENTIAL DISTRIBUTION OF ISCHNODEMUS VARIEGATUS (HEMIPTERA: BLISSIDAE), AN HERBIVORE OF WEST INDIAN MARSH GRASS (HYMENACHNE AMPLEXICAULIS)
Introduction
Hymenachne amplexicaulis (Rudge) Nees (Poaceae) (West Indian marsh grass) is a robust, stoloniferous, semi-aquatic, perennial grass, native to the neotropics. The timing and pathway of introduction of this plant into Florida are unknown but its quality as forage suggests that the introduction may not have been accidental. The grass is also established in Indonesia
(Holm et al. 1979) and in Australia (Csurches et al. 1999) where it is considered a weed of national significance. The Florida Exotic Pest Plant Council listed the grass as a Category I species, which are invasive exotics that are altering native plant communities by displacing native species, changing community structures or ecological functions, or hybridizing with natives (FLEPPC 2007). The aggressive growth of H. amplexicaulis is due in part to rapid adaptation to changes in water levels (Kibbler and Bahnisch 1999a), high production of stolons and perhaps the absence of effective natural enemies. Once the grass invades a wetland, it forms monotypic stands 2-3m high with complete canopy cover. At the end of the growing season, this results in a massive accumulation of biomass. The grass disperses by seeds, which are produced in large quantities, and broken stolons, both of which can travel great distances during flooding events. Negative impacts of the grass in Australia affect the sugarcane industry, water resources, fisheries and ecotourism (Csurches et al. 1999). Plant managers in Florida and Australia find it challenging to control this grass with herbicides due to regrowth from below ground stolons
(Csurches et al. 1999).
Hymenachne amplexicaulis is considered a valuable forage grass in the neotropics, particularly in Mexico, Cuba and Venezuela. Important forage characteristics of this grass
35
include high digestibility, high nitrogen content and adaptation to changes in water levels (Anten
et al. 1998, Kibbler and Bahnisch 1999a). In the Brazilian pantanal, H. amplexicaulis occurs
within four habitats: marsh ponds, waterlogged basins, tall grasslands and forest edges (Pinder
and Ross 1998). Observations in marshes at Myakka River State Park, Sarasota Co., Florida
(27.20 N, 82.20 W) suggest that when subject to inundation, H. amplexicaulis is capable of rapid stem elongation, increase in foliage volume and rapid nodal adventitious root production (R.
Diaz, pers. observ.). Kibbler and Bahnisch (1999a) demonstrated that rapid elongation of the stem maintains the leaves above the water allowing emergent leaves to function at full photosynthetic capacity. In Venezuela, Tejos (1978) found a positive relationship between H. amplexicaulis growth and depth of flooding and that biomass production ranged from 5,911 -
18,162 t/ha/yr during the flood period and from 5,553 - 7,836 t/ha/yr during the dry season.
In 2000, the neotropical bug Ischnodemus variegatus (Signoret) (Hemiptera: Blissidae)
was discovered feeding and causing severe damage to H. amplexicaulis at Myakka River State
Park. Scientists from the Florida Department of Agriculture and Consumer Services (FDACS)
identified I. variegatus as a new record for the continental USA (Halbert 2000). The native
distribution of I. variegatus includes Central and South America and collection records indicate
H. amplexicaulis as the only host (Baranowski 1979, Slater 1987). Like other species in the
Blissidae family (Hemiptera classification follows Henry 1997), Ischnodemus feeds on the sap of
monocotyledonous plants (Slater 1976). Population outbreaks of this insect in central and south
Florida occur from August to November. Feeding effects of I. variegatus diminish carbon
dioxide assimilation, growth rate and biomass of H. amplexicaulis (Overholt et al. 2004). Despite
its potential as a fortuitous biological control agent of H. amplexicaulis, there are no studies that
address the basic biology of I. variegatus or its host range. The purpose of this study was to
36
determine temperature-dependent developmental times and survival, and with this information
generate a map depicting the predicted number of generations/yr of I. variegatus across Florida.
This study is an initial step towards understanding the thermal requirements for I. variegatus
establishment and population growth.
Materials and Methods
Source of Ischnodemus variegatus and Hymenachne amplexicaulis
Ischnodemus variegatus and H. amplexicaulis were collected in Myakka River State
Park, and Fisheating Creek, Glades Co., Florida (26.50 N, 81.70W) and maintained at the
Biological Control Research and Containment Laboratory (BCRCL), Fort Pierce, Florida.
Stolons of H. amplexicaulis were planted in two liter pots and placed in large trays filled with water to maintain permanent flooding conditions. Pots received one tablespoon of Osmocote®
(Scotts, Marysville, Ohio, USA) after transplanting and weekly applications of Miracle-Grow® water soluble fertilizer (Scotts, Marysville, Ohio, USA). Potted plants were placed in small, mesh screened cages (0.90 m x 0.90 m x 0.90 m) located within a walk-in rearing room maintained at 25-300C, 50-80% RH and a 14:10 L:D photoperiod. Field-collected I. variegatus
were released in these cages and monitored every other day for nymphal survival and
colonization. The maintenance of I. variegatus genetic variability was ensured by addition of field-collected individuals to the colony three times a year. Voucher specimens of the plant and the insect were deposited in the Florida herbarium (accession number 208823) and the Florida
State Collection of Arthropods (accession number E2002-6139), respectively.
Laboratory Studies
The lengths of eggs, nymphs and adults were measured from randomly collected individuals from the insect colony. Pictures of individuals placed in a sealed Petri dish were taken through a microscope using a digital camera fitted with Automontage® software
37
(Synchroscopy, Frederick, Maryland, USA) and measured with the image processing software
ImageJ (http://rsb.info.nih.gov/ij/). Length of eggs was measured along the longest axis.
Nymphs and adults were measured from the tip of the rostrum to the most distal point of the
abdomen. Behavioral observations were described from insect colonies and in field settings.
Observations were performed only during the day at least once every two weeks for a two-year
period.
Development Time and Survival
Temperature development studies of I. variegatus were conducted in environmental
chambers at ten constant temperatures (8± 0.5, 13± 0.5, 18± 0.5, 20.5± 0.5, 23± 0.5, 25.5± 0.5,
28± 0.5, 30.5± 0.5, 33± 0.5 and 38± 0.5 0C). Relative humidity and photoperiod were kept constant at 80% and 14:10 L:D, respectively. Environmental variables were confirmed with
HOBO® data loggers placed in each chamber. Ten adult couples were placed in a small cage
(0.3 m x 0.3 m x 0.3 m) and given stems of H. amplexicaulis for feeding and oviposition. Fresh
eggs (ca. 1 d old) were collected from the stems and transferred individually to small (5 cm) Petri
dishes containing moist filter paper. Fifty eggs were placed at each temperature treatment. Egg
development was monitored daily and hatching dates recorded.
Eggs collected from the adult colony were monitored daily and newly hatched nymphs
were used for the nymphal development study. First instars were placed singly in 250 cm3 vials containing a small section of H. amplexicaulis whorl which was placed upright in wet sand. The center of the vial lid was removed in a circular shape and replaced by fine mesh to allow gas exchange. Nymphs were transferred to fresh plant material every two days using a fine brush.
Fifty individuals were exposed to each temperature treatment. Nymphal molts, confirmed by the presence of exuviae, and survival were recorded between 8 and 11 AM every other day until the last individual molted to the adult stage.
38
Developmental Rate and Degree Day Requirement
Developmental time at different temperatures was analyzed using the general linear
model procedure (PROC GLM; SAS Institute 1999) for each instar separately as well as the total
immature stages combined. Whenever significant (P < 0.05) F-values were obtained, means were
separated using the Student-Neuman-Keuls (SNK) test (SAS Institute 1999).
Linear Model
For the egg, nymphal and total immature stages (egg and nymphal stages combined), the
linear portion (20-330C) of the developmental rate curve [R(T) = a + bT] was modeled using the
least squares linear regression (PROC GLM; SAS Institute 1999) where T was temperature, and
a and b were estimates of the intercept and slope, respectively. The temperatures 18 and 380C were not included in the regression analysis since their values were not part of the linear portion of the curve. The base temperature threshold was estimated by the intersection of the regression line at R(T) = 0, T0 = -a/b. Degree-day requirements for each stage were calculated using the
inverse slope of the fitted linear regression line (Campbell et al. 1974)
Nonlinear Model
The nonlinear relationship between developmental rate r(T) and temperature T was fitted
to the Brière model which allows the estimation of the upper and lower developmental thresholds
1/2 (Brière et al. 1999). The Brière-1 model is defined as R(T) = a T (T-T0)(TL-T) ; where R is the rate of development and is a positive function of the rearing temperature T, T0 is the base
temperature threshold, TL is the lethal (upper) temperature threshold, and a is an empirical constant (Brière et al. 1999). The developmental rate of I. variegatus was modeled using the
Marquardt algorithm of PROC NLIN (SAS Institute 1999) which determines parameter
estimates through partial derivations. Temperature data used in the nonlinear model were from
39
18 to 330C. Initial model parameters were calculated by the grid search method (SAS Institute
0 0 1999) with T0 and TL set between 13 to 18 C and 32 to 38 C, respectively.
Weather Data from Florida
Daily minimum and maximum temperatures from Florida were obtained from 98 weather stations recorded by the Applied Climate Information System (CLIMOD, Southeast Regional
Climate Center, http://acis.dnr.sc.gov/Climod/, last accessed February 20, 2007). Daily minimum and maximum temperatures were averaged for the last 5 to 11 years, depending on the availability of data, which provided 365 values for each temperature and station. The maximum period of weather data was from January 1, 1996 to December 31, 2006.
Calculation of Degree-days and Number of Generations for GIS Analysis
Accumulated degree-days for I. variegatus were obtained from DegDay v.1.01 which is a
Microsoft Excel application developed by University of California-Davis
(http://biomet.ucdavis.edu/). This application uses the upper and lower temperature threshold for an organism, and daily average of minimum and maximum temperatures to calculate the accumulated degree-days using the single sine method (Baskerville and Emin 1969). The upper and lower temperature thresholds for I. variegatus immature stages (egg and nymphal stages) were estimated from the Brière-1 nonlinear model as 17.38 and 35.08 0C, respectively. Degree- day requirements for I. variegatus were calculated from the fitted linear regression of the developmental rate function [R(T) = a + bT] as K=1/b (Campbell et al. 1974). The prediction of the number of generations per year was calculated by dividing the cumulative degree-days per station by K, 588.24, required by I. variegatus immature stages to complete development.
Generation of GIS Map for Prediction of I. variegatus Generations in Florida
Weather station name, latitude, longitude and number of I. variegatus generations were tabulated in a Microsoft Excel spreadsheet, saved as IV dBase file and then imported into ArcGis
40
9.0 (ESRI Inc., Redlands, CA, USA). The imported file was converted to shapefile using the
ADD X-Y DATA function followed by the selection of the State Plane Projection. A shapefile of
Florida was obtained from AWhere® Continental USA database (AWHERE, Inc., Denver, CO,
USA) to delineate the range of predictions.
The ArcGis Geostatistical Analyst function (ESRI Inc., Redlands, CA, USA) was used to
generate prediction grids of I. variegatus generations across Florida. Prediction values in
unsampled locations were obtained by surface interpolation of sampled locations. The Inverse
Distance Weighted (IDW) deterministic method was used, whereby predictions are made from
mathematical formulas that generate weighted averages of nearby known values. The IDW
method gives closer points more influence on the predicted value than points that are farther
away (hence the name “inverse distance weighted”). This method was used by Pilkington and
Hoddle (2006) to predict the number of generations of an egg parasitoid in California. The
parameters used in the IDW analysis were
• The number of stations used for interpolation was set to 15 and with a minimum of 10. Due to the large number of weather stations, there were always 15 stations available for interpolation.
• The Power Optimization option was selected generating a Power value p = 1.7021. This means that the weights of each weather station are proportional to the inverse distance raised to the power value p. Therefore, as the distance from the station increases, the weights decrease rapidly.
• The search neighborhood shape was circular because there were no directional influences on the weighting of number of generations per station; thus, equal weight was given to each sample point regardless of the direction from the prediction location (ESRI Inc., Redlands, CA, USA). Ellipse parameters were set to the default values: angle, 0; major and minor semiaxis, 2.3878. Results
Size and Behavior of Stages
Description of I. variegatus eggs and nymphs was reported by Baranowski (1979) and
Slater (1987). The size, color, location and behavior of the different stages found in our studies
41
are as follows: Egg. Length is 2.97 ± 0.13 mm (mean ± SD) (n = 25) (Fig.2-1 a). Eggs are laid in
masses (12 eggs per mass, range 1-38) between the leaf sheath and culm preferentially near the
node. Newly deposited eggs (0 to 5 d) are white and older eggs (6 to 10 d) turn bright red. An
egg parasitoid, Eumicrosoma sp. (Hymenoptera: Scelionidae) was found in Myakka River State
Park and later identified as a possible introduced species for North America (Terry Nuhn 2005
personal communication). This parasitoid attacks young and old eggs (R. Diaz personal
observation) and its presence can be detected by the black coloration of the eggs. Since eggs are
immobile and take longer to develop than other stages, it appears to be the most vulnerable stage
for parasitization or predation. The impact of Eumicrosoma sp. on I. variegatus population is
unknown but previous studies have demonstrated that egg parasitism on hemipterans plays an
important role in regulation of populations (Buschman and Whitcomb 1980, Irvin and Hoddle
2007).
Nymphs
The length of each immature stage is shown in Figure 2-1. Upon hatching, first instars are
bright red in color (Fig.2-1 b) and remain aggregated near the eggs and then migrate to tightly
oppressed spaces between leaves and stems. Feeding and resting occurs in tight spaces between
the leaf sheath and culm, and in the inner whorl. Fourth and fifth instars are darker in color than
early instars (Fig. 2-1 e, f). Laboratory and field observations showed the first to fourth instars
are more often found in aggregations while fifth instars and adults can be observed exploring as individuals. If nymphs or adults are disturbed, they secrete a strong odor from the scent glands
located in the thorax (Fig. 2-1 j) and abdomen. After molting, the cuticles of nymphs and adults
are bright red in color and delicate, but after a couple of hours they darken and harden.
Adults
42
Females are larger than males (Fig. 2-1) and both genders have a distinctive “M” pattern
at the base of the hemelytra. The sclerites at the ventral tip of the abdomen of females are
triangular in shape whereas in males the last sclerites are rounded (Fig. 2-1 h, i). Adults become
highly active during the hottest part of the day and mating individuals can be found in
aggregations. Despite having fully developed wings, adult flying was restricted to short hops of a
few meters or less. Gravid females mostly walked, possibly due the larger size of their
abdomens. Oviposition behavior can be described as follows: females locate the fold of the leaf
sheath by walking around the stem while performing antennation of the surface. Once a fold is
located, the proboscis is extended and briefly inserted at the site for probing; if the site is
accepted then the female extends and inserts the ovipositor at the site and starts laying eggs.
Immature Survival and Developmental Time
The survival of I. variegatus nymphs varied with temperature (Fig. 2-2). Nymphs could
not complete development at extreme low and high temperatures (8, 13 and 380C) and died before molting to the second instar (Fig. 2-2). At 8, 13, and 180C, first instar nymphs typically survived several weeks before dying, while at 380C they usually died a few days after eclosion
(Table 2-1). At 180C, a few nymphs molted to the second instar, but none survived to the third nymphal instar. First instar survival increased between 20.5 to 330C and was the highest, 84%, at
25.50C (Fig. 2-2). Nymphal survival decreased up to the third instar, after which survival
stabilized. The percentage of nymphs that molted to the adult stage was highest, 42%, at 30.50C and lowest, 16%, at 20.50C.
Temperature affected the developmental time for eggs, nymphs and total immature stages
(egg and nymphal stages combined) (Table 2-1). Mean development time from egg to adult was longest, 122 days, at 20.50C and shortest, 40 days, at 30.50C. Egg, first and fifth instar nymphs
had the longest developmental times at each temperature indicating critical stages for I.
43
variegatus survival. Developmental time of each stage significantly decreased between 20.5 to
30.50C (Table 2-1). At 330C, there was a slight increase in developmental time which could indicate the initiation of stressful conditions.
Both linear and nonlinear models were used to determine the relationship between developmental rate (1/D) and temperature (T). Developmental rates of each stage and total immature stages (egg and nymphs) were estimated between 20.5 and 30.50C where the relationship with temperature was approximately linear. Table 2-2 shows the lower threshold temperature and total degree days required to complete development of each immature stage.
The linear model estimated that the lower temperature threshold for all stages ranged from 14 to
16.50C and total degree days required for immature development was 588.
The parameter estimates for the Brière-1 nonlinear model are shown in Table 2-3. The lower temperature threshold predicted for the immature stages ranged from 16.8 to 17.90C (Table
2-3) which may be slightly low since laboratory studies showed that nymphs did not complete
development at 180C. The upper temperature threshold for immatures was predicted to be between 32.5 and 36.10C (Fig. 2-3). The rate of development increased with temperature until the curve reached an optimum and then decreased rapidly as temperatures reached the upper temperature threshold (Fig. 2-3).
GIS Mapping of I. variegatus Generations in Florida
A grid map indicating the predicted number of I. variegatus generations was generated for Florida (Fig. 2-4). Overall, the predicted grids followed a thermal gradient across the state.
Predicted number of generations ranged from 2.36 to 4.84 in Florida counties. Florida counties
located south of Lake Okeechobee had the highest number of generations per year ranging from
3.52 to 4.84. Florida counties located between Orlando and Lake Okeechobee had fewer
generations per year (3.21 to 3.52). Counties where the average maximum temperature in
44
January was below 170C were excluded since laboratory studies and non-linear models predicted high I. variegatus mortality at constant low temperatures.
Discussion
Developmental-time and survival of eggs as well as immature stages were affected by
temperature. No survivorship was observed at extreme low and high temperatures (Table 2-1).
Nymphs died within a few days at 380C and after weeks at lower extreme temperatures,
suggesting that I. variegatus has a broader lower temperature threshold compared to the upper
threshold. A wider range of lower lethal temperature threshold is common for insects (Heinrich
1981, Bayoh and Lindsay 2004). The overall high mortality observed in the first three instars
may have been due to the rearing of I. variegatus as individuals in our experiments, as opposed
to typical aggregations observed in the field. Harrington (1972) observed that Ischnodemus
species were strongly gregarious and nymphs reared in isolation died sooner than nymphs reared
in groups. The benefits of aggregations on survival in early instars is unclear but may be related to an increase in humidity as shown for the cockroach, Blatella germanica (L.) (Dambach and
Goehlen 1999) and the southern green stink bug, Nezara viridula (L.) (Lockwood and Story
1986). Another explanation for the high mortality at extreme temperatures could be constant conditions at which the insects were exposed. Extreme temperatures like 8, 13 and 380C are
typically present for only a few hours a day in the subtropics. During certain hours in winter and
summer, temperatures in I. variegatus infested regions in Florida could reach 0 and 400C, respectively (CLIMOD 2007). Despite these conditions, field sampling confirms that I. variegatus is present throughout the year (R. Diaz unpublished data) demonstrating that this
insect can survive extremes under existing environmental variability. Eggs, first and fifth instar
nymphs took longer to develop than other stages, indicating their importance for I. variegatus
development. The length of time spent during the fifth instar could be explained by the larger
45
amount of food that insects require during the last immature stage before reproduction (Scriber
and Slansky 1981, He at al. 2003, Bommireddy et al. 2004). Longer developmental time in
stages before reproduction was also found in Ischnodemus falicus (Say) and Ischnodemus slossoni Van Duzee (Harrington 1972) which have temperate distributions. Developmental time from egg to adult was three times less at 30.50C (40 d) compared to 20.50C (122 d) demonstrating clearly the influence of temperature on development. The range of temperatures where development was fastest occurred between 28 and 330C (Table 2-1, Fig. 2-3) which is in
agreement with immature survival (Fig. 2-2). These ideal conditions for I. variegatus
development are typical in central Florida from April to October. Developmental rates of I.
variegatus increased almost linearly with temperature until reaching an optimum at 28-300C and then decreasing rapidly (Fig. 2-3). This pattern has been observed in other hemipteran (Scott and
Yeoh 1999), and non-hemipteran insects (Ponsonby and Copland 1996, Mazzei et al. 1999,
Herrera et al. 2005). Our results of developmental rates were obtained at constant temperatures; however, there is a possibility that our values are underestimated since some insects develop faster at variable temperatures (Worner 1992).
Both linear and nonlinear models overestimated the lower temperature threshold since laboratory results indicated that eggs and nymphs did not complete development below 20.50C.
The partial nymphal development at 180C could be an indication that I. variegatus can develop at this temperature for short periods making the prediction of an absolute lower threshold not possible (Herrera et al. 2005). However, since there is a narrow temperature range between 18
(no development) to 20.50C (complete development), we can safely predict that the lower threshold of I. variegatus occurs within this range. The predictions of both models for the lower
threshold for eggs and nymphs were different (Tables 2-2, 2-3). Egg and nymphal lower
46
thresholds ranged from 16.6 to 19.10C and 13.7 to 19.90C, respectively. This indicates a greater
susceptibility of eggs to lower temperatures than nymphs. While nymphs can move and locate
microclimates suitable for development (plant structure, conspecific aggregations), eggs are
immobile and successful development depends on local conditions. This greater resistance to
lower temperatures of nymphs compared to eggs also has been observed in other heteropteran
insects (He et al. 2003, Bommireddy et al. 2004).
The degree-days (588) required to complete development from egg to adult could be
underestimated since the lower threshold is probably higher than 14.70C (Table 2-2). The lower
threshold for development of I. variegatus, 18 to 20.50C, explains its mostly tropical and partially subtropical distribution. The preoviposition period of I. variegatus is about seven days at 280C (R. Diaz unpublished data) and it was not included in the calculations of degree days.
Therefore, the present model probably overestimates the number of I. variegatus generations. A
future model could be improved by including the preoviposition period at different temperatures
to accurately predict the degree-day requirement to complete one generation. Outside of Florida,
I. variegatus has been reported from as far north as Honduras in Central America, as far east as
the Dominican Republic and Trinidad in the West Indies and as far south as northern Argentina
and Uruguay (Slater and Wilcox 1969, Baranowski 1979, Slater 1987, Baranowski and Slater
2005).
The current distribution of H. amplexicaulis and I. variegatus in the continental United
States is limited to central and south Florida (University of Florida, Herbarium 2007; Wunderlin
and Hansen 2008). Further studies on cold tolerance of H. amplexicaulis and I. variegatus would
provide a better understanding of the potential distribution on United States. Prediction of the
potential range and population growth of herbivores could decrease some of the uncertainty
47
about potential ranges of introduced biological control agents. This study estimated the number of I. variegatus generations based on long term data of 98 weather stations across Florida and the degree-days required to complete egg to adult development. The GIS map shows spatially the areas across Florida where I. variegatus could establish and the potential number of generations.
The use of degree-days for mapping insect voltinism has been used recently for insect conservation (nymphalids butterflies, Bryant et al. 2002), pest management (western corn rootworm, Hemerik et al. 2004) and biological control (egg parasitoid, Pilkington and Hoddle
2006). The current northern and southern invasion fronts of H. amplexicaulis are the St. Johns
River (28.080N, 80.750W, Brevard Co.) and Big Cypress National Park (25.920N, 81.30W,
Collier Co.), respectively. The lower winter temperatures in north Florida could be a climatic barrier for the invasion of H. amplexicaulis and I. variegatus which are mostly restricted to the tropics. Degree-day accumulation in central and especially in south Florida, provides ideal conditions to sustain nearly five I. variegatus generations per year (Fig. 2-4) which could facilitate its establishment in case H. amplexicaulis invades the Everglades National Park. If I. variegatus arrives in Australia, the tropical climate of the northern regions would likely provide ideal conditions for its development and population growth. Other countries where climatic conditions for I. variegatus may be ideal include Mexico, Puerto Rico, Venezuela and Cuba, where H. amplexicaulis is highly valued as forage.
Evaluation of herbivores for weed biological control programs includes studies on climate matching between native and adventive ranges, host specificity of the agent and effectiveness in reducing weed density. Temperature-development studies of weed biological control agents provide baseline knowledge that facilitates agent rearing, colonization and prediction of population growth. Temperature experiments revealed that optimal conditions for I.
48
variegatus ranged from 28 to 300C which explains the occurrence of outbreaks late in summer in
Florida (R. Diaz unpublished data) and its subtropical to tropical distribution. Ongoing studies on population dynamics, host range testing and impacts to H. amplexicaulis will elucidate the importance of I. variegatus as a fortuitous biological control agent in Florida.
49
Table 2-1. Mean developmental time in days (mean ± SE) and surviving individuals of immature I. variegatus stages at 10 constant temperatures
Temperature 0 C
Stage1 8 13 18 20.5 23 25.5 28 30.5 33 38
Egg - - - 35.7 ± 0.21 a 15.51± 0.32 b 12.94 ± 0.10 c 10.41 ± 0.20 d 11.43 ± 0.07 e 14.15 ± 0.23 f - 80 80 80 80 80 80 80 80 80 80 N1 2 to 19 3 to 55 38.33 ± 5.75 a 21.79 ± 1.80 b 13.40 ± 1.57 c 11.81 ± 0.56 c 7.52 ± 0.48 d 5.83 ± 0.46 d 8.32 ± 0.37 e 5 to 7 50 50 6 24 20 42 25 30 22 50 N2 29.67 ± 8.17 a 19.13 ± 2.57 b 11.31 ± 1.25 c 10.76 ± 1.24 c 5.80 ± 0.54 c 4.85 ± 0.40 c 5.78 ± 0.48 c 3 16 13 25 20 26 18 N3 12.50 ± 2.00 a 7.83 ± 0.80 b 7.32 ± 0.95 b 5.25 ± 0.46 b 4.96 ± 0.43 b 3.56 ± 0.30 c 10 12 19 20 23 16 50 N4 10.89 ± 0.73 a 7.08 ± 1.11 b 5.53 ± 0.48 b 5.15 ± 0.55 b 4.87 ± 0.37 b 5.00 ± 0.35 b 9 12 17 20 23 14 N5 21.75 ± 1.33 a 10.75 ± 1.11 b 8.59 ± 0.53 c 7.47 ± 0.38 c 8.29 ± 0.40 c 7.82 ± 0.50 c 8 12 17 19 21 11 Only nymphal stage 71.38 ± 6.20 a 47.83 ± 2.40 b 46.12 ± 2.97 b 31.58 ± 1.07 c 28.81 ± 1.01 c 30.09 ± 0.76 c Egg to Adult 121.76 65.88 56.95 41.6 40.23 44.63
1Insect stages: Nymphal first instar (N1), second instar (N2), third instar (N3), fourth instar (N4), fifth instar (N5) Means within a row followed by different letters are significantly different ( P < 0.05; SNK) Analysis of variance of Eggs ( F = 2204.04 ; df = 4, 395; P = 0.0001), N1( F = 46.88; df = 6, 162; P < 0.0001), N2 ( F = 21.37; df = 6, 114; P <0.0001), N3 ( F = 12.17; df = 5, 94; P <0.0001), N4 ( F = 10.91; df = 5, 89; P < 0.0001), N5 ( F = 46.79; df = 5, 82; P < 0.0001), Nymphal stage ( F = 41.95; df = 5, 82; P <0.0001)
Table 2-2. Linear regression parameter estimates describing the relationship between temperature and developmental rates (1/D) of I. variegatus stages
Stage Intercept Slope R2 n Threshold 0C Degree days1 Egg -0.144 0.0087 0.9507 4 16.6 114.94 Nymphs -0.028 0.002 0.9498 5 14.0 500.00 Egg to adult -0.025 0.0017 0.9478 5 14.7 588.24 1 Total degree-day to complete development
51
1 Table 2-3. Parameter estimates (a, To, TL) for the Brière-1 nonlinear model describing the relationship between temperature and developmental rate (1/D) of I. variegatus stages
Parameters estimates (0C)2 Stage 2 A TO 95% confidence TL 95% confidence R Egg 0.00016 17.9 16.74-19.06 32.5 31.12-33.87 0.9900 Nymphs 0.00004 16.8 13.71-19.89 36.1 32.41-39.80 0.9527 Egg to adult 0.00003 17.4 15.84-18.93 35.1 33.70-36.46 0.9820
1 a, empirical constant; T0, Lower temperature threshold; TL, Upper temperature threshold. 2 Degrees centigrade except a.
52
A
B
C
E
F D
J I G H
Figure 2-1. Life stages of I. variegatus and length in mm (mean ± SD): (A) Egg mass on H. amplexicaulis culm, 2.97 ± 0.13 (n=25) ; (B) first instar, 1.45 ± 0.28 (n=23) ; (C) second instar, 2.70 ± 0.39 (n=47); (D) third instar, 3.06 ± 0.31 (n=42); (E) fourth instar, 3.95 ± 0.32 (n=53); (F) fifth instar, 5.45 ± 0.43 (n=46); (G) female, 7.23 ± 0.56 (n=28); male, 6.05 ± 0.22 (n=49), (H) female sclerites at ventral tip of abdomen; (I) male sclerites at ventral tip of abdomen; (J) scent glands in thorax of adult.
53
Figure 2-2. Proportion survival of I. variegatus stages at constant temperatures (0C).
54
Figure 2-3. Developmental rates (1/D) of I. variegatus at different temperatures (0C). Linear regression of egg to adult stages and observed and predicted values by Brière-1 nonlinear model.
55
Figure 2-4. Geographical information system map showing the predicted number of generations of I. variegatus in Florida.
56
CHAPTER III HOST SPECIFICITY OF ISCHNODEMUS VARIEGATUS, AN HERBIVORE OF WEST INDIAN MARSH GRASS (HYMENACHNE AMPLEXICAULIS)
Introduction
West Indian marsh grass, Hymenachne amplexicaulis Rudge (Nees) (Poaceae), is a
perennial emergent weed in wetlands of Florida USA and northeastern Australia. It is considered
native to the neotropics and was introduced into Florida and Australia prior to 1954 and the
1970’s, respectively (University of Florida Herbarium 2007, Csurches et al. 1999). There are
four species of Hymenachne in the neotropics with overlapping distributions in South America
(Catalogue of New World Grasses 2008). This grass is considered a valuable forage in its native
range (Tejos 1978, Enriquez-Quiroz et al. 2006) and its potential as a ponded pasture forage
grass stimulated studies in South Florida in the late 1990’s (Kalmbacher et al. 1998). However,
aggressive growth during long hydroperiods in Florida allowed H. amplexicaulis to outcompete
native vegetation and create large monotypic stands in natural wetlands. Due to the severity of its
impact to native ecosystems, the Florida Exotic Pest Plant Council designated H. amplexicaulis
as a Category I invasive species which are invasive exotics that alter native plant communities by
displacing native species, changing community structure or ecological functions, or hybridizing
with natives (FLEPPC 2007). Similarly, Houston and Duivenvoorden (2002) found that
Australian wetlands invaded by H. amplexicaulis had lower plant species richness, greater relative abundance of exotic fish and altered macroinvertebrate community structure.
The seasonal cycle of H. amplexicaulis in Florida can be described as follows: seeds germinate and shoots begin growing from underwater stolons during the spring. As the water level, day-length and temperature increase, the plant grows aggressively and reaches maximum biomass during late summer (R. Diaz person. observations). Flowering occurs in late fall
(September to November) and is triggered by short days (Tropical Weeds Research Centre
57
2007). Aerial parts of the plant senesce during the winter but stolons and seeds remain dormant
underwater until more favorable conditions appear. While most of local spread is thought to be
through stolons (R. Diaz person. observations), long distance dispersal may occur by either seeds and/or broken stolons (Csurches et al. 1999). Difficult access to infested wetlands limits the ability to contain and control H. amplexicaulis invasions. Where possible, management of H. amplexicaulis can be accomplished through several applications of registered herbicides or changes in wetland hydroperiod and shading (Csurches et al. 1999, Anonymous 2008).
Currently, herbicides constitute the major tool for managing emerged parts of the plant but control of underwater stolons is more difficult. Exploration for pathogens of H. amplexicaulis in
Brazil has not resulted in the identification of promising biological control agents (Soares and
Barreto 2006).
Ischnodemus variegatus (Signoret) (Hemiptera: Blissidae) was found feeding on H. amplexicaulis in Sarasota Co., Florida in 2000 (Halbert 2000). This species is native to South
America (Baranowski 1979, Slater 1987) and the timing and method of the herbivore’s arrival
into Florida are unknown. The insect has been collected in several wetlands in Central and South
Florida and its distribution appears to coincide completely with that of H. amplexicaulis.
Ischnodemus variegatus has five nymphal stages. Females lay eggs in tight spaces within the
leaf sheath and, upon hatching, nymphs form aggregations near the site of eclosion (Diaz et al.
2008a). This sap-sucking bug is commonly found between the leaf sheath and the stem which is
the most suitable place for feeding due to the absence of the silicon dioxide, a feeding deterrent,
in actively growing meristematic areas (Slater 1976). Damage can be recognized by the
accumulation of anthocyanins in the leaves which appear dark red, and later turn brown and die
if the infestation persists. Overholt et al. (2004) found that feeding damage of I. variegatus
58
reduced the photosynthetic capacity and growth rate of H. amplexicaulis under greenhouse
conditions. Temperature dependent studies indicated that the lower threshold for development of
I. variegatus was between 14 to 160C and total degree-days to complete development were 588
(Diaz et al. 2008a). This information allowed the development of prediction models which indicated that I. variegatus could complete from 2.3 to 4.8 generations per year across Florida and its most northern distribution may extend to Leon and Wakulla counties in Northern Florida
(Diaz et al. 2008a).
Host associations of species in the genus Ischnodemus have been described by Slater
(1976), Slater (1987), Slater and Wilcox (1973) and Baranowski (1979). Ischnodemus spp. feed and reproduce on monocotyledonous plants and some species show a remarkable degree of host
specificity, including several marsh grass specialists (Harrington 1972, Slater 1976, Slater and
Wilcox 1973). Table 1-2 shows the host associations of Ischnodemus spp. found in Florida.
Ischnodemus sallei feeds and reproduces on Thalia geniculata in Florida (Slater and
Baranowski1990).
Due to the ecological and economical importance of grasses, we conducted an extensive host specificity investigation of I. variegatus. We hypothesized that because of the longer coevolutionary history of I. variegatus and H. amplexicaulis in their native range, native plants in Florida would be unsuitable hosts for development, and oviposition. The suitability of plant species for development of I. variegatus was determined using first instar nymphs and adults.
We conducted several tests which examined the preference of ovipositing females for H. amplexicaulis and native species. Finally, the presence of I. variegatus in several wetlands in
Florida allowed investigation of the herbivore’s realized host range.
59
Materials and Methods
Origin and Maintenance of Organisms
Laboratory experiments were conducted at the Biological Control Research and
Containment Laboratory (BCRCL), Fort Pierce, Florida. Ischnodemus variegatus and H. amplexicaulis material were collected in Myakka River State Park, Sarasota Co., Florida (27.2o
N, 82.2o W), and Fisheating Creek (26.5o N, 81.7oW). The genetic variability of the insect colony
was maintained by adding field collected individuals at least three times per year. Potted plants
were placed in small, mesh screened cages (0.90 m x 0.90 m x 0.90 m) located within a walk-in
rearing room maintained at 25-30oC, 50-80% RH and a 14:10 L:D photoperiod. Field collected
individuals were released in these cages and monitored every other day for nymphal survival and
colonization. When fresh neonates were needed, adults were held in small cages containing H.
amplexicaulis stems. Eggs were removed every other day and placed in Petri dishes until
eclosion. Adults were handled using an aspirator and nymphs by using a small brush.
Development and oviposition studies were conducted in a walk-in rearing room set at 28oC, 70-
80% RH and 14:10 L:D photoperiod. We used 28oC since a previous study indicated a high survival of individuals at this temperature (Diaz et al. 2008a). Voucher specimens of I. variegatus are maintained in the Florida State Collection of Arthropods (FSCA), Gainesville, FL,
accession # E2002-6139 and the Australian National Insect Collection, Canberra.
Test Plants
Plant species were selected following guidelines of the Technical Advisory Group (TAG)
of the U. S. Department of Agriculture, Animal and Plant Health Inspection Service (USDA
2000). These guidelines follow the centrifugal phylogenetic method developed by Wapshere
(1989) and include species related to the target plant and those which have economical or
ecological importance. Plant selection was biased towards species in the subtribes of Paniceae as
60
well as wetland species that grow in the same habitat as H. amplexicaulis. Plants tested also
included species that are known hosts of other Ischnodemus spp. Additionally, Hymenachne
acutigluma (Steudel) Guilliland, a plant native to Australia, was included to examine the host
specificity of I. variegatus at the generic level. Plants were either collected from the field or purchased at native plant nurseries. Seeds of H. acutigluma were provided by Tim Heard at
CSIRO-Queensland. Test plants were maintained in the greenhouses at the BCRCL in potting soil and received one tablespoon of Osmocote® (18N:7P:12K; Scotts, Marysville, Ohio, USA) after transplanting and weekly applications of Miracle-Grow® water soluble fertilizer
(15N:30P:15K; Scotts, Marysville, Ohio, USA).
No-choice Nymphal Development
The fundamental host range was estimated using neonate nymphs because this insect stage is characterized by poor dispersal capabilities, absence of habituation, sensitization, and
associative learning (Heard 2000). Eggs were collected every other day from the colonies and
placed in Petri dishes until hatching. Five neonate nymphs where transferred using a fine brush
to plastic vials (40dram, 45mm diameter x 102mm tall) containing a piece of stem covered by a
leaf sheath of each test plant species. The end of each vial was covered with a fine mesh for air
circulation. An 8-cm plant piece was placed in moist sand contained at the bottom of the vial
(Fig. 3-1a) and replaced every other day. The number of nymphs tested per plant species was
250, with a few exceptions (see Table 3-2). Nymphal survival and development was monitored
every 48 hrs.
No-choice and Choice Oviposition
Plant species that supported complete development were used in no-choice oviposition conditions. Species included Panicum anceps Michx., Panicum hemitomon Schult, Phanopyrum gymnocarpon (Elliot) Nash, Thalia geniculata L. and H. amplexicaulis as a control. Despite not
61
finding development to adult on P. hemitomon and P. gymnocarpon, these species were included because they occur in the same habitat as the known host in Florida. Fifth instars were collected from the colonies and placed in Petri dishes containing H. amplexicaulis. Newly emerged males and females were placed in a Petri dish containing H. amplexicaulis and allowed to mature sexually for one week. One couple was transferred to a cylindrical cage (5 cm diameter x 20 cm tall) containing a single 20 cm plant stem in wet sand (Fig. 3-1b). Since I. variegatus females actively search for leaf sheaths to lay eggs and hide (Fig. 3-1d) (Diaz et al. 2008a), all cuttings were selected to provide these conditions. Adult survival and number of eggs laid were recorded every two days as cuttings were replaced. At least 15 couples (range: 15 to 25) were tested per plant species. If a male died, it was replaced by a young male from the colony; when a female died, that replicate was terminated.
Plant species used for the choice experiment included P. anceps, P. hemitomon, T. geniculata, Spartina bakeri, Hymenachne acutigluma and H. amplexicaulis as a control. Despite no nymphal development on S. bakeri, H. acutigluma and P. hemitomon, these species were included to provide a wider range of potential hosts for oviposition. Stems (20 cm) of each plant species were placed randomly in a circular fashion in wet sand. A plastic cylindrical cage (30 cm tall x 15 cm diameter) was used to cover the stems (Fig. 3-1c). The effect of adult density on oviposition was tested by placing 1, 5 or 10 couples per cylinder and each density was replicated six times. Adults were collected from the colony and placed in the middle of the arena in a 1- dram vial. Stems were replaced every three days and the number of eggs was counted. If adults died, they were replaced by individuals from the colony to maintain the same density throughout the experiment. After two weeks the experiment was terminated.
62
Field Colonization of Potted Plants
Two sites were established in Sarasota Co. The first site (27.2106 N, 82.2554 W) was
selected in 2005 and consisted of a 100 m long by 3 m wide roadside canal. The approximate
water depth and H. amplexicaulis height were 10 and 80 cm, respectively. One side of the canal
had tall trees which provided partial shade. The H. amplexicaulis stand was considered poor since the plants were short, with small leaves and thin stems. The second site (27.2611 N,
82.2859 W), used in 2006, had a large infestation of H. amplexicaulis in an open marsh in
Myakka River State Park. The approximate water depth and H. amplexicaulis height were 1 and
2 m, respectively. The marsh was flooded periodically during the summer. The H. amplexicaulis stand was considered healthy since the plants were tall, with large leaves and thick stems. The initial population of I. variegatus was estimated by destructive sampling of two contiguous plants at 1 m intervals along a 20 m transect at each site. Then, six 2-liter pots containing uninfested T. geniculata, P. hemitomon, P. anceps, S. bakeri or H. amplexicaulis were randomly placed along the transect and separated by 1 meter. To ensure direct contact, two H. amplexicaulis stems were joined by a cord to each potted plant. After two weeks, the potted plants were recovered from each site and examined in the laboratory for presence of I. variegatus.
Spill-over to Co-occurring Species
Four marshes were selected according to the presence of I. variegatus, H. amplexicaulis and other potential hosts during 2005 and 2006. Three marshes separated by at least 50 meters were located in Myakka River State Park in the summer and fall of 2006. The fourth marsh was located along a roadside ditch in Sarasota Co. in the fall of 2005. The first three marshes were in floodplains located on the east side of the Myakka River and had a depth of >30 cm. Depending of the marsh shape, one or two longitudinal transects were established per site. These marshes in
63
general had low plant species diversity because of the dominance of H. amplexicaulis. Marsh 1 had P. hemitomon, marsh 2 had P. hemitomon and Sacciolepis striata (L.) Nash, and marsh 3 had an unidentified “panicoid” grass. Along each transect, 17 to 20 stations were selected containing a native plant and H. amplexicaulis in its immediate surroundings. At each station, a stem of the native plant and three surrounding stems of H. amplexicaulis were clipped just above the water level and stored separately in plastic bags. Samples of native plants and H. amplexicaulis were dissected in the laboratory and the number of I. variegatus or other Ischnodemus spp. was recorded. Feeding damage on H. amplexicaulis was evaluated by using the top three leaves of each plant. The amount of anthocyanins (red color) per leaf was estimated using a scale (0 = leaf completely green, 1=10% red, 2 = 20% red to 10=100% red). Plant samples were oven dried for one week and dry weight was measured on a Ohaus NavigatorTM balance (precision 0.01g)
(Bradford, Massachusetts, USA).
The fourth wetland (27.2452 N, 82.3524 W) contained H. amplexicaulis and T. geniculata. Twenty plant samples of both species, separated at least by 1 meter, were collected along a longitudinal transect and the presence of Ischnodemus sallei (Signoret) or I. variegatus was assessed in the laboratory. Variables collected from samples included number and stage of insects per stem and presence of egg parasitism (evident from a black coloration of eggs).
Statistical Analysis
The effect of plant species on variables was compared with the general linear model procedure (PROC GLM; SAS Institute 1999). Whenever significant (P < 0.05) F-values were obtained, means were separated using the Student-Neuman-Keuls (SNK) test.
64
Results
No-choice Nymphal Development
A total of 55 plant species, including 34 natives, in 7 families were tested (Table 3-1).
Nymphs molted to the third instar on H. amplexicaulis, Paspalum repens P. J. Bergius, Panicum
anceps, P. hemitomon, Sacciolepis striata, Andropogon virginicus L., Sorghum bicolor Moench,
Eragrostis spectabilis (Pursh) Steud. and Thalia geniculata. Nymphal development to adult was
completed on the known host H. amplexicaulis and the Florida natives P. repens and P. anceps,
both members of the same tribe as H. amplexicaulis; and T. geniculata, a member of the
monocot family Marantaceae (Table 3-1). Survival to adult on the native P. repens (0.4%), P. anceps (2.2%) and T. geniculata (0.3%), was much lower compared to survival on H. amplexicaulis (23.4%), suggesting inferior host quality. There was 100% mortality of first instar
nymphs on the Australian native Hymenachne acutigluma. Nymphs reared on hosts other than H.
amplexicaulis were often observed wandering on the vial walls, and later were found dead on the
sand, which suggests a total rejection of the plant. No development occurred on economically
important grasses except the partial development on S. bicolor.
No-choice and Choice Oviposition
Females laid significantly more eggs (296.5 ± 34.7, mean ± SE) during their lifetime and
survived longer (74.1 ± 9.5 days) on H. amplexicaulis compared with other test species (eggs: F
= 20.7; d.f. = 4, 87; P < 0.0001; longevity: F = 67.4; d.f. = 4, 86; P < 0.0001) (Fig. 3-2). Females
laid an average of 44.9 ± 14.56 eggs and lived 30.6 ± 4.3 days on T. geniculata; no eggs were
found on the other species tested (Fig. 3-2).
Results from the multiple-choice test indicated that females laid significantly more eggs
on H. amplexicaulis compared to other plant species (Fig. 3-3). There was an increase in the
number of eggs laid on suboptimal species with an increase in adult density (Fig. 3-3).
65
Field Colonization of Potted Plants
Initial observations of H. amplexicaulis plants at site 1 indicated the presence of severe
feeding damage by I. variegatus which was reflected by the accumulation of anthocyanins and
necrosis in the leaves. The initial infestation was 13.8 adults and nymphs combined, and 51.8
eggs per stem (Fig. 3-4). Results from potted plants showed that H. amplexicaulis, and to a lesser degree the other species, were colonized by I. variegatus during the 2 week study (Fig. 3-4).
Adults, nymphs and eggs of I. variegatus were found on all test plants but in lower numbers than on H. amplexicaulis (Fig. 3-4).
Hymenachne amplexicaulis plants at site 2 were healthier than those at site 1 and showed less feeding damage despite a high initial density of insects. The initial infestation was 20.0 adults and nymphs, and 23.8 eggs per H. amplexicaulis stem (Fig. 3-5). Lower numbers of adults and nymphs of I. variegatus were found on test species (<1) compared to H. amplexicaulis (48)
(Fig. 3-5). While a high number of eggs (mean per stem = 36) were recorded on H. amplexicaulis, no eggs were found on the other species.
Spill-over to Co-occurring Species
The presence of native grasses and the density of I. variegatus were different among sampled marshes (Table 3-2). There were higher numbers of adults and nymphs on H.
amplexicaulis compared to native plants in all four marshes (Table 3-2). More I. variegatus were
found on the panicoid grass, followed by P. hemitomon and S. striata (Table 3-3). A few
individuals of Ischnodemus brunnipennis were found on P. hemitomon (data not presented).
There was no indication of spill-over of I. variegatus to T. geniculata or I. sallei to H.
amplexicaulis in the site where both species co-occurred. The number of individuals of I. sallei
per stem of T. geniculata was higher than the number of I. variegatus on H. amplexicaulis (Fig.
3-6).
66
Discussion
Host range testing is an essential step for quantifying risks posed by exotic herbivores to native and economically important plant species. The arrival of I. variegatus to Florida wetlands triggered several questions regarding its specificity and potential impacts to the invasive grass H. amplexicaulis as well to native species within the wetland communities and to cultivated grasses.
Therefore, we quantified the fundamental host range of I. variegatus under controlled conditions and, through field experiments, monitored spill-over to suboptimal hosts.
Results of the no-choice developmental study showed that I. variegatus nymphs were highly specific to H. amplexicaulis. Under restrictive conditions, very few nymphs reached the second instar and even fewer reached the adult stage on plants other than H. amplexicaulis, which indicates a narrow fundamental host range. Survival to the third stage was partially related to the taxonomic relatedness of test plants to the invasive weed (Table 3-2). Survival to adult was
23% on H. amplexicaulis while less than 3% survival was found on P. anceps, P. repens and T. geniculata. The relatively low survival of I. variegatus under laboratory conditions suggests that experimental conditions were not ideal. Despite the different soil moisture requirements of
Sorghum bicolor and H. amplexicaulis, the partial nymphal development found on S. bicolor could warrant further attention. Except for T. geniculata, survival to adult was restricted to species in the tribe Paniceae. Most plant species did not support complete development of I. variegatus, despite some survival of early nymphal instars. Further laboratory studies on the suitability of suboptimal hosts could include insect population growth parameters, developmental times and fitness of emerging adults (van Klinken 2000).
Studies under no-choice conditions revealed that I. variegatus females survived longer and laid more eggs on H. amplexicaulis compared to other species. Females lived an average of
67
74 days on H. amplexicaulis, 31 days on T. geniculata and less than 8 days on P. anceps, P.
hemitomon and P. gymnocarpon. Longevity of adults provided only water was similar to
longevity on P. anceps, P. hemitomon, P. gymnocarpon which may suggest that adults did not
feed on these plants. Only 9 out of 18 females laid eggs on T. geniculata while all 18 females
laid eggs on H. amplexicaulis. Oviposition choice experiments indicated that at high densities of adults, females may lay eggs on species that provide suitable oviposition sites (Fig. 3-3).
Females preferred to lay eggs on H. amplexicaulis at 1-couple and 10 couples per cylinder compared to other plants, and there was no preference for a particular location along the stem (R.
Diaz person. observ.). Eggs were found on species in which stems were morphologically similar to H. amplexicaulis, such as H. acutigluma and to a lesser degree, on P. hemitomon. Field observations indicated that some I. variegatus females preferred to oviposit slightly above nodes
(R. Diaz person. observ.) but under high insect densities, this preference was not observed. Lack of unoccupied suitable locations for oviposition on H. amplexicaulis could have forced females to relax discrimination for oviposition sites and move to suboptimal hosts (Schoonhoven et al.
2005).
Field experiments during 2005 and 2006 demonstrated that H. amplexicaulis supported higher densities of I. variegatus compared to other plant species. Spill-over to native plants differed between sites 1 and 2 (Figs. 3-4 and 3-5). The quality of H. amplexicaulis plants at site
1 was inferior to those at site 2, probably due to partial shading and low water level, and this was reflected in greater spill over to native plants. Eggs, nymphs and adults of I. variegatus were found on T. geniculata, P. hemitomon and P. anceps at site 1 but in less numbers than on H. amplexicaulis (Fig.3-4). Despite having a higher density of insects per stem than at site 1, spill
over of I. variegatus to native plants occurred to a lesser degree than at site 2. Hymenachne
68
amplexicaulis plants at site 2 were taller, with a much higher biomass than at site 1, and were
able to support ~20 adults and nymphs per stem with little visible damage (R. Diaz person.
observ.). We suspect that the quality of H. amplexicaulis influences the risk of spill over. Where
quality of the primary host is low, I. variegatus may exhibit a greater tendency to move to
alternative hosts.
Field studies revealed that H. amplexicaulis supported a higher density of I. variegatus
and experienced greater feeding damage compared to co-occurring native species.
Morphological features of the “panicoid” grass, such as the gap size between the leaf sheath and
the stem, as well as the stem diameter, may explain the spill-over observed. Under the local field
conditions, we could not detect spill-over of I. sallei to H. amplexicaulis nor of I. variegatus to T.
geniculata (Fig. 3-6). Phylogenetic relationships among plants could not explain the use of T.
geniculata (Marantaceae), suggesting that other factors such as plant chemistry (Bernays and
Chapman 1994) or phylogenetic relationships among Ischnodemus spp. might be involved in I.
variegatus host selection. Field sampling in St. Lucie and Sarasota counties indicated that I.
sallei is commonly found associated with T. geniculata (R. Diaz person. observ.; Table 3-1).
Future studies could evaluate potential competition, sharing or displacement of I. sallei by the exotic I. variegatus in T. geniculata.
Other Hymenachne spp. and T. geniculata, P. repens and S. bicolor occur in the native
range of H. amplexicaulis, allowing possible interaction with I. variegatus (New York Botanical
Garden 2007, Catalogue of New World Grasses 2008). However, collections in the native range
have only found I. variegatus in association with H. amplexicaulis (Slater and Wilcox 1969,
Baranowski 1979). This suggests that despite the presence of potential alternative hosts, I. variegatus sustains persistent populations only on H. amplexicaulis. The recognition of primary
69
hosts and potential spill-over to secondary hosts in members of the family Blissidae was
recognized by Slater (1976, 1987), who has studied its phylogeny and host associations of this
group for several decades. According to Slater and Wilcox (1973), most blissids tend to use one
host species, but the temporary use of a secondary host has been observed. Furthermore, these
authors suggested that most secondary hosts do not meet all physiological and ecological
requirements of the insects, and that after a short time “colonies on such hosts tend to die out”.
The population dynamics of I. variegatus, coupled with host quality of H. amplexicaulis,
may provide insight into the timing of spill-over. Field monitoring of I. variegatus and H. amplexicaulis in Sarasota Co. indicated that 1) insect populations increase during the summer until reaching a peak in September and October and 2) H. amplexicaulis grows more aggressively in open, deep water marshes with high sediment influx compared to shaded, low water wetlands (R. Diaz unpublished data, Csurches et al. 1999). Therefore, we hypothesize that the risk of spill-over on suboptimal hosts may be higher in shaded areas experiencing high insect densities later in the growing season (fall in Florida). Further periodic sampling of suboptimal hosts growing close to as well as separated (Schooler et al. 2003) from H. amplexicaulis would be required to test this hypothesis. According to other studies, spill-over of weed biological control agents may occur when populations reach certain threshold levels (Dhileepan et al. 2006,
Schooler et al. 2003, Blossey et al. 2001). The presence of I. variegatus on H. amplexicaulis stands in Florida provides an ideal field setting for studies on host switch and permanent utilization of suboptimal species predicted by the fundamental host range.
Changes in host utilization by specific insect herbivores depend on the level of novel selection pressures and the genetic variation upon which the selection will act (van Klinken and
Edwards 2002). In the present system Ischnodemus variegatus had a much higher fitness when
70
feeding on H. amplexicaulis compared to other hosts; and, the fortuitous arrival of I. variegatus
into Florida suggests that the local population could have a low genetic variability. Under this
scenario, a selection pressure for improved performance, and consequently permanent utilization,
of suboptimal hosts (native plants) in Florida may be unlikely. However, selection pressures
favoring utilization of suboptimal hosts could be related to the availability of enemy free-space
(Bernays and Graham 1988), a reduction in primary host density and seasonal availability during
primary host absence. The permanent utilization by I. variegatus of suboptimal hosts in Florida
wetlands should be determined by long term monitoring studies.
Conclusion: Field surveys in the native range indicated that I. variegatus had been collected only from H. amplexicaulis (Baranowski 1979, Baranowski and Slater 2005).
Nevertheless, the presence of I. variegatus in Florida allowed us to study the interactions of a putatively monophagous herbivore with other plant species. Overall, I. variegatus had a narrow fundamental host range. Higher adult survivorship and longevity as well as higher fecundity were found on H. amplexicaulis compared to other species. Field and laboratory tests suggested that H. amplexicaulis plant quality and I. variegatus density affect the degree of spill-over to suboptimal host plants. We conclude that feeding on species other than H. amplexicaulis results in high fitness costs to I. variegatus which may be a strong selection pressure for maintenance of a narrow host range.
71
Table 3-1. No-choice nymphal survival of I. variegatus on test plants1 st Phylogenetic classification Origin2 Habitat/Uses3 1 Instar Percent survival to Initial Survived Second Third Fourth Fifth Adult (n, days) Family Poaceae Tribe Paniceae Subtribe Paspalinae Amphicarpum muhlenbergianum (Schult.) Hitchc. N Wetlands 250 0.0 0 0 0 0 0 Axonopus fissifolius (Raddi) Kuhlm N Wetlands 250 0.0 0 0 0 0 0 Axonopus furcatus (Flüggé) Hitchc. N Wetlands 250 0.0 0 0 0 0 0 Hymenachne amplexicaulis (Rugde) Nees I Wetlands 188 79.8 61.7 42.6 29.8 23.4 23.4 (44, 45.3) Hymenachne acutigluma (Steud.) Guilliland Q Wetlands 250 0.0 0.0 0.0 0.0 0.0 0.0 Paspalum repens P.J.Bergius N Wetlands 258 12.8 9.7 8.1 4.3 0.4 0.4 (1, 40) Paspalum vaginatum Sw. N Salt marsh 14 0.0 0 0 0 0 0 Paspalum floridanum Michx. N Wetlands 250 0.0 0 0 0 0 0 Paspalum blodgettii Chapm. N Rocky pinelands 30 0.0 0 0 0 0 0 Paspalum conjugatum P.J.Bergius N Wetlands 225 0.0 0 0 0 0 0 Paspalum urvillei Steud. I Wetlands 190 0.0 0 0 0 0 0 Paspalum notatum Flüggé I Pastures 250 0.0 0 0 0 0 0
72 Phanopyrum gymnocarpon (Elliott) Nash N Wetlands 250 0.0 0 0 0 0 0 Subtribe Cenchrinae Pennisetum purpureum Schumach. I Wetlands 212 3.8 0.5 0 0 0 0 Pennisetum setaceum (Forssk.) Chiov. I Disturbed sites 300 0.0 0 0 0 0 0 Subtribe Digitariinae Digitaria ciliaris (Retz.) Koeler 250 0.0 0 0 0 0 0 Subtribe Melinidinae Melinis minutiflora P.Beauv. I Dry/Disturbed sites 299 0.0 0 0 0 0 0 Urochloa mutica (Forssk.) T.Q.Nguyen I Wetlands 246 1.6 0 0 0 0 0 Subtribe Panicinae Echinochloa walteri (Pursh) A.Heller N Wetlands 200 0.0 0 0 0 0 0 Panicum amarum Elliott N Salt marsh 110 0.0 0 0 0 0 0 Panicum anceps Michx. N Wetlands 225 6.7 4.0 4.0 4.0 2.2 2.2 (5, 45) Panicum hemitomon Schult. N Wetlands 245 0.0 0 0 0 0 0 Panicum maximum Jacq. N Wetlands 345 3.2 2.0 1.4 0 0 0 Panicum repens L. I Old fields 250 0.4 0 0 0 0 0 Panicum virgatum L. I Wetlands 80 0.0 0 0 0 0 0 Sacciolepis striata (L.) Nash N Salt marsh 250 0.0 0 0 0 0 0 N Wetlands 247 4.5 1.2 1.2 0.8 0 0 Subtribe Setariinae
Table 3-1. Continued 1 st Phylogenetic classification Origin2 Habitat/Uses3 Instar Percent survival to Initial Survived Second Third Fourth Fifth Adult (n, days) Setaria sphacelata (Schumach.) M.B.Moss ex Stapf & C.E.Hubb. I Disturbed sites 250 0.0 0 0 0 0 0 Stenotaphrum secundatum (Walter) Kuntze N Wetlands 250 0.0 0 0 0 0 0
Tribe Andropogoneae Andropogon brachystachyus Chapm. N Wetlands 190 0.0 0 0 0 0 0 Andropogon glomeratus (Walter) Britton et al. N Wetlands 226 0.4 0 0 0 0 0 Andropogon virginicus L. N Wetlands 241 3.3 2.5 1.2 0 0 0 Andropogon virginicus L. var. glaucus Hack N Scrub/Dry pinelands 250 0.0 0 0 0 0 0 Hemarthria altissima (Poir.) Stapf & C.E.Hubb. I Cultivated pasture 241 0.0 0 0 0 0 0 Imperata cylindrica (L.) P.Beauv. I Disturbed sties 250 0.0 0 0 0 0 0 Saccharum officinarum L. I Cultivated 250 0.0 0 0 0 0 0 Sorghum bicolor Moench I Cultivated 36 5.6 2.8 2.8 0 0 0 Tripsacum dactyloides (L.)L. N Wetlands 250 0.0 0 0 0 0 0 Tripsacum floridanum Porter ex Vasey N Rocky pinelands 250 0.0 0 0 0 0 0 73 Zea mays L. I 285 0.0 0 0 0 0 0 Tribe Arundineae Arundo donax L. I Wet to dry sites 250 0.0 0 0 0 0 0 Tribe Cynodonteae Cynodon dactylon (L.) Pers. I Disturbed sites 192 6.3 1.0 0 0 0 0 Muhlenbergia capillaris (Lam.) Trin. N Wetlands 250 0.0 0 0 0 0 0 Tribe Eragrostideae Eragrostis elliottii S.Watson N Disturbed sites 248 0.8 0.4 0 0 0 0 Eragrostis spectabilis (Pursh) Steud. N Wetlands 226 3.5 3.1 0.9 0 0 0 Tribe Oryzeae Oryza sativa L. I Wetlands 807 1.1 0.1 0 0 0 0 Luziola fluitans (Michx.) Terrell & H.Rob. N Wetlands 50 0.0 0 0 0 0 0 Tribe Zoysieae Spartina bakeri Merr. Sporobolus indicus (L.) R.Br. I Dry hammocks 250 0.0 0 0 0 0 0 Zoysia sp. I Disturbed sites 250 0.0 0 0 0 0 0
Family Alismataceae Sagittaria lancifolia L. N Wetlands 250 0.0 0 0 0 0 0
Table 3-1 Continued 1 st Phylogenetic classification Origin2 Habitat/Uses3 Instar Percent survival to Initial Survived Second Third Fourth Fifth Adult (n, days) Cyperus ligularis L. N Disturbed sites 65 3.1 1.5 0 0 0 0 Family Haemodoraceae Lachnanthes caroliana (Lam.) Dandy N Wetlands 90 0.0 0 0 0 0 0 Family Marantaceae Thalia geniculata L. N Wetlands 304 1.0 0.3 0.3 0.3 0.3 0.3 (1, 41) Family Pontederiaceae Pontederia cordata L. N Wetlands 250 0.0 0 0 0 0 0 Family Typhaceae Typha latifolia L. N Wetlands 165 1.2 0 0 0 0 0
1Higher classification based on the Catalogue of new world grasses (Poaceae) (http://mobot.mobot.org/W3T/Search/nwgc.html).
2 N= native to Florida, I= introduced, Q= not present in Florida, 3 Habitat data from Wunderlin and Hansen (2004). 74
Table 3-2. Mean number of I. variegatus per stem and plant parameters collected from spill-over studies in Florida during 2006 Mean I. variegatus Damage scale2 Dry weight 3 Water level Location/plant species Stations per stem1 (0 to 10) (g/tiller) (cm) Marsh 1 H. amplexicaulis 10.43 ± 1.78 4.13 ± 0.32 a 7.36 ± 0.25 a 20 40 P. hemitomon 0.50 ± 0.30 a - - Marsh 24 H. amplexicaulis 1.20 ± 0.18 0.80 ± 0.15 c 7.18 ± 0.29 a 18 30 S. striata 0.11 ± 0.08 a - - H. amplexicaulis 1.55 ± 0.42 0.62 ± 0.13 c - 20 40 P. hemitomon 0.10 ± 0.06 a - Marsh 3 H. amplexicaulis 7.55 ± 1.39 2.90 ± 0.41 b 10.93 ± 0.77 b 17 50 Panicoid 3.7 ± 0.75 b - 17.33 ± 1.31 c 1 Adult and nymphs, means were significantly different when comparing H. amplexicaulis versus native plant (1- way ANOVA, F > 5.9, P < 0.05), different letters indicate comparisons among native plants only; 2Average of top three leaves (F = 39.8, d.f. = 3,74; P < 0.0001 ) ; 3 Above water biomass (F = 40.5, d.f. = 3, 72; P < 0.0001); 4 Marsh 2 contains two transects
75
B C
A D
Figure 3-1. Laboratory set-up for I. variegatus host range tests. A) No-choice development, B) No-choice oviposition, C) Multiple-choice oviposition, and D) Female laying eggs.
76
100 350 a a 300 80 s)
y Longevity a 250 Eggs ) ) (d E E 60 S S 200 ± ( (± s
g y t g i
v l e e 150 g b 40 ta n o o T l
an 100 e
M b 20 b b 50 b b
0 0 ulis lata ceps on pon ater xica nicu . an itom car y w ple . ge P hem mno Onl . am T P. . gy H P
Figure 3-2. Mean (± SE) adult longevity and number of eggs laid under non-choice conditions. Plant species with different letters within a variable are significantly different (P<0.05).
77
90 H. amplexicaulis H. acutigluma 80 P. hemitomon P. anceps m
e 70 T. geniculata S. bakeri a st 60 per )
E 50 S
± a 40
30 a b ean eggs ( a b M 20 b b 10 bb b b b b 0 1-couple 5-couples 10-couples
Figure 3-3. Mean (± SE) number of eggs laid at three different adult densities. Plant species in the same treatment with different letters are significantly different (P<0.05). 1 couple (F = 15.23 ; d.f.= 5, 138; P < 0.0001), 5 couples (F = 9.34, d.f. = 5, 138; P < 0.0001), 10 couples (F= 9.89 ; d.f.= 5, 138 ; P < 0.0001)
78
60
55 m e t Adults and Nymphs 50 Eggs ) per s
SE 45 (±
r 20 a
15 a numbe n a e
M 10 b 5 b b b c c 0 al is ta n s iti ul la o ep In ca cu om nc s- xi i it a li le en m P. au p g e ic m T. . h ex . a P pl H m . a H
Figure 3-4. Open field test results at site 1. Adult and nymphs: (F = 13.52 ; d.f. = 3, 19; P <0.001); eggs (F = 5.28; d.f.= 3,19; P<0.001). Plant species with different letters within a variable are significantly different (P < 0.05).
79
60 a 50
40 a
em Adult and Nymphs st
r 30 Eggs pe )
E 20 ber (± S
1
ean num b M b b b b b b b 0 al is s ri ta n iti al ep ke la o in cu nc a cu om s- xi a . b i it li le P. S en m ua p g he ic m T. . ex . a P pl H m . a H `
Figure 3-5. Open field test results at site 2. Adult and nymphs: (F = 14.0; d.f. = 3, 26; P <0.001); eggs:(F= 8.99, d.f.= 3,26; P <0.001). Plant species with different letters within a variable are significantly different (P < 0.05).
80
Male I. sallei on T. geniculata I. variegatus on H. amplexicualis Female 5th Instar 4th Instar 3rd Instar 2nd Instar 1st Instar Red egg White egg Black egg
10 8 6 4 2 0 0246810
Mean (±SE) number of indivuals per stem
Figure 3-6. Age distribution (mean ± SE) of all the stages of I. variegatus and I. sallei on their hosts found at marsh 4
81
CHAPTER IV LIFE HISTORY PARAMETERS OF ISCHNODEMUS VARIEGATUS (SIGNORET) (HEMITERA: BLISSIDAE) REARED ON TWO CLOSELY RELATED GRASSES
Introduction
Historically, grasses were not considered a major target group in classical weed
biological control programs. Of the 949 releases of exotic agents in Julien and Griffiths’s
Catalogue (1998), for instance, none have been directed against grass species. The paucity of
biological control programs of grass weeds may be related to potential risk of non-target attack
to economically important species in this group (Wapshere 1990, Pemberton 2002).
Additionally, grasses were thought to support few specific insect herbivores due to simple plant architecture, low presence of secondary metabolites and the occurrence of feeding deterrents such as silica (Tscharntke and Greiler 1995, Massey et al. 2006). Despite this, intensive sampling in the native range has yielded several potential biological control agents of major grass weeds including Cogongrass, Imperata cylindrica (L.) (reviewed by Van Loan et al. 2002), Common
Reed, Pragmites australis (Cav.) Trin. ex Steudel (reviewed by Blossey et al. 2002) and smooth cordgrass, Spartina alterniflora Loisel. (Grevstad et al. 2003). However, the only insect subjected to host specificity testing following the centrifugal phylogenetic method (Wapshere
1989) is the planthopper Prokelisia marginata (Van Duzee). Results indicated that P. marginata had a high degree of host specificity towards S. alterniflora when tested on conspecifics and closely related grasses (Grevstad et al. 2003). Thus, the common assumption that grasses lack specialist herbivores may not be entirely true.
The genus Hymenachne belongs to tribe Paniceae and contains seven species worldwide which inhabit mostly tropical wetlands. There are two native species in Asia, one in Australia, and four in America, and several species are considered valuable as forage. All Hymenachne spp. are perennial, decumbent aquatic species that are dioceious and have a C3 type photosynthetic
82
pathway (Watson and Dallwitz 1992). Collections of the four Hymenachne spp. in South
America reveal a sympatric distribution across wetland habitats (Catalogue of New World
Grasses 2008). There is no information about the phylogenetic relationships among species in this genus.
Tropical and subtropical wetlands are sensitive ecosystems that are vulnerable to the invasion of exotic macrophytes and concomitant loss of biodiversity. Hymenachne amplexicaulis
(Rudge) Nees (Poaceae) is a Neotropical, stoloniferous, perennial grass adapted to wetland conditions (Csurches et al. 1999). The grass is a major weed in central and south Florida, as well as in northern Australia. Herbarium specimens confirmed that H. amplexicaulis has been present in Florida since at least 1954 (University of Florida Herbarium 2007). Due to its aggressiveness, the Florida Exotic Pest Plant council designated H. amplexicaulis as a Category I invasive species, which include invasive exotics that alter native plant communities by displacing native species, change community structure or ecological functions, or hybridize with natives (FLEPPC
2007). This grass was imported into Australia from Venezuela in the late 1970s to evaluate its potential as forage for cattle (CSIRO 1973) but its invasiveness was noted in the mid 1990s
(Csurches et al. 1999). Based on its native distribution and climatic adaptations, H. amplexicaulis has the potential to invade all seasonally flooded wetlands in Australia (Charleston 2006), where it poses a serious threat to the sugar industry, water resources, fisheries, and ecotourism (Csurhes et al. 1999). Due to its ecological and economic importance, the Australian government designated this grass a Weed of National Significance (Thorp and Lynch 2000).
Hymenachne amplexicaulis grows aggressively in flooded conditions and produces large
amounts of biomass. Long distance dispersal occurs by seeds and broken stolons, especially
during flooding. It persists in wetlands due to the presence of underwater stolons, which require
83
several applications of chemical herbicides to control (Anonymous 2008). Flowering is triggered
by short days and a single panicle produce ca. 4200 seeds (R. Diaz unpublished data).
Hymenachne amplexicaulis also forms dense monotypic stands 2-3 m tall in deep water wetlands
with high nutrient influx (Csurches et al. 1999) whereas in shallower areas, the grass performs
poorly.
The sap-feeding bug Ischnodemus variegatus (Signoret) (Hemiptera: Blissidae) was
found in Florida in 2000 (Halbert 2000). This insect is native to South America and its pathway
and date of arrival to Florida are unknown. Laboratory and field studies revealed that I. variegatus negatively influenced growth of H. amplexicaulis (Overholt et al. 2004). Temperature dependent studies determined that 588 degree-days were required to complete one generation and the lower threshold to complete development (egg to adult) was estimated to be 15-170C
(Diaz et al. 2008a). Host specificity studies demonstrated that 23% I. variegatus nymphs completed development on H. amplexicaulis, were as only 2% completed development on closely related species. During oviposition choice tests, females laid more eggs on H. amplexicaulis compared to other plant species (Diaz et al. 2008b). In addition, field sampling in
Florida indicated that H. amplexicaulis sustained larger populations of I. variegatus compared to other plant species. However, some plant species growing in close proximity to H. amplexicaulis were at risk of a temporary colonization when I. variegatus populations were high (late summer) and H. amplexicaulis quality was poor (Diaz et al. 2008b). Several species in the genus
Ischnodemus spp. show a remarkable degree of host specificity. Feeding and reproduction are restricted to monocotyledonous plants and some species are marsh grass specialists (Harrington
1972, Slater 1976). Host associations of species in the genus Ischnodemus have been described by Slater (1976), Slater (1987), Slater and Wilcox (1973) and Baranowski (1979). Adults of I.
84
variegatus are poor fliers and gravid females mostly walk to locate oviposition sites. Females
locate the leaf sheath fold by antennation; upon location, the proboscis is extended and briefly
inserted at the site for probing. If the site is suitable, then the female extends and inserts her
ovipositor and starts laying eggs. The nymphal stage of I. variegatus has five instars, which feed and rest in tight spaces between the leaf sheath and culm, and in the inner whorl (Diaz et al.
2008a). Ischnodemus variegatus can be found in aggregations of immatures and adults (R. Diaz person. observ.)
Hymenachne acutigluma (Steudel) Guilliland is native to Australia and shares a similar ecological niche as H. amplexicaulis. Both species are morphologically similar but can be
distinguished by the base of the leaf which clasps the culm in H. amplexicaulis, but not in H.
acutigluma. The use of H. acutigluma as a forage grass in ponded pastures in Australia has been
suggested since 1990. However, in flooded conditions, H. acutigluma has a lower photosynthetic
rate and reduced photosynthetic leaf area compared to H. amplexicaulis (Kibbler and Bannisch
1999b) and this could explain the invasiveness of the latter in Queensland wetlands.
The presence of H. amplexicaulis, the known host of I. variegatus, and a close relative,
H. acutigluma, in Australia stimulated questions about the level of host specificity by I. variegatus. Insects are typically not considered to be host-specific on grasses (Bernays 1985) and therefore not generally considered as candidates for biological control (Julien and Griffiths
1998). We tested this hypothesis by comparing performance of I. variegatus on two congeneric grasses, the native host and a species on which it had never been exposed. We hypothesized that a specialist grass herbivore would incur high fitness costs on a novel congeneric host compared
to its native host. Immature developmental time, adult longevity, ovipositional preference and
85
population growth of I. variegatus were compared on H. amplexicaulis and H. acutigluma under
choice and no-choice conditions.
Materials and Methods
Origin and Maintenance of Organisms
All experiments were conducted in quarantine at the UF/IFAS Biological Control
Research and Containment Laboratory (BCRCL), Fort Pierce, Florida. Ischnodemus variegatus
and H. amplexicaulis samples were collected in Myakka River State Park, Sarasota Co., (27.2o
N, 82.2o W), and Fisheating Creek (26.5o N, 81.7oW), Glades Co., Florida during 2007. Insect
colonies were maintained in greenhouses in screen cages (1.0 x 0.7 x 0.6 m) containing several
pots of H. amplexicaulis. Seedlings or stolons were planted in 2 liter pots and placed in plastic
trays filled with water to simulate permanent flooding conditions. Pots received 16.2 g of
Osmocote® (18N:7P:12K; Scotts, Marysville, Ohio, USA) after transplanting and weekly
applications of Miracle-Grow® water soluble fertilizer (24N:8P:16K; Scotts, Marysville, Ohio,
USA). Voucher specimens of I. variegatus were deposited in the Florida State Collection of
Arthropods (FSCA), Gainesville, FL (accession # E2002-6139) and in the Australian National
Insect Collection, Canberra. Seeds of H. acutigluma were harvested in 2003 at Marrakai Station
(12.78o S, 131.37o E), Northern Territory, Australia by Arthur Cameron. Seeds were sown upon arrival at the BCRCL quarantine and seedlings transplanted to the same growing media and fertilization regime as H. amplexicaulis. Voucher specimens of H. amplexicaulis and H. acutigluma were deposited in the Botany section’s herbarium of the Division of Plant Industry,
Florida Department of Agriculture and Consumer Services, Gainesville, FL.
A series of tests were conducted to determine the relative performance or preference of I. variegatus on the two grass species. Most tests were conducted on cut plant material, but nymphal development was compared on both cuttings and on whole plants.
86
Development of First Instars on Cuttings
Specificity of I. variegatus was estimated using neonate nymphs because this insect stage is characterized by poor dispersal capabilities and absence of habituation (Heard 2000). Eggs were collected every other day from the colonies and placed in Petri dishes until hatching. Five neonate nymphs where transferred using a fine brush to plastic vials (45mm diameter x 102mm height, Fisher Scientific) containing a layer of wet sand in the bottom and a piece of stem covered by a leaf sheath of each plant species. The vial lid was covered with a fine mesh for air circulation. The number of nymphs tested per plant species was 250. Nymphal survival and development was monitored every 48 hrs.
Development of Third Instars on Growing Tips and Cuttings
A factorial design with plant species (H. amplexicaulis or H. acutigluma) and source
(growing tips or cuttings) was used to compare development of third instar I. variegatus. Five
third instars of I. variegatus were collected from the H. amplexicaulis-caged colony and
transferred to each growing tip or cutting, with 10 replicates per treatment. Growing tips (15 cm)
of each grass species were covered with mesh screens. Cuttings were harvested from growing
tips of each plant species, placed in individual containers with wet sand and covered with mesh
screen. Total cutting length was 18 cm with 3 cm buried in sand and 15 cm exposed to the
insects. Growing tips were replaced every 7 days and cuttings twice per week. The following variables were recorded for individuals that reached the adult stage: (1) developmental time
(days), (2) gender, (3) body length (mm) (rostrum to tip of abdomen, inset Fig. 2-1) and (4)
weight (mg). Adults were placed inside a Petri dish with a ruler and photographed using a digital
camera fitted with Automontage® software (Synchroscopy, Frederick, Maryland, USA) and
measured with the image processing software ImageJ (http://rsb.info.nih.gov/ij/). Adults were
weighed using an Ohaus analytical balance (precision: 0.1 mg) (Bradford, Massachusetts, USA).
87
Since developmental time was recorded every 7 days for the growing tips and every 3 days for
the cuttings, we could not calculate the mean developmental time.
Adult Longevity and Oviposition Test in No-choice Conditions
Two trials were conducted to evaluate the effect of previous experience on the longevity and oviposition of adults under no-choice conditions. Third instars were reared on H. amplexicaulis or H. acutigluma and adults obtained were paired and kept in the same grass
species (n=10). Couples of newly-emerged adults (1 or 2 days after molt) of I. variegatus reared
since first instars on H. amplexicaulis were placed on H. amplexicaulis or H. acutigluma (n = 10
pairs). Couples obtained from either treatment were transferred with soft forceps to cuttings of
each plant species. Cuttings were harvested from growing tips and then placed individually in
containers with wet sand. Cuttings were covered with a clear plastic cylindrical cage (5 cm
diameter x 30 cm height). Total cutting length was 28 cm with 3 cm buried in sand and 25 cm
exposed to the insects. Cuttings were replaced every 3 days and survival of adults and number of
eggs was recorded. The experiment was terminated when the female died. If the male died
during the experiment, it was replaced by a new male from the colony. Egg fertility was
evaluated only for adults with previous experience. Weekly samples of eggs laid per female were
collected and placed in a Petri dish with moist filter paper to determine egg fertility.
Oviposition Choice Test
Choice oviposition tests were conducted to simulate field conditions in Australia where
both hosts were growing sympatrically in the presence of the insect. Ten couples of newly-
emerged adults (1 or 2 days after molt) from the H. amplexicaulis colony were exposed to the
two plant species. Cuttings were harvested from growing tips of each plant species and placed
together (separated by 10 cm) inside cylindrical plastic cages (15 cm diameter x 45 cm height)
with wet sand. Adults were released from an open plastic vial (2.5 cm diameter x 6.6 cm tall)
88
placed in the middle of the arena. Stems were replaced every 3 days and the number of eggs laid
was recorded. If the male died, it was replaced by new male from the colony. The experiment
was terminated when the female died.
Population Growth Under No-choice Conditions
Thirteen females and ten males were placed inside cages (n = 4) (90 cm length x 90 cm width x 90 cm height) containing a tray with three potted plants of either H. acutigluma or H.
amplexicaulis. Initial plant height averaged 80 cm and trays were watered daily. Sixty days after
adult inoculation, the plants were harvested and the number of I. variegatus stages was recorded.
The length of fifth instars was measured for all individuals (n=48) found in each H. acutigluma
cage and from a random sample of 112 individuals from the H. amplexicaulis cages. The method
used to measure body length was the same as described above.
Statistical Analysis
Data obtained from the first instar experiment was analyzed using a 1-way ANOVA using plant species as the factor. The third instar experiment was analyzed using 2-way ANOVA
(PROC GLM, SAS Institute 1999) with plant species (H. acutigluma or H. amplexicaulis) and
source (growing tips or cuttings) as factors. Number of eggs laid, longevity of adults and female
fertility under no-choice conditions, and number of individuals per stage in the population
growth experiment were analyzed using one-way ANOVA (PROC GLM, SAS Institute 1999)
with plant species as the factor. Percent hatching (fertility) and nymphal survival were
transformed using the arcsine square root transformation (Zar 1999). The number of ovipositing
females was compared between plant species using a G-test of independence with the Yates
correction for continuity (Zar 1999). Longevity of adults under no-choice conditions and
ovipositional preference in the choice experiment were analyzed using Chi-square goodness of
fit test (Zar 1999).
89
Results
Development of First Instars on Cuttings
Plant species affected survival of I. variegatus nymphs. All neonate nymphs died after a couple of days and none molted to the second instar on H. acutigluma while 80% molted to the second stage on H. amplexicaulis (F = 531.4; d.f. = 1, 95; P < 0.0001) (Table 4-1).
Development of Third Instars on Growing Tips and Cuttings
Plant source (growing tip or cutting) had no effect on adult length or female weight.
Females reared on H. acutigluma were 9% shorter (F = 17.8; d.f = 1, 59; P < 0.0001) and 49% lighter (F = 11.4; d.f = 1, 54; P = 0.0004) than those obtained from H. amplexicaulis (Table 4-1).
There was no difference in male length reared on either plant species (pooled average: 6.7 ± 0.05 mm) (F = 0.17; d.f = 1, 45; P = 0.678) (Table 4-1).
Fewer nymphs reached the adult stage on H. acutigluma than on H. amplexicaulis (Table
4-1; F = 21.9; d.f. = 1, 46; P < 0.0001). There was no difference in survival of H. amplexicaulis
between cuttings and growing tips (Table 1; F = 0.15; d.f. = 1, 21; P = 0.703). However, there
was significantly greater survival to the adult stage on cuttings versus growing tips for H.
acutigluma (Table 4-1; F = 5.82; d.f. = 1, 24; P = 0.024). Overall, survival to the adult stage was
2.5 times lower on H. acutigluma compared to H. amplexicaulis.
The developmental time for most nymphs of I. variegatus to reach the adult stage on H.
acutigluma was longer compared to H. amplexicaulis (Fig. 4-1). This trend was clearly evident
on cuttings where all nymphs completed development by day 25 on H. amplexicaulis, but not until day 35 on H. acutigluma (Fig. 4-1).
Adult Longevity and Oviposition Trial in No-choice Conditions
There was no interaction between plant species and insect experience (third instar vs. adult) on longevity (F = 1.50, d.f. = 1, 38; P = 0.228), days to first oviposition (F = 0.04, d.f. = 1,
90
30; P = 0.837), egg production (F = 2.43, d.f.= 1, 38; P = 0.128) or clutch size (F = 0.00; d.f. =
1, 298; P = 0.999). Thus, further statistical analyses were conducted with pooled data over insect experience. Plant species had a significant effect on adult longevity, days to first oviposition, and egg clutch size (Table 4-1). Adults lived half as long on H. acutigluma compared to H. amplexicaulis (Table 4-1; F = 19.47, d.f. = 1, 38; P < 0.0001). All females reared on H. amplexicaulis oviposited whereas only 62% oviposited on H. acutigluma (χ2 = 6.62; d.f. =1; P <
0.05). Females started ovipositing 66% later on H. acutigluma than on H. amplexicaulis (Table
4-1). In addition, females laid approximately seven times fewer eggs on H. acutigluma than on
H. amplexicaulis (Table 4-1; F = 64.06; d.f. = 1, 38; P < 0.0001; Fig. 4-2). Fertility (eggs hatching ) was not statistically different between plant species (Table 4-1; F = 0.02, d.f. = 1, 84;
P = 0.90).
Oviposition Choice Test
Under choice conditions, females laid significantly fewer eggs on H. acutigluma than on
H. amplexicaulis (Table 4-1; χ2 = 5.8; d.f. = 1; P <0.05).
Population Growth Under No-choice Conditions
Cages with H. acutigluma contained significantly (F > 6.0, df. = 1, 7; P < 0.05) fewer I. variegatus individuals than cages with H. amplexicaulis (Fig. 4-3). After 60 days, fewer first, second, third and fifth instars and males were found on H. acutigluma than on H amplexicaulis
(Fig. 4-3). The length of fifth instars collected from H. acutigluma (5.6 ± 0.1 mm) was significantly less than the length of those collected from H. amplexicaulis (5.8 ± 0.1 mm) (F =
4.6; d.f = 1, 159; P = 0.033).
91
Discussion
Successful colonization and establishment of specialized exotic herbivores in new geographical areas is determined in part by the availability of their primary host. In the case of insects used for classical weed biological control, the occurrence of congeners of the target species may create serious concerns due to higher risk of non target damage (Paynter et al. 2008,
Pemberton 2000a). Local congeneric hosts often share morphological and chemical characteristics that could elicit feeding or oviposition responses in the agent. Host specificity testing performed under quarantine conditions evaluates the performance of the agent on the primary host and its closest relatives. Comparisons of life history parameters of specialized herbivores reared on old association hosts and potential new associations provide baseline information about host suitability, and possibilities for non-target effects (van Klinken 2000).
This study confirmed that H. acutigluma is a poor-quality host for I. variegatus compared to H. amplexicaulis. Several life history parameters of I. variegatus were negatively affected when reared on H. acutigluma. The immature development study revealed that neonate nymphs reared on H. acutigluma were unable to complete their development and did not molt to the second instar. Third instars reared on H. acutigluma exhibited lower survival to the adult stage, a longer developmental time and a reduction in the size and weight of females compared to those reared on H. amplexicaulis. Feeding on growing tips attached to rooted plants resulted in lower nymphal survival on H. acutigluma and a longer total developmental time on both plant species.
It could be possible that plants translocated defensive compounds to the wounded area (Baldwin
1999) resulting in reduced insect performance. The deleterious effects could be cumulative, suggesting that feeding on H. acutigluma from the second instar could result in even greater fitness costs, i.e. increased mortality. Thus, early instars may be more susceptible to inferior host
92
quality compared to later instars. This is why naïve neonates are used in no-choice development
(or starvation) tests during the host specificity testing process.
Adult reproductive fitness parameters of I. variegatus also were low when held on H. acutigluma. The no-choice reproductive performance study indicated that individuals feeding on
H. acutigluma had a shorter lifespan, fewer ovipositing females, longer time to first oviposition and lower egg production. Previous experience (reared from third instar to adult on each species) did not provide any reproductive advantage since performance was similar to that obtained from adults directly transferred from H. amplexicaulis. The longer time to first oviposition might reduce the total number of eggs laid per female in a given lifespan. Under choice conditions, females laid more eggs during their lifespan on H. amplexicaulis than on H. acutigluma. Similar results were obtained in multiple choice oviposition tests where females laid more eggs on H. amplexicaulis compared to H. acutigluma, Thalia geniculata L. and Panicum hemitomon Schult.
(Diaz et al. 2008b). Because H. amplexicaulis may occur in monotypic stands in the native range
(Anten et al. 1998), I. variegatus may not be highly discriminatory in choosing oviposition sites since the chance of encountering a suboptimal host plant species is low. Ischnodemus variegatus females survived on average 2 months and laid eggs on H. amplexicaulis throughout their lifetime, which indicates that females could be synovigenic. Permanent feeding on inferior hosts such as H. acutigluma may depress I. variegatus lifetime fecundity.
Plant components that affect herbivore performance include nitrogen, carbon, trace elements and defensive compounds (Awmack & Leather 2002). Some third instar individuals completed development on H. acutigluma, which indicates that I. variegatus was able to overcome physical barriers to plant penetration, and thus we can speculate that a major component of the poor performance on H. acutigluma involves the nutritional quality of the sap.
93
Other reasons for poor performance of sap-feeding hemipterans include changes in the amino
acid composition in the phloem (Sandström and Pettersson 1994, Cuda et al. 1995, Karley et al.
2002), seasonal changes in host quality (Brodbeck et al. 2004, Bi et al. 2007), lack of feeding stimulants and presence of defensive compounds in grasses (Ciepiela & Sempruch 1999).
The cage experiment demonstrated that the I. variegatus population grew at a slower rate on H. acutigluma compared to H. amplexicaulis over the 60 day period. Despite the low number of adults obtained in the H. amplexicaulis cage, the larger numbers of fifth instars suggests a faster growth of the population in the next generation compared to H. acutigluma. In retrospect, the duration of this experiment could have run longer (at least for 90 days) to assess the difference in populations after multiple generations. A faster growing I. variegatus population on
H. amplexicaulis may deplete host defenses at a faster rate due to a large number of feeding individuals. On the other hand, the slower growth, fewer number and smaller size of I. variegatus on H. acutigluma, indicated this host may not provide suitable conditions for population outbreaks as observed in natural infestations of H. amplexicaulis in Florida (R. Diaz unpublished data). However, some exotic insect herbivores may become adapted to novel hosts in a relatively short period. Carrol and Fox (2007) found that the soapberry bug Jadera haematoloma (Herrich-Schaeffer) (Hemiptera: Rhopalidae) initially had poor performance on a novel host. However, after 30-50 years J. haematoloma, became more efficient in exploiting the novel host. This finding suggests that weed biological control agents that show poor performance on non-target plants during initial host range studies should be monitored even several decades after release.
94
Several species of the genus Hymenachne share the same ecological niche and occur sympatrically in South and Central America as I. variegatus (New York Botanical Garden 2008,
Catalogue of New World Grasses 2008). However, voucher specimens and host associations of I. variegatus do not report its presence on hosts other than H. amplexicaulis (Slater and Wilcox
1969, Baranowski 1979). One might hypothesize that I. variegatus and H. amplexicaulis have a long coevolutionary history in their native range, which suggests that plant chemistry unique to
H. amplexicaulis is vital to fitness as well as adaptation to host defenses. This could explain the poor performance of I. variegatus observed on H. acutigluma.
The present study demonstrates that contrary to conventional wisdom that grasses are unsuitable targets to be considered in biological control programs, specialist arthropods have evolved to maximize fitness on certain grass species. Further studies are necessary to (i) quantify the population growth rates of I. variegatus on both species; (ii) establish the effect of plant quality (especially nutrient status and insect damage) on fitness parameters; (iii) determine the relationship between plant species density and colonization by I. variegatus; and (iv) investigate the behavioral responses of nymphs while feeding on a sub-optimal host. A field and laboratory study on the use of other neotropical species of Hymenachne by I. variegatus would also provide useful information.
95
Table 4-1. Development and reproductive performance parameters of I. variegatus exposed to H. amplexicaulis and H. acutigluma1
H. amplexicaulis H. acutigluma Growing tip Cutting Growing tip Cutting Development of first instars (no-choice) Survival to second instar (%) - 79.8a - 0.0 b
Development of third instars to adults (no-choice) Survival (%) 67.5 ± 8.8a 73.0 ± 8.2a 18.3 ± 6.4b 41.3 ± 8.6c Development time (median, days) 18.6 18.6 31.1 20.5 Length - male (mm) 5.7 ± 0.1a 5.6 ± 0.1a 5.7 ± 0.1a 5.5 ± 0.1a Length - female (mm) 7.1 ± 0.1a 7.2 ± 0.2a 6.5 ± 0.1b 6.5 ± 0.2b Weight - female (mg) 1.6 ± 0.1a 1.7 ± 0.1a 1.3 ± 0.1b 1.0 ± 0.1b
Adult longevity and oviposition (no-choice) Longevity (days) - 65.8 ± 4.4a - 32.3 ± 5.9b Reproductive females (%) - 100.0a - 61.9b a b First Oviposition (days) - 8.5 ± 0.5 - 12.9 ± 1.9 a a Egg hatch (%) - 76.2 ± 4.5 - 73.2 ± 6.2 a b Total eggs per female - 251.4 ± 25.1 - 35.2 ± 12.7
Choice oviposition Total number of eggs laid - 200.1 ± 16.6a - 73.4 ± 6.5b
1 Means within a line followed by the same letter are not significantly different, P > 0.05.
96
100
80 ) (% s t 60
ged adul 40 Emer
20 H. amplexicaulis cutting H. acutigluma cutting H. amplexicaulis growing tip H. acutigluma growing tip 0
15 20 25 30 35 40
Days Figure 4-1. Cumulative number of I. variegatus adults emerging from third instar nymphs over time on H. amplexicaulis and H. acutigluma
97
2500 H. amplexicaulis Adult 3rd Instar 2000 s egg of r
e 1500 b m u
tive n 1000 H. acutigluma la
u Adult rd m 3 Instar u C 500
0 0 20406080100 Days
Figure 4-2. Cumulative number of I. variegatus eggs collected under no-choice conditions
98
1000 22 b a a 20 H. acutigluma b 18 ge ge 800 H. amplexicaulis ca ca 16 r r 14 pe pe l s 600 12 s a dua dual 10 vi vi a 400 b ndi ndi 8 i
l a b a i l a 6 Tot Tot 200 a a b 4 a a 2 a a 0 0 s ar ar ar ar ar Egg inst inst inst inst inst Male Female 1st 2th 3th 4th 5th
Figure 4-3. Number of I. variegatus surviving on H. amplexicaulis and H. acutigluma after 60 days. The scale on the left y axis refers to the immature stages while the scale on the right y axis refers to the adults. Dashed lines indicate the original number of males and females per cage. Differences (P > 0.05) were observed for every stage except females, fourth instars and eggs.
99
CHAPTER V
POPULATION DYNAMICS OF ISCHNODEMUS VARIEGATUS, AN ADVENTIVE HERBIVORE OF WEST INDIAN MARSH GRASS (HYMENACHNE AMPLEXICAULIS) IN FLORIDA
Insect herbivores often are reunited with their host plant in exotic ranges due to
accidental arrival, natural migration or intentional introduction, as in the case of classical
biological control. Exotic invasive weeds lack insect communities present in their indigenous
range (Strong et al. 1984), suggesting that specialist herbivores may enjoy a large and often
concentrated food source with potential enemy and competition free space. This scenario is the
major premise for the success of classical biological control of exotic weeds (Williams 1954,
Keane and Crawley 2002). Monitoring the population dynamics of exotic insect herbivores in the
adventive range is a major step towards understanding their potential ecological impacts (Louda
et al. 2003). This information may provide baseline data to describe damage inflicted to the host plant and non-targets, the seasonality of the populations, presence of local natural enemies, and the influence of biotic and abiotic factors as sources of mortality, among others.
The arrival and colonization of the neotropical bug Ischnodemus variegatus (Signoret)
(Hemiptera: Blissidae) in wetlands infested with the exotic West Indian Marsh Grass
Hymenachne amplexicaulis (Rudge) Nees (Poaceae) in Florida triggered several questions about the host specificity and population dynamics of this sap-feeding herbivore, including the extent to which this insect would exert sufficient pressure on the exotic weed to suppress its growth and perhaps reduce its densities in Florida wetlands.
Hymenachne amplexicaulis is a robust, stoloniferous, semi-aquatic, perennial grass, native to the neotropics. The timing and pathway of its introduction into Florida is unknown but its quality as forage (Enriquez-Quiroz et al. 2006), suggests that the introduction may not have been accidental. The Florida Exotic Pest Plant Council classified the grass as a Category I
100
species, which are invasive exotics that are altering native plant communities by displacing
native species, changing community structures or ecological functions, or hybridizing with
natives (FLEPPC 2007). The aggressive growth of H. amplexicaulis is due to rapid adaptation to
changing water levels (Kibbler and Bahnisch 1999a), high production of stolons and perhaps the
absence of effective natural enemies. Once the grass invades a wetland, it forms monotypic
stands 2-3m high with complete canopy cover. At the end of the growing season, this results in
massive accumulation of biomass. Grass dispersal occurs during summer flooding via seeds or
broken stolons. Negative impacts of the grass in Australia affect the sugarcane industry, water
resources, fisheries and ecotourism (Csurches et al. 1999).
Ischnodemus variegatus was discovered feeding and causing severe damage on H.
amplexicaulis at Myakka River State Park (27.2o N, 82.2o W; Sarasota Co., Florida) in 2000
(Halbert 2000), and has since been observed at all locations where H. amplexicaulis occurs in
Florida. Ischnodemus variegatus is considered a new record for the United States and its timing and method of arrival are unknown. The native distribution of I. variegatus includes Central and
South America (Baranowski 1979, Slater 1987). Feeding effects of I. variegatus diminished carbon dioxide assimilation, growth rate and biomass of H. amplexicaulis (Overholt et al. 2004).
Temperature-dependent studies determined that 588 degree-days were required to complete one generation and the lower threshold to complete development (egg to adult) was estimated to be
15-170C (Diaz et al. 2008a). Host specificity studies revealed that I. variegatus is highly specific
to H. amplexicaulis; however, under localized outbreak conditions spill-over to nearby grasses
does occur (Diaz et al. 2008b). The objective of this study was to better understand the seasonal
dynamics of I. variegatus in the adventive range.
101
Materials and Methods
Study Sites
Wetlands at Myakka River State Park (Sarasota Co.) and floodplains along Fisheating
Creek (Glades Co.) were selected as study sites (Fig. 5-1). The Myakka River flows through 45
square miles of the Myakka River State Park, which is located in Sarasota County in southwest
Florida. The land cover in the upper river basin is dominated by a mosaic of pastures, hardwood
forest, palms, citrus groves and row crops. Rainfall is seasonal with most of the rain falling
between April and October (Kushlan 1990). Heavy rain triggers floods that discharge water and
sediment into the riverine marshes. Vegetation of the marsh community adjacent to the Myakka
River is composed of: (a) emergent plants dominated by Panicum hemitomon Schult. (native),
Urochloa mutica (Forssk.)T.Q. Nguyen and Hymenachne amplexicaulis (exotics), and
Polygonum spp.; (b) free-floating plants, including Salvinia minima Baker, Lemna spp. and in
lesser abundance, Eichhornia crassipes (Mart.) Solms; (c) littoral plants, including Quercus spp,
Sabal minor (Jacq.) Pers. and Sabal palmetto (Walter) Lodd. ex Schult. & Schult.f. The park
also contains flag marshes that are dominated by Pontederia cordata L., Saggitaria spp., Thalia geniculata L. and other species with flag-like leaves. Flag marshes occur where the wet season
water depth is between 0.3 and 1 m and the hydroperiod extends more then 200 days per year
(Kushlan 1990).
Fisheating Creek Wildlife Management Area is located in Glades Co. (Fig. 5-1) and
extends for about 40 miles along the course of the only free-flowing tributary to Lake
Okeechobee. The floodplains sampled were located adjacent to the creek and were mostly
grasslands (Fig. 5-2).
102
Myakka River State Park Marsh Description
Three open water marshes infested with H. amplexicaulis were located in Myakka River
State Park. Marsh 1 (27.24633N, -82.31052W) and marsh 2 (27.26155N, -82.28622W) were located adjacent to Myakka river (Fig. 5-3). These marshes are flooded during the summer and the soil remains moist during the winter. The floral composition of marshes 1 and 2 is 80% H. amplexicaulis growing in monotypic stands and a lower proportion of emergent plants such as P. hemitomon and U. mutica. Water flows through these marshes during the rainy season depositing sediments. Marsh 3 (27.25911N, 82.27621W) is a flag marsh, which was located 1.2 km east of the river (Fig. 5-3) and receives less flooding influence from the river. Water in marsh 3 is more stagnant compared to marshes 1 and 2 (R. Diaz person. observ.). The floral composition of marsh 3 is 50% H. amplexicaulis intermixed with patches of P. cordata, Saggitaria spp. and
Typha spp. (R. Diaz unpublished data). During a small outbreak of I. variegatus in October
2006, two hammocks containing an infestation of H. amplexicaulis also were sampled (Fig. 5-3).
These hammocks were adjacent to the Myakka Lake and contained dense coverage of broad- leaved evergreen plants including Quercus spp., Gordonia lasianthus (L.)J. Ellis, Magnolia virginiana L. and Persea borbonia (L.) Spreng., among others.
Environmental Variables
Gauge height, discharge and precipitation data for Myakka River were obtained from the
USGS National Water Information System, Web Interface (http://waterdata.usgs.gov/).
Temperature data for Myakka River State Park were obtained from the CLIMOD database
(http://acis.dnr.sc.gov/Climod/). Day length in Sarasota Co. was obtained from the U.S. Naval
Observatory Data Services website (http://aa.usno.navy.mil/data/).
103
Myakka River State Park Insect and Plant Sampling Methodology
Twenty stations were located at 10m intervals in a single transect that traversed each
marsh during the winter of 2002. At each station, one plant was randomly selected and plant
height was measured from the bottom of the marsh to the newest open leaf. The plant was
removed from the soil and each live I. variegatus stage was counted from the above water
portion. Ischnodemus variegatus eggs were classified according to development as white (newly
laid), red (mature egg) and black (parasitized). Ischnodemus variegatus samples collected during
fall of 2002 revealed the presence of an entomopathogen. Infested samples were sent to Dr.
Drion Boucias at the Department of Entomology and Nematology in Gainesville for
identification. The presence of I. variegatus per tiller infested with the entomopathogen was recorded during each sampling date. If a nymph or adult of I. variegatus had white hyphae, the tiller was reported as positive for the entomopathogen. Data were collected at least once per month but during the outbreak of 2004 samples were collected more frequently.
Myakka River State Park Plant Damage Methodology
Hymenachne amplexicaulis damage was assessed on the apical three expanded leaves of each tiller sampled. For each leaf, the proportion of green, red or brown color was evaluated on a scale from 0 to 10. A score of 10 indicated that 100% of the leaf had a particular color. A more detailed sampling was conducted during October 2006. The following data were collected from the transects in the open marshes 1, 2, 3 and the two additional transects in hammock 1 and 2: tiller height from the bottom of the marsh to the newest open leaf, tiller weight using the above ground vegetative part of the tiller, panicle length and weight, number of I. variegatus present in the above water portion of the tiller and leaf color score in the trop three leaves.
104
Fisheating Creek Insect and Plant Sampling Methodology
Ten stations were located at 10m intervals in a single transect along the longitudinal
section of each grassland. Data collection was similar to that described for Myakka River State
Park. Due to difficulty in accessing the site, the frequency of sampling was less than in the park.
Between-tiller Distribution of Life Stages
Mean and variances were calculated for adults, nymphs and eggs on each sampling date
in each marsh at Myakka River State Park. These data were fitted to Taylor’s power law, which
2 regresses the log10 variance on the log10 mean (Taylor 1961): Log10 s = b log10 m + log10a, where
m is the mean number of insects per tiller in each marsh on each sampling date, s2 is the variance associated with m, b is the slope of the regression line, and a is the anti-log of the intercept.
Regression lines were first plotted with the Poisson distribution line and visually inspected. Data points at densities less than the density at the intersection of the two lines were eliminated from the analysis to avoid confounding true randomness from pseudo-randomness (Perry and
Woiwood 1992). The slope of the Taylor’s power function is considered a measure of aggregation (Taylor et al. 1988). A slope = 1 suggest randomness; slope <1, uniformity; slope >
1, aggregation (Taylor 1961). Slopes for adults, nymphs and eggs were tested for differences from randomness (slope = 1) using a t test (SAS Institute, 1999). The parameters of Taylor’s power function (a, b) were integrated into Wilson & Room’s (1983) model that describes the relationship between the proportion of plants infested and the mean number of individuals per
b-1 b-1 -m[loge (am )/(am -1)] sampling unit: P(I ) = 1- e , where P (I)= predicted proportion of
plants infested, m = mean number of insects per plant, and a and b are the parameters from the
Taylor power function. This model provides additional insight into the degree of aggregation of
organisms. Lower P (I) values for a given mean represent aggregation. Agreement of model
105
predictions with observed values was analyzed by regressing model predicted values on observed
field values (Wilson and Room 1983; SAS Institute 1999).
Intraspecific Association Between Adults and Nymphs
Intraspecific association between I. variegatus adults and nymphs was determined using
data accumulated over 5 years (2002-2006) from marshes 1 and 2 (Myakka River State Park).
The degree of association was measured following the method described by Southwood (1978).
This method records the presence or absence of I. variegatus per tiller and employs a 2 x 2
contingency table. The table was tested using a χ2 test of independence at a 5% level of
significance to determine whether the hypothesis of independence should be accepted or
rejected. Hulbert’s coefficient of determination (C8) was used to measure the strength of
association and to determine whether the association was positive or negative. C8 values range
from -1 to +1, where: +1 is complete positive association, -1 is a complete negative association
and 0 is no association. Hulbert’s coefficient of determination is defined:
ad - bc Obs χ 2 − Min χ 2 c8 = where a, b, c and d are the cells of the 2 x 2 contingency | ad - bc | Max χ 2 − Min χ 2 table; Obs χ2 is the observed value of χ2; Min χ2 is the value of χ2 when the observed a differs from its expected value â by less than 1.0; Max χ2 is the value of χ2 when a is as large or small as the marginal total of the 2 x 2 table will permit (Southwood 1978).
Prediction map of H. amplexicaulis in Florida
Geographic coordinates of voucher specimens of H. amplexicaulis from the native range were obtained from the New York and Missouri Botanical Gardens. A shapefile was generated using Awhere® software (AWHERE, Inc., Denver, CO, USA) and imported to DIVA-GIS software (Hijmans et al. 2001) for generation of distribution maps. The modeling function was selected in DIVA and a prediction map with the climatic suitability of areas in Florida was
106
generated using maximum average temperature during the warmest month, minimum average temperature during the coldest month and precipitation seasonality.
Statistical Analysis
Even though an established sampling protocol was followed, due to the infrequent sampling at Fisheating Creek, standard parametric and nonparametric analyses were not applied.
Consequently, most data are presented graphically as means and standard errors. Plant parameters and insect numbers collected during October 2006 at Myakka River State Park were compared using one-way ANOVA (PRO GLM, SAS Institute 1999) with location as the dependent variable.
Results
Hymenachne amplexicaulis and Ischnodemus variegatus Population Dynamics at Myakka River State Park and Fisheating Creek
Hymenachne amplexicaulis grew rapidly under warmer conditions and rising water levels present from April to November in Myakka River State Park over the course of this study. Gauge height in the marshes during 2003, 2004 and 2005 indicated the occurrence of flood conditions during several weeks (Fig. 5-4). However, rapid tiller growth maintained apical H. amplexicaulis leaves above water allowing photosynthesis (R. Diaz person. observ.).
A pattern in the I. variegatus population density was detected during this 5 year study at
Myakka River State Park. Numbers of I. variegatus per tiller were high during the fall of 2002 when the sampling started. Once temperatures and water level decreased, the population decreased; in 2003, very few insects were found. During the summer and fall of 2004, the insect population started to increase in June and reached its highest density in October (Fig. 5-5). The insect population decreased during December 2004, and throughout 2005, very few insects were
107
found. In 2006, a small increase of the I. variegatus population was observed in July but total
insect numbers per tiller were lower compared to outbreaks as in fall 2002 and 2004.
Hymenachne amplexicaulis height and number of I. variegatus life stages per tiller were considerably lower at Fisheating Creek compared to Myakka River State Park. The average tiller height during the October 2004 was 74 cm compared to the average for Myakka River of 140 cm. The largest number of I. variegatus recorded per tiller was 8.1 nymphs, which occurred during November 2004 compared to Myakka River State Park of 15.8 nymphs.
Egg Dynamics and Parasitism at Myakka River State Park
The number of eggs per tiller reached a maximum 40 eggs during the outbreak of the summer and fall of 2004 and up to 10 eggs per tiller during 2006 (Fig. 5-6). The highest egg parasitism proportion reached up to 0.30 during 2004 and 2006. Adults of a scelionid wasp were recovered from black eggs. Terry Nuhn from the Systematic Entomology Laboratory (USDA,
ARS) identified this wasp as Eumicrosoma sp., a new record from North America (Terry Nuhn
person. communication; SEL reference number: 0506023). There was no distinct reproductive
diapause or overwintering stage observed for I. variegatus because eggs and immatures were
found throughout the year.
Presence of Beauveria bassiana at Myakka River State Park
The white hyphae covering I. variegatus collected from Myakka River State Park was
identified as Beauveria bassiana (Bals.) Vuillemin (Drion Boucias person. communication). This
ubiquitous entomopathogen was most noticeable during the outbreaks of 2002 and 2004 (Fig. 5-
7). The presence per tiller of B. bassiana during both years occurred in December and January
(Fig. 5-7). Despite its presence, there were few I. variegatus individuals infected with B.
bassiana during 2004 (R. Diaz unpublished data). During 2003, 2005 and 2006, B. bassiana was
not observed.
108
Between-plant Distribution and Intraspecific Association of Adults and Nymphs
The variance to mean relationship was supported by Taylor’s power law for all I.
variegatus stages, as indicated by the high coefficients of determination (r2) (Fig. 5-8). Slope of regressions ranged from 1.24 (adults) to 1.37 (nymphs), and all slopes were significantly > 1 (t- test, P < 0.001). Slopes for adults and nymphs were statistically different (t = 2.61; d.f.= 1, 198;
P = 0.01). The relationship of the mean number of adults, nymphs and eggs per tiller to the
proportion of tillers infested was adequately described by the Wilson & Room (1983) model
(Fig. 5-9). Regressions of the observed proportion of tillers infested on the values predicted from
the Wilson & Room model were all significant (F test, P < 0.0001 in all cases) and r2 values
were 0.87, 0.90 and 0.93 for eggs, nymphs and adults, respectively. Predicted curves for all I.
variegatus stages were to the right of a Poisson distribution line (random distribution, P (I) = 1-
e-m), suggesting an aggregated distribution.
The total number of adults and nymphs found in the samples was 7,089. The cell frequencies were Ad and Ny present (Ad=adult, Ny=nymph) =426; Ad present and Ny absent =
123; Ad absent and Ny present= 266; Ad and Ny absent = 1671. The observed χ2 was 865.33 (P
< 0.001) and Hulbert coefficient (C8) was 0.69. This analysis indicates that there was a strong
positive association between I. variegatus adults and nymphs.
Seasonal Dynamics of Tiller and Water Height and Insect Density at Myakka River State Park During the Outbreak in 2004
Tiller height increased gradually at beginning of the 2004 growing season (February-
March);, rapid growth then was observed during July to September due to rising water levels
(Fig. 5-10). Density per tiller of I. variegatus remained low during January to July; however, a
rapid increase was detected between August and October (Fig. 5-10). Evaluation of tiller
damage, reflected by reduction in green coloration on the top three leaves was, inversely related
109
to the cumulative number of I. variegatus adults and nymphs (Fig. 5-11). Accumulation of anthocyanins on distal portion of the leaves, scored as red, started during the first week of
August (Fig. 5-11). The red color score was positively correlated with number of I. variegatus adults and nymphs per tiller (Fig. 5-12). Following the accumulation of anthocyanins, leaves turned necrotic with a brown color by the first week of November (R. Diaz unpublished data).
Despite the rapid growth of I. variegatus and leaf damage during the 2004 outbreak, H. amplexicaulis was still able to allocate large amounts of biomass to tiller production and below water stolons, which resulted in the production of thousands of seeds per panicle (R. Diaz unpublished data).
Comparison of Hymenachne amplexicaulis Plant Parameters, Damage and Ischnodemus variegatus Density During October 2006
Hymenachne amplexicaulis plants growing in hammocks were shorter and the weight of the panicle lower than plants growing in the open deep water of marshes 1 and 2 (Table 5-1).
Ischnodemus variegatus adults and nymphs density was 13.1 per tiller in the open marsh 2 compared with 5.6 and 7.2 in the hammocks. Ischnodemus variegatus egg density was lower in marsh 3 and highest in marsh 1. Red coloration on leaves was similar in marsh 2, and the two hammocks (Table 5-1).
Discussion
Seasonality of Ischnodemus variegatus Populations
Seasonal fluctuations of I. variegatus in Myakka River State Park during this 5 year study followed a distinctive 2 year cycle. Populations increased during fall of 2002, 2004 and partially during 2006. The hurricane season of 2004 was particularly active with water levels rising rapidly during the summer. Hymenachne amplexicaulis was able to grow rapidly during this period in order to allocate resources for above water leaf production. Kibbler and Bahnisch
110
(1999a) demonstrated that rapid elongation of the stem maintains the leaves above the water
allowing emergent leaves to function at full photosynthetic capacity. Despite the adverse
flooding conditions of 2004, I. variegatus populations remained high during late fall of 2004.
Survival of I. variegatus during this major flooding event may have been facilitated by (a)
migration from lower nodes to the whorl of the plant that was constantly above the water and (b)
adaptation to float on the water surface until a tiller was located. Populations of I. variegatus
remained low during the 2005. Temperature and water levels during spring of 2005 could have
affected the survival of H. amplexicaulis and I. variegatus. Early in the spring of 2005, H.
amplexicaulis shoots and seedlings were relatively small when several heavy rains elevated the
water levels drastically submerging the plants for several days. These conditions could have
decimated populations of I. variegatus in the marshes sampled. The following year populations increased during the middle of summer until reaching a peak in October 2006.
The entomopathogen B. bassiana and the egg parasitoid Eumicrosoma sp. at Myakka
River State Park responded to the increase of I. variegatus populations during 2002 and 2004 by
increased prevalence. There is no published information on the native range, biology or host
range of the Eumicrosoma sp. found in this study. However, other Eumicrosoma spp. are
recognized egg parasitoids of other blissid bugs (McColloch and Yuasa 1914, Coracini and
Samuels 2002, Sodomaya 2007). Field sampling in Florida indicated that Eumicrosoma
beneficum Gahan appeared to regulate populations of the blissid Blissus insularis Barber (Reinert
1978). Ischnodemus variegatus populations were able to increase during the summer and fall of
2004 without major noticeable effects by these natural enemies. Further studies are needed to
identify the factors affecting the epizootic of B. bassiana and establish the effect of I. variegatus
aggregations on the spread of the pathogen and parasitoid.
111
Ischnodemus variegatus did not exhibit reproductive diapause or a distinctive diapause
stage during winter. This could influence the I. variegatus ability to survive colder conditions
experienced in northern latitudes. Many tropical and subtropical insects do not enter a diapause
stage due to favorable climatic conditions throughout the year (Denlinger 1986). Other more
temperate species of the genus Ischnodemus overwinter as adults (Harrington 1972, Wheeler
1996). Based on lower developmental thresholds of I. variegatus, Diaz et al. (2008a) modeled
the potential northern distribution up to Leon County at latitude N 30o. A climate matching model using herbarium specimens of H. amplexicaulis from the native range predicted that the most northern distribution in Florida could be Alachua Co. (Fig. 5-13). Field sampling indicated that the most northern distribution of H. amplexicaulis is Brevard County at latitude N 28.1o (Fig.
5-13) (EDD MapS 2008). Sampling in the most northern areas of H. amplexicaulis and I.
variegatus occurrence could provide insights into their adaptation to colder conditions. Based on
the model prediction and the known occurrence of H. amplexicaulis in Florida, it is unlikely to
pose a threat further north. On the other hand, the model predicted that southern areas in Florida
are more suitable for H. amplexicaulis, which poses a serious threat to complex watershed of the
Everglades National Park.
The 2 year cycle observed at Myakka River State Park is probably more related to
stochastic events disrupting population growth than an intrinsic life history strategy of I.
variegatus. Once H. amplexicaulis invades a riverine marsh, it forms thick perennial clonal
monocultures providing a reliable food source for I. variegatus. Winter survival is probably the
more critical aspect limiting I. variegatus population increase during the following season. The
die-back of H. amplexicaulis starting in late November was followed by a decline in I.
variegatus populations. Therefore, the reduction in host quality and quantity combined with cold
112
temperatures during winter may be part of the normal cycle in Central Florida. Flooding events
during winter and early spring also could leave tillers underwater decimating critical
overwintering insect populations.
Ischnodemus variegatus completed several generations per year in Central Florida during
active years and its population growth was related to the accumulation of degree days during the
growing season. Temperature-dependent studies indicated that the life cycle of I. variegatus was
shortest (40 days) and longest (122 days) at 30.5 and 20.5oC, respectively (Diaz et al. 2008a).
Based on degree days required to complete one generation, an interpolation model predicted that
I. variegatus may have up to 3.8 generations per year in Central Florida (Diaz et al. 2008a). Ideal
temperature conditions for I. variegatus growth, where the monthly low average temperature is
above 20.5oC, occur between May and October in Central Florida. This explains in part the
outbreaks observed during the summer and fall at Myakka River State Park. The eruptive nature
of I. variegatus in certain wetlands in Florida could be explained partially by the absence of the
aforementioned natural enemies. It remains to be seen whether this eruptive nature of I.
variegatus also is observed in the native range. It is noteworthy that I. falicus (Say), a native
North American species, also experiences outbreaks on its host Spartina pectinata Bosc ex Link
(Wheeler 1996). Other species in the family Blissidae that are well known eruptive pests of grasses include Blissus leucopterus (Say), B. insularis, Cavelerius saccharivorus Okajima, and
Dimorphopterus gibbus (F.) (reviewed by Sweet 2000).
Occurrence of Aggregation in Ischnodemus variegatus
The between-plant distribution of I. variegatus was found to be aggregated in all stages analyzed, with the slope of Taylor’ power law being significantly >1 in all cases. Moreover, the
Hulbert’s coefficient of association between adults and nymphs was positive and close to one which indicates that both stages are likely to be found associated. This suggests that there may be
113
large fitness benefits for I. variegatus by living in aggregations. Positive associations may occur when species have overlapping niches or interact in a manner that favors mutual occurrence
(Cole 1949). The presence of scent glands in the thorax and abdomen of many species of
Blissidae (Slater and Wilcox 1969) suggests that aggregation is mediated in part by pheromones
(Aldrich 1998). Harrington (1972) found that I. falicus and I. slossoni Van Duzee were strongly gregarious, and suggested that pheromones may be responsible for this behavior. Aggregations increase humidity as shown for the cockroach, Blatella germanica (L.) (Dambach and Goehlen
1999) and the southern green stink bug, Nezara viridula (L.) (Lockwood and Story 1986) providing adequate conditions for survival. The combination of aposematic coloration of the first through fourth instars, release of a strong odor from scent glands when disturbed, and the aggregative behavior of I. variegatus (Diaz et al. 2008) could provide an effective defense mechanism against predators. Aldrich (1998) suggested that the correlation between aposematic coloration and the tendency to aggregate is the norm for Heteroptera. Despite these adaptations for defense, I. variegatus may not yet have a complex of specialized natural enemies in the
Florida. Monocultures of H. amplexicaulis contained less diversity of arthropods compared with native vegetation in Central Florida (Appendix A). Adult heteropterans also use attractant pheromones to locate mates (Aldrich 1998). Ischnodemus variegatus adults can be found in groups of mating couples in field conditions during the summer (R. Diaz person. observ.). The use of attractant pheromones by either male, female or both must be considered a cue for migration (Aldrich 1998), and could play a role in the colonization of surrounding tillers by I. variegatus adults.
Aggregations of insects also could facilitate the exploitation of their host plant. Phloem feeding homopterans create sinks that are supplied with larger amounts of nutrients (Larson and
114
Whitham 1991, Girousse et al. 2005). Previously infested leaves allowed less salivation and faster phloem ingestion by the aphid Aphis fabae Scop (Prado and Tjallingii 1997). The grass feeding aphids Schizaphis graminum (Rondani) and Diuraphis noxia (Kurdjumov) altered their host plants (barley and wheat) in a manner that they ingest increased concentrations of amino acids (Sandström et al. 2000). It seems plausible that I. variegatus aggregative behavior in older nodes of H. amplexicaulis (Diaz et al. 2008a) could be associated with a nutritional advantage.
Feeding Damage and Plant Performance
The appearance of red coloration followed by browning and necrosis of the H. amplexicaulis leaves was associated with a high density of I. variegatus during the summer and fall of 2004. Marshes with high I. variegatus infestations across Myakka River State Park had a distinctive red coloration that was clearly visible from a distance. Similar sampling in non- outbreak years of 2003 and 2005 revealed the top three leaves remained deep green throughout the summer and fall (R. Diaz unpublished data). Some plants under stress tend to accumulate anthocyanins (Chalker-Scott 1999). Blissids affect grasses by direct destruction of cells by the insertion of the stylets (Anderson et al. 2006), blockage of sieve elements by saliva sheaths and by a direct withdrawal of nutrients (Painter 1928). Field measurements of H. amplexicaulis infested with I. variegatus confirmed that carbon dioxide assimilation was ca. 35% less than of non-infested plants and the rate of assimilation was related to the insect density (Overholt et al.
2004). Johnson and Knapp (1996) concluded that feeding by Ischnodemus falicus (Link) was associated with photosynthetic inhibition on Spartina pectinata.
Hymenachne amplexicaulis performance was affected by location in the wetland. Deep open water marshes with direct influence by the river had the tallest and most vigorous plants.
Density of I. variegatus per tiller was particularly high at marsh 2 compared to both hammocks; however, this did not seem to negatively influence any plant trait evaluated. Shaded hammocks
115
had the shortest plants and were less vigorous compared to open water marshes. It could be
possible the less vigorous plants growing in shaded areas are more susceptible to I. variegatus
feeding damage.
The dynamics of I. variegatus, H. amplexicaulis and water levels during 2004 were particularly informative. The plant was able to grow and start allocating large amounts of energy during the spring and summer relatively free of I. variegatus damage. This asynchrony between
the population growth of the plant and its herbivore could be explained in part by an adaptation
of H. amplexicaulis to colder conditions, which allowed growth earlier in the season than I.
variegatus. It may be possible that populations of H. amplexicaulis in Florida came from colder
areas in South America compared to I. variegatus.
In summary, Ischnodemus variegatus populations decreased late in the year due to
reductions in temperatures and host quality. Flooding events during winter and early spring
constituted a great source of mortality for populations. Natural enemies did not appear to be a
major source of mortality for I. variegatus. Aggregation is major adaptation of I. variegatus to
successfully utilize its host. Due to late occurrence of outbreaks in Central Florida, the impact of
I. variegatus on H. amplexicaulis vegetative and reproductive allocation appears to be minimal
especially in deep open marshes.
116
Table 5-1. Plant parameters, damage and I. variegatus numbers in different wetlands at Myakka River State Park during October 20061
Plant Internode Panicle I. variegatus Top three leaves Adults- Marsh Length Weight Length Weight (g) Length Nymphs Eggs Green Red Brown 19.9 ± 0.7 marsh1 117.4 ± 5.7 b 3.3 ± 0.3 c a 0.8 ± 0.08 a 27.6 ± 0.6 a 1.3 ± 0.59 c 9.2 ± 1.1 a . . . 20.1 ± 0.5 0.08 ± 0.03 marsh2 135.3 ± 4.3 a 11.0 ± 0.7 a a 0.7 ± 0.06 a 26.2 ± 0.7 ab 13.1 ± 2.74 a 8.5 ± 0.5 a 5.4 ± 0.5 a 4.5 ± 0.5 a b 13.5 ± 0.5 marsh3 129.9 ± 4.3 ab 4.0 ± 0.2 c d 0.3 ± 0.05 b 16.8 ± 0.9 c 0.3 ± 0.12 c 2.3 ± 0.4 c . . . 17.9 ± 0.7 Hammock M 99.7 ± 3.8 c 7.3 ± 0.2 b b 0.6 ± 0.12 a 25.6 ± 0.12 ab 5.6 ± 1.2 b 5.3 ± 1.1 b 4.3 ± 0.8 a 4.7 ± 0.8 a 1.0 ± 0.38 a 15.5 ± 0.7 Hammock S 97.4 ± 3.9 c 7.3 ± 0.3 b c 0.4 ± 0.09 b 23.7 ± 2.1 b 7.2 ± 1.55 b 3.9 ± 0.5 b 3.7 ± 0.4 a 5.3 ± 0.4 a 1.1 ± 0.24 a 1 Same letters within a column are not significantly different.
117
A
B
Figure 5-1. Location of study sites in Central Florida. A) Myakka River State Park (Sarasota Co.), B) Fisheating Creek (Glades Co.).
118
Figure 5-2. Study sites at Fisheating Creek (Glades Co.) during 2003 to 2006.
119
Hammock 2
Hammock 1
Marsh 2
Marsh 3
Marsh 1
Figure 5-3. Study sites at Myakka River State Park (Sarasota Co.) during 2002 to 2006.
120
4.5 4.0
) 3.5 (m 3.0 ght
i 2.5 e 2.0 e h 1.5 ug a
G 1.0 0.5 0.0 C)
0 35 e ( 30 ur at 25
mper 20 e
e t 15 g a r
e 10 v
A 5 ___ Maximum, _ _ _ Average...... Minimum
) 14 h ( t 13 h 12 lig y
a 11 D 10
02 02 03 03 03 03 04 04 04 04 05 05 05 05 06 06 06 06 09/ 12/ 03/ 06/ 09/ 12/ 03/ 06/ 09/ 12/ 03/ 06/ 09/ 12/ 03/ 06/ 09/ 12/
Date
Figure 5-4. Environmental variables at Myakka River State Park. Dashed line in gauge height graph represent flooding conditions.
121
16 200 14 Marsh 1 12 150 10 8 100 6 4
50 ) r 2 tille (cm
0 0 ) r e 21 SE p
) 200 ±
E 18 Marsh 2 S 15 ± 150 und (
12 o 100 mph ( 9 ny
6 bove gr d
50 a
n t
a 3
h t
0 0 ig e
10 150 h r
an adul Adults ille e 8 Marsh 3 Nymphs T M Tiller height 100 6
4 50 2
0 0
002 003 003 003 003 004 004 004 004 005 005 005 005 006 006 006 006 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 1/2 10/ 1/ 4/ 7/ 10/ 1/ 4/ 7/ 10/ 1/ 4/ 7/ 10/ 1/ 4/ 7/ 10/
Sampling date
Figure 5-5. Number of I. variegatus adults and nymphs and tiller height at three different marshes at Myakka River State Park (Sarasota Co.)
122
50 0.6 Marsh 1 40 0.5 0.4 30 0.3 20 0.2 10 0.1 ) r E
0 0.0 S
tille 80 0.5 ± r ( e s p Marsh 2 g )
0.4 g
E 60 e S
d e ± 0.3 (
s 40 g sitiz g 0.2 a r e a f p o 20
0.1 f r e o b n
m 0 0.0 tio u 8 0.10 r o n e p g Marsh 3 o r a 0.08 r 6 P e v 0.06 A 4 Eggs Parasitized 0.04 2 0.02
0 0.00
04 04 04 04 04 05 05 05 05 05 05 06 06 06 06 06 06 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ 3 5 7 9 11 1 3 5 7 9 11 1 3 5 7 9 11
Sampling date
Figure 5-6. Number of I. variegatus eggs and proportion of parasitism in three different marshes at Myakka River State Park (Sarasota Co.), 2004-2006
123
0.6 0.5 Marsh 1 0.4 0.3 0.2 0.1
ana 0.0 1.0 bassi . Marsh 2 B
0.8
ith 0.6 w s r 0.4 tille
f 0.2 o n 0.0 tio r
o 0.6 p o r 0.5 Marsh 3 P 0.4 0.3 0.2 0.1 0.0
02 03 03 03 03 04 04 04 04 05 05 05 05 06 06 06 06 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ 11 2 5 8 11 2 5 8 11 2 5 8 11 2 5 8 11
Samping date
Figure 5-7. Proportion of tillers with I. variegatus infested with B. bassiana at Myakka River State Park (Sarasota Co.), 2003-2006
124
3 Adult 2
1
0
-1 y= 0.38 + 1.25x; n = 94; r2 = 0.95
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 Log mean 3 Nymph 2
1 ance 0
-1 y= 0.48 + 1.37x; n = 88; r2=0.96
Log vari -1.0 -0.5 0.0 0.5 1.0 1.5 Log mean 4 Egg 3 2 1 0 -1 y= 0.76+ 1.31x; n=52; r2=0.95 -2 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 Log mean
Figure 5-8. Regressions of the log variance on the log mean for I. variegatus adults, nymphs and eggs.
125
1.0 0.8 Eggs 0.6 0.4 0.2 0.0
s 0 1020304050 r Mean insect per tiller tille
d 1.0
ste 0.8 fe
in 0.6
f Nymphs
o 0.4 n
tio 0.2 r o
p 0.0 o r 024681012 P Mean insect per tiller
1.0 0.8 0.6 Adults 0.4 Observed 0.2 Predicted Poisson 0.0 0246810 Mean insect per tiller
Figure 5-9. Relationship of proportion of tiller infested to the mean number of I. variegatus adults, nymphs and eggs
126
3.5 30 Inflorescence present Sept.-Dec. Tiller 3.0 Water gage 25 Adults and nymphs
2.5 r
) 20 tille r e (m p 2.0 Vegetative growth ls ght
March- October Senescence a 15 u d hei Dec.-Feb. i e v
1.5 i d abl i
10 in n Var a
1.0 e M 5 0.5
0.0 0
03 03 04 04 04 04 04 04 04 04 04 04 04 04 05 05 05 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ /1/ 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3
Sampling date
Figure 5-10. Seasonal dynamics of tiller and gauge height and I. variegatus density during 2004 at Myakka River State Park (Sarasota Co.)
127
80 10 Green score = -1.6x + 9.86 70 r2 = 0.9819
phs 8 60
6
and nym 50 t
l Insects Green color lor score co adu 40 Red color 4 of Leaf er
b 30 2 Num 20 Insects = 15.15x - 0.41 r2 = 0.984 10 0 8/6/2004 8/28/2004 9/17/2004 10/8/2004 11/5/2004
Sampling date
Figure 5-11. Regression of accumulative number of adults and nymphs and leaf color score during 2004 at Myakka River State Park (Sarasota Co.)
128
10 y = 0.0624x + 1.6197 r2 = 0.14 8 e
r scor 6 o d col e r
4 Leaf 2
0 0 20406080 Density of adults and nymphs per tiller
Figure 5-12. Regression of number of adults and nymphs with leaf red score during Summer and Fall of 2004 at Myakka River State Park (Sarasota Co.)
129
Local infestations
Figure 5-13. Model prediction of climate suitability for Hymenachne amplexicaulis using herbarium specimens from New York and Missouri Botanical Garden
130
CHAPTER VI DAMAGE OF ISCHNODEMUS VARIEGATUS (HEMIPTERA: BLISSIDAE) TO HYMENACHNE AMPLEXICAULIS UNDER THE INFLUENCE OF VARIOUS FERTILIZER AND SOIL MOISTURE LEVELS
Introduction
Ischnodemus variegatus (Signoret) was found in Myakka River State Park in 2000 and
has since been observed at most locations where Hymenachne amplexicaulis (Rudge) Nees
(Poaceae) occurs in Florida. Ischnodemus variegatus is a new record for the United States and its timing and method of arrival are unknown. The native range of I. variegatus includes Central and South America (Slater 1987). Feeding damage of I. variegatus diminished carbon dioxide assimilation, growth rate and biomass of H. amplexicaulis (Overholt et al. 2004). Temperature- dependent studies determined that 588 degree-days were required to complete one generation and the lower threshold to complete development (egg to adult) was estimated to be 15-170C
(Diaz et al. 2008a). This insect is the dominant insect herbivore exploiting the sap fluids of H. amplexicaulis in Florida (Appendix A). Most of the studies on host injury by blissids have been conducted on chinch bugs, Blissus spp., which are major pests of grasses in USA (Painter 1928,
Miles 1968, Sweet 2000). Painter (1928) confirms that the primary food supply for Blissus spp.
is the phloem. Blissids affect grasses by direct destruction of cells by stylets (Anderson et al.
2006), blockage of sieve elements by saliva sheaths and by a direct withdrawal of nutrients
(Painter 1928).
Laboratory and field observations in Florida during 2004 indicated that Hymenachne
amplexicaulis infested with I. variegatus exhibited an accumulation of anthocyanins (red
coloration) in the leaves (Chapter 5), which is a common reaction of certain grasses to stress
(Costa-Arbulú et al. 2001). Because I. variegatus forms large aggregations inside protected
structures of the plant (leaf sheath, whorl), the increase in anthocyanins and later browning of the
131
leaves could be explained as a cumulative systemic response by H. amplexicaulis to phloem
depletion and vessel stoppage. However, the effect of insect damage on plant performance has
not been evaluated.
Factors influencing the performance of invasive aquatic plants in the adventive range
include biotic and abiotic factors such as hydroperiod, nutrient availability and herbivory levels
(Myers and Bazely 2003). Wheeler and Center (2001) found the nitrogen content of a floating
weed was different among Florida lakes and this affected the dynamics of its biological control
agent. Hymenachne amplexicaulis in Florida occurs in freshwater habitats with different
environmental conditions. Plants growing in riverine wetlands are more robust than plants found
in depressional marshes probably due to water quality and sediment influx (R. Diaz person.
observ.). In addition, several wetlands infested with H. amplexicaulis are located in the vicinity
of agricultural lands in Central Florida suggesting that fertilizer runoff could be present in the
water. Nutrient supply is an important factor that determines the structure of plant communities
(Grime 2001). Success of exotic plants is correlated with the nutrient conditions of terrestrial and
aquatic ecosystems. In the case of aquatic habitats, higher nutrient conditions of lakes were suggested as a cause for the invasion of Hydrilla verticillata (L. f.) Royle and other exotic plants in United States (Andres and Bennet 1975, Dye 1995). The objective of this study was to evaluate the effect of I. variegatus density, soil moisture and nutrient levels on the performance of H. amplexicaulis seedlings.
Materials and Methods
Insect and Plant Cultures
All I. variegatus and H. amplexicaulis samples were collected in Myakka River State
Park, Sarasota Co., Florida (27.2o N, 82.2o W). Plants were maintained in a greenhouse at the
UF/IFAS Biological Control Research and Containment Laboratory (BCRCL), Fort Pierce,
132
Florida. The I. variegatus colony was maintained in cages containing H. amplexicaulis potted
plants placed inside climate controlled rooms (25-30oC, 50-80% RH and 14:10 light/dark
photoperiod). All experiments were conducted in a greenhouse during the summer (May to
October) of 2004, 2005 and 2006.
Effect of Fertilizer and I. variegatus on H. amplexicaulis Performance
Hymenachne amplexicaulis seedlings were allowed to grow up to ca. 20 cm tall in pots
(18 cm height, 17 cm diameter) and placed individually in trays filled with water. Six fourth instars of I. variegatus were placed on each potted plant and covered with a clear acrylic cylinder
(45 cm height, 15 cm diameter) with six holes (6 cm diameter) and tops covered by a fine mesh
to allow air circulation. Insects were replaced when necessary to maintain a constant density on
all plants (6 insects per plant). Two nutrient levels were established (ten replicates per
treatment): 1) 100 ml of liquid fertilizer every 4 days (850 mg;15N-30P-15K; Miracle-Gro,
Scotts Miracle-Gro Products Inc., Marysville, OH), and 2) 100 ml of tap water every 4 days. The
following plant parameters were measured at the 4 day intervals: 1) plant height, 2) chlorophyll
content, 3) leaf damage, and 4) stem thickness. Plant height was measured from the soil to the
base of the most distal open leaf. Chlorophyll content (SPAD units) was measured using a
chlorophyll meter (Minolta SPAD-502) in the middle of the top three open leaves. The
chlorophyll meter measures the absorbances of the leaf in red and near infrared regions and using
these measures the meter calculates a numerical SPAD value which is proportional to the
chlorophyll content of the leaf (Konica-Minolta 2008). Leaf damage was measured as a score
from 0 to 10 where 0 meant no red (anthocyanins) or brown (necrotic) spots on the leaves and 10
meant 100% of the leaves area was red or brown. Stem thickness was measured 3 centimeters
above the ground with a caliper. After 30 days, the cylinders were removed and plants were cut
133
at the base. Plant material was placed inside a drying oven at 60oC for 2 weeks, and aboveground biomass (dry weight) was recorded.
Effect of Soil Moisture and I. variegatus Density on H. amplexicaulis Performance
Hymenachne amplexicaulis seedlings were allowed to grow up to ca. 15 cm tall in pots (8 cm height) and then assigned randomly to individual treatments. Plants were covered with cylindrical transparent cages (35 cm height, 5 cm diameter) with a rectangular window (10 cm length, 2 cm wide) and the tops covered with fine mesh for air circulation (Fig. 6-1). A two-way factorial design was used with different levels of insect density and soil moisture content (8 replicates per treatment). Three levels of insect density were evaluated: 1) control (no insects), 2) medium (2 fourth instars), and 3) high (4 fourth instars). Soil moisture content consisted of three levels: 1) dry (no addition of water), 2) saturated (water level maintained at 4 cm from the base of pots, and 3) flooded (water level maintained 1 cm below the top of the pot) (Fig. 4). Water level was maintained by placing the pots inside a larger container as shown in Fig. 6-1. Despite no addition of water in the dry treatment, soil remained moist during several weeks due to initial conditions and presence of misters in the greenhouse. Cages were monitored twice per week and missing insects were replaced when necessary by fifth instars. After 14 days, the cylinders were removed and plants were cut at the base. Plant material was placed inside a drying oven at 60oC for two weeks, and aboveground biomass was recorded (dry weight).
The same experiment was repeated a second time using higher insect densities and lower initial moisture in the dry treatment. Insect density treatments were: 1) control (no insects), 2) medium (4 fourth instars), and 3) high (6 fourth instars). All plants were assigned to each soil moisture treatment 5 days prior to the insect inoculation, which resulted in lower moisture content in the dry treatment. After 16 days, the cylinders were removed and plants were cut at
134
the base. Plant material was dried at 60oC for 2weeks, and aboveground biomass was recorded
(dry weight). The same plant parameters described above also were recorded.
Effect of Soil Moisture, I. variegatus Density and Fertilizer on H. amplexicaulis Performance
A three way factorial design experiment was conducted using soil moisture level, insect density and fertilizer concentration as treatment variables (8 replicates per treatment). The soil moisture levels were: 1) low (50 ml of water every 4 days), 2) saturated (water level maintained at 4 cm from the base of pots) and 3) flooded (water level maintained 1 cm below the top of the pot). Ischnodemus variegatus density was: 1) control (no insects), 2) medium (3 fourth instars per plants), and 3) high (6 fourth instars per plant). The following three fertilizer levels also were established: 1) 50 ml of liquid fertilizer (100 mg/50ml) (15N-30P-15K; Miracle-Gro, Scotts
Miracle-Gro Products Inc., Marysville, OH), 2) 50 ml of liquid fertilizer (30 mg/50ml), and 3) control (50 ml of water every 4 days). Water level was the block (large container) (Fig. 6-1) and the different bug densities (small pots) and fertilizer levels were distributed randomly in the block. Insect density was monitored once per week and missing individuals were replaced by fifth instars. Variables measured at harvest (21 days after insect inoculation) included plant height, dry biomass, and total leaf area. Leaf area was assessed by taking digital pictures of all leaves and their area was measured using ImageJ ® software. (http://rsb.info.nih.gov/ij/).
Statistical Analysis
Variables measured in the fertilizer experiment were compared between nutrient treatments by repeated measures analysis of variance (Zar 1999). The two-way and three-way factorial experiments were analyzed using ANOVA (PROC GLM procedure, SAS Institute
1999). Significant differences between means were indicated by Student-Newman-Keuls test (P
< 0.05).
135
Results
Effect of Fertilizer and I. variegatus on H. amplexicaulis Performance
Not surprisingly, Hymenachne amplexicaulis responded favorably to the addition of liquid fertilizer. Fertilized plants had higher values of chlorophyll content, lower damage scores and were taller than control plants (Fig. 6-2, Table 6-1). Plant biomass was 3.4 times higher in fertilized plants compared to control plants (Table 6-1). Leaves in the fertilizer treatment were greener and had less accumulation of anthocyanins compared to control plants (Fig. 6-3).
Effect of Soil Moisture and I. variegatus Density on H. amplexicaulis Performance
No interaction was detected between water level and insect density for all variables analyzed (F ≤ 2.25, P ≥ 0.075) in the first and second experiment. Therefore, the type and magnitude of the impact of each factor was similar regardless of the level of the other factor.
High insect density negatively affected the performance of Hymenachne amplexicaulis (Fig. 6-4,
6-5; Table 6-2, 6-3). Plant biomass was reduced by ca. 35% and 23% in the high density treatment compared to medium and control treatments in the first and second experiment, respectively (Table 6-2 and 6-3). Chlorophyll content, expressed as SPAD units, was lower in the high compared to the medium and control treatments. Damage scores increased gradually following insect inoculation in the high and low insect densities (Fig. 6-4, 6-5). Damage scores were higher in the second factorial experiment compared to the first factorial experiment, which was a function of the number of insects used at each density.
No differences were detected among plant parameters in the water level treatment except for plant biomass between dry and flooding treatments in the first factorial experiment (Table 6-
2). In the second factorial experiment, water level had a greater impact on the performance of H. amplexicaulis. Plants in the dry treatment performed poorly compared to plants in the saturation and flooding treatments. Plant biomass and plant height were higher in the flooding compared to
136
the dry treatments (Table 6-3). Chlorophyll content was lower in the dry compared to the
saturation and flooding treatments.
Effect of Soil Moisture, I. variegatus Density and Fertilizer on H. amplexicaulis Performance
No interaction was detected between water level, fertilizer and insect density for plant
weight and height (Table 6-4). Plant weight was 36% and 25% less in the low soil moisture
versus saturated and flooded, and control versus high and medium fertilizer, respectively (Fig. 6-
6a). There was no difference in plant weight among insect levels (Fig. 6-6a). Plant height was
reduced by 26% and 12% in the low soil moisture compared to saturated and flooded treatments,
and control compared to high and medium fertilizer, respectively (Fig. 6-6 b). Plant height was
reduced by 11 % in the medium and high insect densities compared to control treatment (Fig. 6-6
b).
Differences were observed in leaf damage between soil moisture content, fertilizer level
and insect density (Table 6-4, Fig. 6-6 c, d). There also were interactions between fertilizer and
water level as well as insect density and fertilizer level (Table 6-4, Fig. 6-6 c, d). Plants in
saturated or flooded treatments and high fertilizer had greater damage scores compared to low
moisture treatments (Fig. 6-6 c). Plants that received insects and no fertilizer had more leaf
damage compared to plants at all fertilizer levels (Fig. 6-6 d).
Discussion
This study showed that plants exposed to high nutrient levels performed better than
control plants (no fertilizer) despite the feeding pressure of I. variegatus. Fertilizer runoff from
surrounding agricultural areas and human induced eutrophication of lakes are major factors that
determine the aggressiveness of aquatic weeds (Dye 1995). Stands of H. amplexicaulis in riverine wetlands could be considered areas of high productivity due to sediment deposition,
137
higher levels of oxygen in the water and potential fertilizer runoffs. Under these favorable
conditions, it is possible for H. amplexicaulis to withstand severe damage by I. variegatus infestations without major impact on the plant or faster recovery may occur after a temporary infestation. Moreover, H. amplexicaulis responded favorably to the addition of liquid fertilizer by allocating three times more resources to aboveground biomass compared to the control plants.
Changes in life history parameters of I. variegatus feeding on plants exposed to different fertilization regimes were not evaluated in this study. It is possible that feeding on healthier host plants could provide a fitness advantages to I. variegatus as has been demonstrated with other herbivores of aquatic weeds (Wheeler and Halpern 1999).
Ischnodemus variegatus density and water level had clear impact on H. amplexicaulis overall performance. Different plant parameters (e.g. biomass, height, chlorophyll content) decreased with increasing insect densities. Plants with high densities of nymphs had brown necrotic spots on the leaf blade in the first week of the experiment suggesting a rapid breakdown of leaf pigments. Behavioral observations demonstrated that I. variegatus forms aggregations between the leaf sheath and the stem where feeding scars can be found (Diaz et al. 2008a). Leaf damage appears initially in the most distal part of the leaf blade and then progresses until covering the entire leaf area. Observations by Anderson et al. (2006) indicate that chinch bug stylets injured bundle sheet cells in sorghum that contained significant amounts of starch and carbohydrates. Some plants under stress tend to accumulate anthocyanins (Chalker-Scott 1999).
Accumulation of anthocyanins on H. amplexicaulis leaves is similar to damage symptoms on corn (Zea mays L.) and buffalograss (Buchloë dactyloides (Nutall)) caused by feeding of Blissus spp. (Hemiptera: Blissidae) (Negron and Riley 1990; Baxendale et al. 1999). Overholt et al.
(2004) found that I. variegatus feeding damage caused a reduction in the biomass of H.
138
amplexicaulis under greenhouse conditions. The present study provides a better understanding of
the interactions between insect damage and plant performance in relation to plant quality found
in the field.
Higher performance of H. amplexicaulis was obtained under flooding conditions,
demonstrating plant adaptability to wetland conditions (Kalmbacher et al. 1998, Kibbler and
Bahnisch 1999a). In contrast, low moisture conditions (dry treatment) had a negative effect on
the plant, which is in agreement with field observations that showed low tolerance to drought of
H. amplexicaulis (Medina and Motta 1990). In the Brazilian pantanal, H. amplexicaulis occurs within four habitats: marsh ponds, waterlogged basins, tall grasslands and forest edges (Pinder and Ross 1998). Observations in marshes at Myakka River State Park, Sarasota Co., Florida
(27.20 N, 82.20 W) suggest that when subjected to inundation, H. amplexicaulis is capable of rapid stem elongation, increase in foliage volume and rapid nodal adventitious root production
(R. Diaz, pers. observ.). Kibbler and Bahnisch (1999a) demonstrated that rapid elongation of the stem maintains the leaves above the water allowing emergent leaves to function at full photosynthetic capacity. In Venezuela, Tejos (1978) found a positive relationship between H. amplexicaulis growth and depth of flooding and that biomass production ranged from 5,911 -
18,162 t/ha/yr during the flood period and from 5,553 - 7,836 t/ha/yr during the dry season.
In this study, plants growing under dry soil moisture conditions with no insects exhibited
brown spots and reduced accumulation of anthocyanins in the leaves blades compared to plants
with insects. Although the performance of I. variegatus was not evaluated in this study, feeding on water-stressed plants can negatively affect sap-feeding herbivores despite the increase in leaf nitrogen concentration (Huberty and Denno 2004). Hymenachne amplexicaulis may experience
139
periods of drought during the late fall and spring in Central Florida ,which may explain in part the absence of I. variegatus during those months (Chapter V).
When all three factors (water level, fertilizer, and insect density) were incorporated into the three-way factorial experiment, insect density had no impact on plant biomass. This result was quite surprising because the two-way factorial experiments demonstrated otherwise.
However, insect density had a negative effect on plant height suggesting that infested plants would not respond as quickly to rising water levels in the field. Hymenachne amplexicaulis suffered lower damage by I. variegatus in the medium and high nutrient levels compared to the control treatments. Thus, plants under stress due to poor nutritional conditions exhibit greater feeding damage compared to those that received fertilizer. Overall, the results obtained in this study suggest that plants growing in riverine wetlands with greater influx of nutrients and higher water levels might exhibit elevated tolerance to I. variegatus damage.
140
Table 6-1. Hymenachne amplexicaulis variables measured 30 days after the inoculation of I. variegatus
Variable Fertilizer Control F, P Biomass (g) 11.08 ± 0.6 3.22 ± 0.2 177.21, <.0001 SPAD reading 36.83 ± 1.8 20.07 ± 0.9 65.25, <.0001 Damage score 1.0 ± 0.0 4.70 ± 0.33 121.00, <.0001 Height (cm) 68.17 ± 3.8 50.57 ± 1.6 18.20, 0.0016 Stem thickness (mm) 6.25 ± 0.2 5.83 ± 0.3 1.37, 0.2682
141
Table 6-2. Effect of Ischnodemus variegatus density and soil moisture on different Hymenachne amplexicaulis parameters after 14 days1 Insect density Soil Moisture Variable Control Medium (two) High (four) Dry Saturation Flood Biomass (g) 0.41 ± 0.04 a 0.40 ± 0.04 a 0.26 ± 0.03 b 0.30 ± 0.03 a 0.42 ± 0.04 b 0.34 ± 0.03 ab Plant height (cm) 23.37 ± 1.62 a 20.67 ± 1.56 a 15.62 ± 1.28 b 17.32 ± 1.61 a 20.70 ± 1.80 a 21.70 ± 1.41 a Number leaves 3.58 ± 0.14 a 3.77 ± 0.11 a 3.77 ± 0.11 a 3.80 ± 0.14 a 3.61 ± 0.15 a 3.75 ± 0.09 a Damage score 0.00 ± 0.00 a 2.09 ± 0.22 b 4.13 ± 0.33 c 1.86 ± 0.41 a 2.90 ± 0.47 a 2.08 ± 0.40 a Stem thickness (mm) 4.93 ± 0.14 a 4.83 ± 0.12 a 4.09 ± 0.19 b 4.50 ± 0.18 a 4.73 ± 0.15 a 4.64 ± 0.18 a SPAD reading 20.31 ± 0.55 a 19.66 ± 0.66 a 16.39 ± 1.35 b 19.53 ± 0.66 a 19.08 ± 1.14 a 17.99 ± 0.98 a 1 Means followed by the same letter are not significantly different, P > 0.05 142
Table 6-3. Effect of Ischnodemus variegatus density and soil moisture on different Hymenachne amplexicaulis parameters. Second Experiment after 16 days.1 Insect density Soil Moisture Variable Control Medium (four) High (six) Dry Saturation Flood Biomass (g) 0.33 ± 0.02 a 0.30 ± 0.02 a 0.24 ± 0.02 b 0.22 ± 0.02 a 0.30 ± 0.02 b 0.36 ± 0.02 c Plant height (cm) 21.75 ± 1.23 a 15.65 ± 0.85 b 13.34 ± 0.86 c 12.94 ± 1.05 a 16.84 ± 0.85 b 20.97 ± 1.19 c Damage score 0.00 ± 0.00 a 5.25 ± 0.24 b 6.46 ± 0.30 c 4.46 ± 0.70 a 3.46 ± 0.57 b 3.79 ± 0.58 b Stem thickness (mm) 4.72 ± 0.11 a 4.20 ± 0.15 b 3.64 ± 0.16 c 3.70 ± 0.18 a 4.31 ± 0.12 b 4.55 ± 0.14 b SPAD reading 16.67 ± 0.42 a 15.17 ± 0.64 ab 13.50 ± 0.89 b 13.20 ± 0.83 a 15.51 ± 0.51 b 16.61 ± 0.60 b 1 Means followed by the same letter are not significantly different, P > 0.05
Table 6-4. Factorial analyses of the effect of soil moisture, fertilizer dose and insect density on Hymenachne amplexicaulis seedling biomass, height and damage.
Factor Weight Height Damage Mean F P Mean F P Mean F P Square Square Square Water level 1.91 23.35 <.0001 1733.09 31.10 <.0001 3.80 4.26 0.016 Fertilizer dose 0.73 8.94 <0.001 277.81 4.99 0.008 10.50 11.85 <0.001 Insect density 0.11 1.35 0.261 495.98 8.90 <0.001 87.35 97.74 <0.001 Water x fertilizer 0.14 1.74 0.143 116.02 2.08 0.085 3.68 4.11 0.003 Water x insect 0.06 0.70 0.596 76.40 1.37 0.246 0.86 0.97 0.420 Fertilizer x insect 0.06 0.77 0.545 55.22 0.99 0.414 3.19 3.57 0.008 Water x fertilizer x insect 0.06 0.78 0.622 71.67 1.29 0.253 1.54 1.73 0.090
a b c
Figure 6-1. Schematic representation of soil moisture levels used for the factorial experiments: a) dry: no water, b) saturation: water maintained 4 cm from the base of the pots, and c) flood: water level maintained at 1 cm below the top of the pot.
143
45 80 B 40 A 70 SE) ± SE) (
60 ) ±
35 m
50 c ( ng ( 30 40 ght eadi 25 30 hei D r
20 20 ant SPA 10 Pl 15 F= 45.4, d.f.=1,10, P < 0.0001 F = 14.9, d.f.= 1,10, P < 0.003 0 )
6 E C D 7 S ± ( SE) 5 )
± 6 m m e ( 4 (
5 s s
3 e Scor n e k
g 4 a 2 ic th
3 m Dam 1 e t
F = 49.0, d.f. = 1,10, P < 0.0001 F = 0.2, d.f.= 1,10, P = 0.70 S 0 2 0 4 9 13182127310 4 9 1318212731
Days since insect release
Figure 6-2. Changes in Hymenachne amplexicaulis parameters recorded from plants with fertilizer (filled circles) or control (empty circles).
144
A B
accumulation of anthocyanins B
A
Figure 6-3. Hymenachne amplexicaulis leaf damage with fertilizer (A) and control (B).
145
25 High SE)
± Medium 20 ( No bugs m) c
( 15 ght
hei 10 ant Pl 5
) 4 E S (±
3 e or
Sc 2 ge
ma 1 a D 0
SE) 20 ±
18 eading ( AD r 16 SP
14 02468101214
Days after insect release
Figure 6-4. Changes in H. amplexicaulis parameters under control (no bugs), medium (2 fourth instars) and high (4 fourth instars) I. variegatus densities, averaged over all soil moisture treatments
146
5.0 24 )
E A B S High 22 4.5 SE) ± (± 20 )
Medium ( ) m 4.0 Control 18 m c (m (
s 16
s 3.5 e ght
n 14 k 3.0 hei ic 12 t h n t 10 a
m 2.5 Pl e t 8 S 2.0 24 7
) C D 6 SE) SE 22 ± ±
5 (
20 e ng (
4 or
18 c adi s e 3
16 ge
D r 2 14 ma a
SPA 1 12 D 0 10 02468101214160246810121416
Days since insect release
Figure 6-5. Changes in H. amplexicaulis parameters under control (no bugs), medium (4 fourth instars) and high (6 fourth instars) I. variegatus densities, averaged over all soil moisture treatments
147
a 40 a a a a 35 b b b 30 b m) (±SE)
c 25 (
t 20 a 15 10 heigh
t 5 0 Plan Flood Satura Low High Medium Control 036 Soil moisture Fertilizer Bugs a 0.9 a a ±SE) a ( a ) 0.8 g a a ( b 0.7
ass b
m 0.6
o b i 0.5 t B an
l 0.4 P Flood Satura Low High MediumControl 036
2.5 2.0 c 1.5
) 1.0 Flood E
S Saturation ± 0.5 Low water 0.0
score ( Control Fert. Medium Fert. High Fert 3.5 mage
a 3.0 d
d 2.5 f 2.0 Lea 1.5 0 bugs 1.0 3 bugs 0.5 6 bugs 0.0 Control Fert. Medium Fert. High Fert
Figure 6-6. Effect of soil moisture, fertilizer and I. variegatus density on Hymenachne amplexicaulis weight (a), height measured 21 days after insect inoculation (b); and, interactions of soil moisture and fertilizer level (c) and insect density with fertilizer level on leaf damage (d)
148
CHAPTER VII GENERAL CONCLUSIONS AND PERSPECTIVES
The arrival of exotic organisms to new habitats is a major threat to native ecosystems. The
arrival and establishment of I. variegatus in wetlands infested with H. amplexicaulis in Florida allowed the study of different ecological interactions between these exotic organisms. Due to a lack of information on the basic biology, host range, effects and seasonal dynamics of I. variegatus, these subjects were considered critical baseline information for this system.
The temperature-dependent development study provided basic information about the range of temperatures at which I. variegatus can survive and develop. No survival was observed at extreme low and high temperatures. Nymphs died within a few days at 380C and after weeks at lower extreme temperatures, suggesting that I. variegatus has a broader lower temperature
threshold compared to the upper threshold. The overall high mortality observed in the first three
instars may have been due to the rearing of I. variegatus as individuals in our experiments, as
opposed to typical aggregations observed in the field. Harrington (1972) observed that
Ischnodemus species were strongly gregarious and nymphs reared in isolation died sooner than
nymphs reared in groups. Developmental time from egg to adult was three times less at 30.50C
(40 d) compared to 20.50C (122 d), which clearly demonstrated the influence of temperature on development. The range of temperatures where development was fastest occurred between 28 and 330C, which is in agreement with immature survival. These ideal conditions for I. variegatus
development are typical in central Florida from April to October. Egg and nymphal lower
thresholds ranged from 16.6 to 19.10C and 13.7 to 19.90C, respectively. This indicates a greater
susceptibility of eggs to lower temperatures than nymphs. Whereas nymphs can move and locate
microclimates suitable for development (plant structure, conspecific aggregations), eggs are
immobile and successful development depends on local conditions. This greater resistance to
149
lower temperatures of nymphs compared to eggs also has been observed in other heteropteran
insects (He et al. 2003, Bommireddy et al. 2004). The degree-days (588) required to complete
development from egg to adult could be underestimated since the lower threshold is probably
higher than 14.70C. The lower threshold for development of I. variegatus, 18 to 20.50C, explains its mostly tropical and partially subtropical distribution.
This study estimated the number of I. variegatus generations based on long term data of
98 weather stations across Florida and the degree-days required to complete egg to adult development. The GIS map shows spatially the areas across Florida where I. variegatus could establish and the potential number of generations. The current northern and southern invasion fronts of H. amplexicaulis are the St. Johns River (28.080N, 80.750W, Brevard Co.) and Big
Cypress National Park (25.920N, 81.30W, Collier Co.), respectively. The lower winter temperatures in north Florida could be a climatic barrier for the invasion of H. amplexicaulis and
I. variegatus, which are mostly restricted to the tropics.
Field surveys in the native range indicated that I. variegatus had been collected only from
H. amplexicaulis (Baranowski 1979, Baranowski and Slater 2005). Nevertheless, the presence of
I. variegatus in Florida allowed us to study the interactions of a putatively monophagous
herbivore with other plant species. Overall, I. variegatus had a narrow fundamental host range.
Higher adult survivorship and longevity as well as higher fecundity were found on H.
amplexicaulis compared to other species. Field and laboratory tests suggested that H.
amplexicaulis plant quality and I. variegatus density affect the degree of spill-over to suboptimal
host plants. We conclude that feeding on species other than H. amplexicaulis results in high
fitness costs to I. variegatus, which may be a strong selection pressure for maintenance of a
narrow host range.
150
The comparison of life history parameters between I. variegatus reared on H.
amplexicaulis and H. acutigluma revealed further negative fitness effects of feeding on a non-
host. Hymenachne acutigluma is a poor-quality host for I. variegatus compared to H.
amplexicaulis. The immature development study revealed that neonate nymphs reared on H.
acutigluma died after a couple of days and did not molt to the second instar. Third instars reared
on H. acutigluma had a lower survival to the adult stage, a longer developmental time and a
reduction in the size and weight of females compared to those reared on H. amplexicaulis.
Feeding on growing tips attached to rooted plants resulted in lower nymphal survival on H.
acutigluma and a longer developmental time on both plant species. Adult reproductive fitness
parameters of I. variegatus also were low on H. acutigluma. The no-choice reproductive
performance study indicated that individuals feeding on H. acutigluma had a shorter lifespan,
fewer ovipositing females, longer time to first oviposition and lower egg production. Previous
experience (reared from third instar to adult on each species) did not provide any reproductive
advantage because performance was similar to that obtained from adults directly transferred from
H. amplexicaulis. The longer time to first oviposition might reduce the total number of eggs laid
per female in a given lifespan. Under choice conditions, females laid more eggs during their
lifespan on H. amplexicaulis than on H. acutigluma. Similar results were obtained in multiple choice oviposition tests where females laid more eggs on H. amplexicaulis compared to H. acutigluma, Thalia geniculata and Panicum hemitomon. Females of I. variegatus also lived on average 2 months and laid eggs on H. amplexicaulis throughout their lifetime, which indicates that females could be synovigenic. Permanent feeding on inferior hosts such as H. acutigluma
apparently depressed I. variegatus lifetime fecundity.
151
Seasonal fluctuations of I. variegatus in the Myakka River State Park followed a
distinctive 2 year cycle. The hurricane season of 2004 was particularly strong with water levels
rising rapidly during the summer. Despite the adverse flooding conditions of 2004, I. variegatus
populations remained high during late fall of 2004. The appearance of red coloration and later
browning and necrosis on the H. amplexicaulis leaves was associated with a high density of I. variegatus during the summer and fall of 2004. Observations during summer and fall of 2004 revealed that the accumulation of damage by I. variegatus did not seem to negatively influence the vegetative and reproductive allocation of H. amplexicaulis in open deep water marshes with high sediment influx by the Myakka river.
Ischnodemus variegatus sampling demonstrated the absence of reproductive diapause or a distinctive diapause stage during winter. This could influence the ability of I. variegatus to survive colder conditions experienced in northern latitudes. Many tropical and subtropical insects do not enter a diapause stage due to favorable climatic conditions throughout the year (Denlinger
1986). Other more temperate species of the genus Ischnodemus overwinter as adults or nymphs
(Harrington 1972). Currently, the most northern distribution of H. amplexicaulis is located in
Brevard County at latitude 28.1o North.
The 2 year cycle observed at Myakka River State Park probably is more related to stochastic events disrupting population growth than an intrinsic life history strategy of I. variegatus. Once H. amplexicaulis invades a riverine marsh, it forms thick perennial clonal monocultures providing a reliable food source for I. variegatus. Winter survival is probably the more critical aspect limiting I. variegatus population increase the following season. The die-back of H. amplexicaulis starting in late November was followed by a decline in I. variegatus
populations. Therefore, the reduction in host quality and quantity combined with cold
152
temperatures during winter may be part of the normal cycle in Central Florida. Flooding events
during winter and early spring also could leave tillers underwater decimating critical
overwintering insect populations.
The dynamics of I. variegatus, H. amplexicaulis and water levels during 2004 were particularly informative. The plant was able to grow and allocate large amounts of energy during the spring and summer relatively free of I. variegatus damage. This mismatch between the timing for population growth of the plant and its herbivore could be explained in part by an adaptation of H. amplexicaulis to colder conditions, which allowed growth earlier in the season than I. variegatus. It is possible that the origin of the H. amplexicaulis populations in Florida was from colder areas in South America compared to I. variegatus.
The impact of insect density, fertilizer and water level on H. amplexicaulis performance
revealed that under favorable conditions the plant can withstand small infestations of I.
variegatus. Plants exposed to high nutrient levels performed better than control plants (no
fertilizer) despite the feeding pressure of I. variegatus. Hymenachne amplexicaulis stands in
riverine wetlands could be considered areas of high productivity due to sediment deposition,
higher levels of oxygen in the water and potential fertilizer runoffs. Moreover, H. amplexicaulis
responded favorably to the addition of liquid fertilizer by allocating three times more resources
to aboveground biomass compared to the control plants.
Ischnodemus variegatus density and water level had clear impact on H. amplexicaulis
overall performance. Different plant parameters (e.g. biomass, height, chlorophyll content)
decreased with increasing insect densities. Plants with high densities of nymphs had brown
necrotic spots on the leaf blade in the first week of the experiment suggesting a rapid breakdown
of leaf pigments. Higher performance of H. amplexicaulis was obtained under flooded
153
conditions, demonstrating plant adaptability to wetland situations (Kalmbacher et al. 1998,
Kibbler and Bahnisch 1999a). In contrast, low moisture conditions (dry treatment) had a negative impact on the plant, which is in agreement with field observations that showed low tolerance to drought of H. amplexicaulis (Medina and Motta 1990).
Further studies could explore in greater detail the host range of I. variegatus. Since I found that spill-over to other plants under high density in the field, the following information will be useful to better understand potential use of non-hosts: (i) quantify the population growth rates of
I. variegatus on species in the fundamental host range; (ii) establish the effect of plant quality
(especially nutrient status and insect damage) on fitness parameters; (iii) determine the relationship between plant species density and colonization by I. variegatus; and (iv) investigate the behavioral responses of nymphs feeding on a sub-optimal host.
Field sampling in Florida wetlands was a bit challenging especially during the hurricane season of 2004. Extensive flooding in the Myakka River State Park destroyed four of the eight filed cages located in the exclusion study. Cage design and construction apparently was not strong enough to sustain the pressure of moving waters. Another limitation with field sampling involved the accessibility to the Fisheating Creek sites. The mobility of the medium-size boat used in this study was limited by water levels that in many cases where too shallow.
Laboratory experiments were conducted by me with the great assistance of several short- term scholars from Zamorano University (Honduras) and Earth University (Costa Rica). This close interaction with younger scientists facilitated hands-on training in different aspects of biological control, the exchange of ideas on pest management under different socio-economic conditions and the establishment of future cooperative projects in developing countries. I
154
consider the training program at BCRCL a positive step towards building a network of potential collaborators.
155
APPENDIX A IMPACTS OF HYMENACHNE AMPLEXICAULIS ON PLANT AND ARTHROPOD COMMUNITIES
Introduction
Invasion of exotic species poses a serious threat to Florida’s sensitive ecosystems.
International trade, tourism, agricultural and urban disturbance have increased the probability of establishment of exotic plants. In Florida wetlands, exotic species spread rapidly due to floods, large interconnected waterway systems, and increased use of commercial and recreational boats.
Fertilizer and sediment runoff from agricultural lands and waste water from beef and dairy operations contribute to successful establishment of aquatic exotic plants.
Wunderlin and Hansen (2008) reported 1,316 exotic plants species naturalized in Florida,
with 125 species being serious threats to natural areas (FLEPPC 2005). Of those, 65 are
considered highly invasive because they are disruptive to native plant communities. West Indian
Marsh Grass, Hymenachne amplexicaulis (hereafter referred as Hymenachne), is one of many
species currently invading sensitive wetlands in Florida.
Hymenachne is a native of South America and the West Indies and has spread to most
countries of the neo-tropics. The pathway and timing of the introduction of this grass into Florida
are uncertain; however, the first herbarium record was from a ponded pasture in Palm Beach
County in 1957 (University of Florida Herbarium). This suggests that the grass could have been
intentionally introduced as forage. The next record was from a wet pasture in Collier county in
1977 (University of Florida Herbarium). Current records confirm that Hymenachne is present in
wetlands and rivers in 16 counties in Florida.
Hymenachne is present in most the counties included in the Charlotte Harbor National
Estuary Study Area (University of Florida Herbarium and personal communication with aquatic
plant managers). Large monocultures of Hymenachne can be found in the rivers, canals and
156
wetlands located in the Myakka and Peace River Basins. Nutrient enrichment, especially with nitrogen and phosphorous, of surface water due to runoffs from agricultural fields and geological deposits of phosphate may have facilitated the establishment and dominance of Hymenachne in these rivers (Charlotte Harbor Environmental Center, Inc. 2002).
Large infestations of exotic grasses can reduce biodiversity in aquatic ecosystems. Recent studies in wetlands demonstrate that exotic grasses are capable of simplifying the plant diversity and reducing or changing the arthropod community (Herrera and Dudley 2003, Houston and
Duivenvoorden 2002, Talley and Levin 2001, Posey 1988). These changes can be linked to alteration of trophic structure, and habitat usage by birds, fish and other vertebrates. Despite the large areas infested and visible reduction of wetland plants, no studies have been conducted to quantify the impact of Hymenachne on Florida native plant and arthropod communities. The objectives of this study were to quantify the impact of Hymenachne on native flora and macroinvertebrates assemblages in floodplain marshes.
Material and Methods
Study Area
The Myakka River flows through 45 square miles of Myakka River State Park, which is located in Sarasota County in southwest Florida. The land cover in the upper river basin is dominated by a mosaic of pastures, hardwood forest, palms, citrus groves and row crops.
Rainfall is seasonal with most of the rain falling between April and October (Kushlan 1990).
Heavy rain triggers floods during the summer in the park. In the last four years, increased hurricane activity generated large discharges of water into marshes next to the river. Vegetation of marshes adjacent to the Myakka River is composed of: (a) emergent plants dominated by
Panicum hemitomon (native), Brachiaria mutica and Hymenachne amplexicaulis (exotics), and
Polygonum spp.; (b) free-floating plants, including Salvinia minima, Lemna spp. and in lesser
157
abundance, Eichornia crassipes; (c) littoral plants, including Quercus spp, Sabal minor and
Sabal palmetto. The park also contains flag marshes that are dominated by Pontederia cordata,
Saggitaria spp., Thalia geniculata and other species with flag-like leaves. Flag marshes occur where the wet season water depth is between 0.3 and 1 m and the hydroperiod extends more then
200 days per year (Kushlan 1990).
Sampling
Three floodplain marshes (27016’N, 82016’W; Fig. A-1) located in the east side of the
Myakka River were monitored. In each marsh, there were sites dominated by either H.
amplexicaulis (exotic) or P. hemitomon (native) and separated at least by 10 meters. At each site
a linear transect was established randomly along the longer axis of the patch. Four sampling
stations separated by 10 meters were located along the transect. Sites were sampled once for
water quality and plant diversity during August 2006. At each sampling station the following
variables were measured: plant diversity and coverage, water quality, aerial and aquatic
arthropod abundance, light attenuation and above ground biomass. Plant diversity was measured
within 0.25 m2 quadrats and cover was assessed with the Daubenmire coverage scale (1= 1-5%,
2= 6-25%, 3=26-50%, 4=51-75%, 5=76-95%, 6=96-100%) (Elzinga et al 2001).
Water quality parameters were collected once during August 2006 with a YSI 556 Multi
Probe System between 9 and 11 AM at a depth of 20 cm and included temperature, pH, conductivity and dissolved oxygen. Aerial and aquatic samples were collected once in the summer (August) and once in fall (November) of 2006. Aerial arthropod abundance was estimated using sweeping nets (24x20 mesh per inch, BioQuip, California). Ten strokes separated by one meter were collected at each station. All the material collected in the net was transferred to plastic bags containing 70% EtOH. Aquatic arthropods were collected at a depth of ca. 30 cm using a D-shaped aquatic net (500 micron Nytex, BioQuip California). The aquatic net
158
was vigorously agitated inside the plant community for a period of 30 seconds covering an area of 1.5 m wide and 2 m long. All insect samples were identified to family level using the taxonomic keys from Borror et al. (1989) and later pined for preservation.
Light attenuation was estimated using a light meter (Extech Digital Foot Candle/Lux
Light Meter). Measurements were collected at the top of the canopy and at the bottom (water surface). Above ground biomass was estimated at each station using a small PVC quadrat (0.50 m2) and hand clippers. Plants were removed from above the substratum, placed in plastic bags and then dried for ten days at 700C.
Statistical Analysis
Macroinvertebrate diversity among native and Hymenachne sites was compared using the
Shannon-Wiener Index, Simpson diversity Index and Evenness. Abundance of macroinvertebrate orders was tested using ANOVA (Proc GLM, SAS Institute) with plant community as a factor.
We performed the analysis only for the summer sampling due to a missing aquatic sample in the fall. Low water levels in Marsh 1 during the fall of 2006 limited the collection of samples. The exotic blissid Ischnodemus variegatus was found in large quantities in areas where Hymenachne amplexicaulis was present and its abundance was not included in the analysis.
Results
Marsh Characteristics
Hymenachne amplexicaulis colonized the deeper part of marshes where standing water was present most of the year. Floating mats where observed in the deeper part of the marshes and they might contribute to short distance dispersal inside the marsh during flooding events. Water quality parameters were not different between the marshes (Table A-1).
159
Plant Community and Biomass Accumulation
Plant communities had very few species. Invaded sites were dominated by Hymenachne
amplexicaulis and native sites by either Panicum hemitomon or Leersia hexandra (Table A-2).
Scores of plant cover were higher in the Hymenachne sites indicating complete dominance of the
space available compared with native sites. There was a greater accumulation of biomass in the
Hymenachne sites, especially in Marsh 1. This could be attributed to the deeper water levels
(Table A-1). During flooding events in the summer of 2004 and 2005 at Myakka River State
Park, the only plant emerging from the water was Hymenachne amplexicaulis, demonstrating its
potential for outcompeting native species. We hypothesize that Hymenachne is more problematic
in deeper marshes located adjacent to the Myakka River due to the direct influence of nutrient
runoff and drastic changes in the hydroperiod.
Macroinvertebrates Summer
A total of 1841 invertebrates were collected in the native sites versus 628 in the
Hymenachne sites during the summer of 2006. Fifteen invertebrate orders were collected from
the six sites (Table A-3, A-4). Insects compromised most of the invertebrates collected in the
aerial and aquatic samples. The insect orders Coleoptera and Hemiptera had more families in the
native than in the Hymenachne sites (Table A-3). Insect families common in the samples
included Formicidae, Chironomidae, Curculionidae, Noteridae, and Cicadellidae. These families
had greater number of individuals in the native sites. Insect herbivore families such as Acrididae,
Tettigonidae, Tetrigidae, Curculionidae and Cicadellidae were less abundant in the Hymenachne
sites (Table A-3). The native blissid Ischnodemus brunnipennis was found in large numbers in the native sites (Table A-3). This sap-sucking herbivore feeds primarily in the phloem of
Panicum hemitomon and its population builds during the summer and fall. Ichneumonidae,
160
Braconidae and other parasitic Hymenoptera were only found in the native sites, which may
suggest the presence of more complex food webs functioning in these sites.
The Shannon-Wiener Index, Simpson Index and Evenness were not different in the
Hymenachne and native sites (Table A-4). However, there was a clear simplification of
macroinvertebrate fauna in the Hymenachne sites (Table A-3, A-4). Mean abundance of
macroinvertebrate orders Diptera, Coleoptera, Hemiptera, Orthoptera and Araneae were
significantly greater in the native sites compared to the Hymenachne sites (Fig. A-2).
Macroinvertebrates Fall
Overall, there was a lower number of invertebrates collected during the fall than in the
summer. A total of 813 invertebrates were collected in the native sites versus 657 in the
Hymenachne sites during the fall of 2006. Thirteen invertebrate orders were collected in the
aerial and aquatic samples. Insect orders with greatest number of families included Hemiptera
and Coleoptera. The insect families Chironomidae, Hydrophilidae and Cicadellidae were more
abundant in the native than in Hymenachne sites (Table A-5). Insect families containing species
only in the native sites included: Berytidae, Coreidae, Naucoridae, Pentatomidae, Tingidae,
Hesperidae and Chrysopidae. One specimen of the parasitic order Strepsiptera (cicadellid host) was found in the native site in Marsh 2. Insect herbivore families such as Acrididae,
Curculionidae and Cicadellidae were less abundant in the Hymenachne sites (Table A-5). The exotic blissid Ischnodemus variegatus was found in large numbers in Hymenachne sites (Table
A-5) and it was the most common insect species on these sites. Similar to I. brunnipennis, I.
variegatus is a specialist phloem feeder and reproduces primarily on Hymenachne amplexicaulis.
Host range studies conducted in greenhouses at the Biological Control Research and
Containment Laboratory (BCRCL)-Fort Pierce determined that I. variegatus can maintain stable
161
populations only on Hymenachne amplexicaulis. However, under outbreak conditions I. variegatus can spill-over temporally to nearby grasses.
The Shannon-Wiener Index, Simpson Index and Evenness were not different in the
Hymenachne and native sites (Table A-6). Additionally, there was no difference between mean abundance of invertebrate orders between the sites (Fig. A-3).
Discussion
The invasion of Hymenachne amplexicaulis in Myakka River is an ongoing process that creates major changes in the wetland communities. Hymenachne was able to out-compete native plant species and dominate in the deeper parts of the marshes. The adaptation to rapid changes in the water level and nutrient uptake could provide a competitive advantage for Hymenachne in
Florida, especially during summer flooding. Physiological adaptations to rising in water levels include rapid elongation of stems, formation of adventitious roots, presence of aerenchyma in the stem, leaf and root tissues, and changes in leaf:shoot ratio (Kibbler and Bahnisch 1999a). In the
Brazilian Pantanal, its native range, Hymenachne grows within four plant formations: marsh ponds, waterlogged basins, tall grasslands and forest edges (Pinder and Ross 1998). Observations of marshes in Myakka River State Park suggest that when subject to inundation, Hymenachne is capable of fast stem elongation, increase in foliage volume and rapid nodal adventitious root production (R. Diaz observations). This adaptation to wetter conditions could be used as a management strategy since Hymenachne is known to be less tolerant to drought (Medina and
Motta 1990). Changes in the hydroperiod in managed wetlands could ameliorate the invasion of
Hymenachne in Florida.
The total plant biomass in the Hymenachne invaded sites was larger than in the native ones. This trend is confirmed from Australian wetlands where Hymenachne is an aggressive exotic weed. Houston and Duivenvoorden (2002) found that invaded wetlands in Queensland
162
had up to 30 fold greater biomass allocation than native sites. Hymenachne is also recognized in the native range as important forage for cattle. In Venezuela, Tejos (1978) found a positive relationship between Hymenachne growth and depth of flooding. Biomass production ranged from 5,911 - 18,162 t/ha/yr during the flood period and from 5,553 - 7,836 t/ha/yr during the dry season.
Invasive plants usually change the faunal composition in the adventive range because they reduce the diversity of foods available, simplify habitats and reduce chances of colonization. The simplification of the macroinvertebrate composition in the Hymenachne sites may be explained by the lack of adaptation of native fauna to this grass, reduction of physical space due to higher density of stems, and greater accumulation of leaf litter and sediments. The simplification of faunal assemblages in areas invaded by exotic plants has been extensively documented (Herrera and Dudley 2003, Houston and Duivenvoorden 2002, Tadley and Levin
2001, Posey 1988, Degomez and Wagner 2001, but see Hager and Vinebrooke 2004). In this study, the changes in macroinvertebrate fauna was evident particularly in the summer sampling where Hymenachne sites had lower abundance of individuals, especially in the orders Diptera,
Coleoptera and Hemiptera compared to native sites. The only insect species that was collected in large numbers in Hymenachne sites was the exotic true bug, Ischnodemus variegatus, which is a
Hymenachne specialist native to South America. Abundance of other insect herbivore families was much lower in the Hymenachne sites. The paucity of predators (spiders, reduviids, dragonflies) that responded to the increase in I. variegatus abundance during the fall of 2006, coupled with the low abundance of insect herbivores present in the Hymenachne sites, support the hypothesis that the colonization of exotic plants by native insects is a slow process.
163
Freshwater marshes in Myakka River are threatened by the invasion of Hymenachne amplexicaulis due the drastic changes in the flora and fauna and allocation of large biomass in the wetland. This study demonstrated that sites invaded by Hymenachne contain simple macroinvertebrate assemblages. The expansion and impact of this highly competitive grass in
Florida might increase due to the interconnection of watersheds and erratic changes in the hydroperiod associated with summer floods.
164
Marsh 3
Marsh 2
Marsh 1
Figure A-1. Location of marshes at Myakka River State Park (27016’N, 82016’W).
165
90 b 80 Hymenachne 70 Native e
t b i
s 60
r e
p 50 e
ag 40
er a b v 30 A a 20 a a a b b 10 aa a a aa 0
a ra a a a s ra r er ae at r te te te e er he p p ip opt n pt Di eo n ra ho don Ot l em e A rt O Co H m O Hy
Figure A-2. Mean abundance of arthropods in native and Hymenachne sites. Summer 2006. (ANOVA, d.f. 1,4, P <0.05).
60 Hymenachne 50 Native
e 40 t
30
erage per si 20 v A 10
0
a a a e a a r r r a at r rs te te te e n te e p p p an o h o r op Ot Di le mi A Od n o C He me Hy
Figure A-3. Mean abundance of arthropods in native and Hymenachne sites. Fall 2006
166
Table A-1. Water quality data collected in native (N) and Hymenachne (H) sites1 in three marshes at Myakka River State Park Variable Marsh 1 Marsh 2 Marsh 3 N H N H N H Depth (cm) 59.8 101.8 57.8 64.3 60.3 65.8 s.d. 2.99 2.75 6.13 4.11 3.30 2.75 Temperature (0 C) 26.4 26.4 25.9 26.0 26.1 26.2 s.d. 0.08 0.24 0.20 0.24 0.13 0.49 Dissolved oxygen (mg L-1) 2.91 3.04 1.24 1.39 1.57 1.18 s.d. 0.42 0.16 0.28 0.23 0.40 0.56 pH 3.14 4.67 5.10 5.00 4.70 4.90 1 Mean and S.D. of four samples
Table A-2. Average emergent stems per unit of area and cover in native (N) and Hymenachne (H) sites1 in three marshes at Myakka River State Park Marsh 1 Marsh 2 Marsh 3
Species N Cover H Cover N Cover H Cover N Cover H Cover Hymenachne amplexicaulis 0.25 <1 48.25 6 0.5 1 41.25 5 - - 40 5 Panicum hemitomon 32 4 - - 68.5 5 25 2 39 4 13.25 1 Polygonon sp. 1.25 <1 - - 3 1 3 1 - - - - Salvinia minima - - Present <1 ------Pontederia cordata - - - - 0.5 1 - - - - 0.25 <1 Leersia hexandra 21 3 ------Vine 0.25 <1 - - 1.5 1 - - 1.75 1 - -
Biomass (g 0.5m-2) 180.3 530.0 250.3 337.8 126.5 295.0 Biomass S.D. 40.2 129.6 104.5 72.8 56.7 71.0 Species richness 5 2 5 3 2 3 Light attenuation 632 519.5 756.5 1117 589.5 379 Light SD 90.5 222.8 511.4 44.0 386.0 251.6 1Coverage scale: 1=1-5%, 2=6-25%, 3=26-50%, 4=51-75%, 5=76-95%, 6=96-100%
167
Table A-3. Macroinvertebrates collected from aquatic and aerial samples during Summer 2006 Taxon Native sites Hymenachne sites
Marsh Marsh Marsh Marsh Marsh Marsh 1 2 3 1 2 3 INSECTA Hymenoptera 6 14 68 54 7 5 Formicidae 1 13 67 54 7 5 Vespidae 2 0 0 0 0 0 Ichneumonidae 1 0 0 0 0 0 Braconidae 0 1 0 0 0 0 Parasitic Hymenoptera 2 0 1 0 0 0 Orthoptera 28 3 11 2 3 7 Acrididae 22 1 6 2 3 2 Tettigonidae 4 0 5 0 0 5 Gryllacrididae 0 1 0 0 0 0 Tetrigidae 2 1 0 0 0 0 Diptera 288 173 110 127 14 87 Unknown Diptera 74 3 26 8 11 4 Chironomidae (larvae) 214 170 84 117 3 81 Culicidae larvae 0 0 0 0 0 1 Diptera larvae 0 0 0 2 0 1 Coleoptera 81 121 133 36 48 61 Anthicidae 0 0 1 0 2 0 Silvanidae 0 1 0 0 0 0 Dysticidae 2 0 0 0 0 0 Tenebrionidae 2 1 5 0 1 3 Dryopidae 1 1 0 1 0 0 Carabidae 0 1 0 0 0 0 Chrysomelidae 3 0 6 0 1 1 Curculionidae 3 29 44 4 1 22 Haliplidae 0 3 4 0 3 0 Hydrophilidae 4 32 4 1 5 4 Noteridae 10 18 52 26 18 21 Scarabidae 0 2 3 0 0 2 Staphylinidae 33 4 1 4 12 7 Unknown Coleoptera 7 3 0 0 0 0 Unknown larvae 16 26 13 0 5 1 Hemiptera 99 346 268 47 44 94 Aphididae 0 0 1 0 0 1 Belastomatidae 2 20 5 0 2 2 Blissidae 2 1 12 0 0 2 Cicadellidae 58 9 47 11 21 33 Delphacidae 5 0 6 1 0 5 Dictyopharideae 2 0 2 0 0 0 Pleidae 0 0 0 1 0 0 Hebridae 0 0 1 0 0 0
168
Table A-3. Continued Taxon Native sites Hymenachne sites
Marsh Marsh Marsh Marsh Marsh Marsh 1 2 3 1 2 3 Ischnodemus brunnipennis 20 305 190 4 2 35 Ischnodemus variegatus 0 0 0 26 17 4 Ischnodemus sp. 0 0 0 2 1 1 Lygaeidae 0 1 0 0 0 0 Membracidae 0 0 1 0 0 0 Miridae 1 0 0 0 0 0 Naucoridae 1 1 0 1 0 0 Notonectidae 1 0 0 0 0 0 Pentatomidae 6 2 0 0 0 0 Phymatidae 0 0 1 0 0 0 Reduviidae 0 0 1 0 0 11 Tingidae 0 0 0 1 0 0 Unknown Hemiptera 1 7 1 0 1 0 Thysanoptera 2 8 0 0 2 0 Lepidoptera 1 0 1 0 2 0 Collembola 0 0 0 0 3 0 Entomobriidae 0 0 0 0 3 0 Odonata 3 1 1 1 1 4 Coenagrionidae 1 0 0 0 0 2 Nymph 2 1 1 1 1 2 Trichoptera 0 0 0 0 0 1 Mantodea 0 0 1 0 0 0 Mantidae 0 0 1 0 0 0 Plecoptera 0 7 0 2 0 0 larvae 0 7 0 2 0 0 Blattaria 0 0 0 0 0 1 Blattidae 0 0 0 0 0 1 ARACHNIDA Araneae 31 19 13 5 3 17 Acari 4 0 0 0 0 1 Ixodida 0 0 0 0 0 1 Acariformes 4 0 0 0 0 0
Insecta Order richness 8 8 8 7 9 8 Coleoptera family richness 9 11 9 4 8 7 Hemiptera family richness 9 6 10 6 3 6 Total abundance 543 692 606 274 127 278 Total abundance without I. variegatus 543 692 606 248 110 274
169
Table A-4. Total count of arthropods and diversity indexes. Summer 2006. Native sites Hymenachne sites Marsh Marsh Marsh Marsh Marsh Marsh ORDER 1 2 3 1 2 3 Hymenoptera 6 14 68 54 7 5 Orthoptera 28 3 11 2 3 7 Diptera 288 173 110 127 14 87 Coleoptera 81 121 133 36 48 61 Hemiptera 99 346 268 21 27 90 Thysanoptera 2 8 0 0 2 0 Lepidoptera 1 0 1 0 2 0 Collembola 0 0 0 0 3 0 Odonata 3 1 1 1 1 4 Trichoptera 0 0 0 0 0 1 Mantodea 0 0 1 0 0 0 Plecoptera 0 7 0 2 0 0 Blattaria 0 0 0 0 0 1 Araneae 31 19 13 5 3 17 Acari 4 0 0 0 0 1 Shannon-Index 2.9 2.6 2.7 2.6 3.7 3.1 Order Richness (S) 10 9 9 8 10 10 Total Abundance 543 692 606 248 110 274 Simpson Diversity Index D: 0.34 0.34 0.29 0.34 0.27 0.26 1-D: 0.66 0.66 0.71 0.66 0.73 0.74 1/D: 2.92 2.9 3.45 2.95 3.65 3.80 Evenness 0.61 0.59 0.65 0.65 0.71 0.66
170
Table A-5. Macroinvertebrates collected from aquatic and aerial samples during Fall 2006 Native Hymenachne
Marsh Marsh Marsh Marsh Marsh Marsh Taxon 1 2 3 1 2 3 INSECTA Hymenoptera 0 2 13 2 1 3 Formicidae 0 1 11 1 1 1 Parasitic Hymenoptera 0 1 2 1 0 2 Orthoptera 3 3 6 0 0 3 Acrididae 3 3 6 0 0 3 Diptera 8 287 87 89 102 148 Bibionidae 0 18 16 14 0 18 Chironomidae (larvae) . 236 58 50 100 109 Culicidae 2 26 3 5 0 7 Culicidae larvae . 0 0 2 0 1 Unknown larvae . 1 1 0 0 0 Unknown Diptera 6 3 8 18 2 13 Coleoptera 1 84 53 28 81 49 Carabidae 1 1 3 0 0 1 Chrysomelidae 0 1 1 3 0 0 Coccinellidae 0 1 0 1 0 1 Curculionidae 0 10 13 0 1 9 Dysticidae . 0 0 1 0 0 Haliplidae . 1 6 0 4 1 Hydrophilidae . 26 8 10 19 5 Noteridae . 9 11 8 49 19 Phalacridae 0 0 2 0 0 1 Staphylinidae . 8 6 2 1 1 Tenebrionidae . 1 0 0 0 1 Unknown Coleoptera 0 0 0 2 0 1 Unknown larvae 0 26 3 1 7 9 Hemiptera 73 50 95 168 560 380 Aphididae 1 1 3 12 0 0 Belastomatidae . 1 2 0 4 0 Berytidae 0 1 0 0 0 0 Cicadellidae 15 11 32 3 16 17 Coreidae 0 0 2 0 0 0 Delphacidae 0 4 13 0 2 4 Derbidae 0 0 0 0 0 1 Gerridae . 0 0 0 0 1 Rhyparochromidae 0 0 2 0 0 1 Hebridae . 0 1 0 1 0 Ischnodemus brunnipennis 1 30 4 0 0 0 Ischnodemus sp.1 0 0 5 0 0 0 Ischnodemus variegatus 54 0 24 153 536 354 Lygaeidae 1 0 0 0 0 1 Naucoridae . 1 0 0 0 0
171
Table A-5. Continued
Taxon Native Hymenachne Marsh Marsh Marsh Marsh Marsh Marsh 1 2 3 1 2 3
Pentatomidae 0 1 7 0 0 0 Tingidae 1 0 0 0 0 0 Thysanoptera 6 1 0 2 1 0 Lepidoptera 3 0 3 0 3 2 Hesperidae 1 0 0 0 0 0 Unknown larvae 0 0 2 0 0 2 Unknown moth 2 0 1 0 3 0 Odonata 2 3 10 2 2 7 Coenagrionidae 2 1 6 0 2 1 Unknown nymph . 2 4 2 0 6 Ephemeroptera larvae 0 0 0 0 1 0 Neuroptera 4 0 0 1 2 0 Neuroptera larvae 1 0 0 1 2 0 Chrysopidae 1 0 0 0 0 0 Unknown Neuroptera 2 0 0 0 0 0 Strepsiptera 0 1 0 0 0 0 Collembola 0 0 4 0 0 0 Sminthuridae . 0 4 0 0 0 Araneae 14 29 46 41 3 19
Insecta Order richness 9 8 8 7 9 7 Coleoptera family richness 1 9 9 6 5 9 Hemiptera family richness 5 8 9 2 4 6 Total abundance 114 460 317 333 756 611 Total abundance without I. variegatus 60 460 293 180 220 257
172
Table A-6. Total count of arthropods and diversity indexes. Fall 2006.
Native sites Hymenachne sites Marsh Marsh Marsh Marsh Marsh Marsh ORDER 1 2 3 1 2 3 Hymenoptera 0 2 13 2 1 3 Orthoptera 3 3 6 0 0 3 Diptera 8 287 87 89 102 148 Coleoptera 1 84 53 28 81 49 Hemiptera 19 50 71 15 24 26 Thysanoptera 6 1 0 2 1 0 Lepidoptera 3 0 3 0 3 2 Odonata 2 3 10 2 2 7 Ephemeroptera 0 0 0 0 1 0 Neuroptera 4 0 0 1 2 0 Strepsiptera 0 1 0 0 0 0 Collembola 0 0 4 0 0 0 Araneae 14 29 46 41 3 19 Shannon-Index 4.0 2.3 3.6 2.6 2.5 2.7 Species Richness (S) 9 9 9 8 10 8 Total Abundance 60 460 293 180 220 257 Simpson Diversity Index D: 0.19 0.44 0.21 0.33 0.36 0.39 1-D: 0.81 0.56 0.79 0.67 0.64 0.62 1/D: 5.17 2.28 4.81 3.05 2.76 2.60 Evenness 0.85 0.52 0.79 0.65 0.54 0.62
173
LIST OF REFERENCES
Agrawal, A. A., Konaten, P. M., Mitchell, C. E., Power, A. G., Godsoe, W., and J. Klir onoms. 2005. Enemy release? An experiment with congeneric plant pairs and diverse above- and belowground enemies. Ecology 86: 2979-2989.
Anonymous. 2008. Hymenanchne or Olive Hymenachne-Weed Management Guide. Tropical Weeds Research Centre. Queensland, Australia.
Anten, N. P. R., M. J. A. Werger, and E. Medina. 1998. Nitrogen distribution and leaf area indices in relation to photosynthetic nitrogen efficiency in savanna grasses. Plant. Ecol. 138: 63-75.
Aldrich, J. R. 1998. Chemical ecology of the Heteroptera. Ann. Rev. Entomol. 33: 211-238.
Anderson, W.G., Heng-Moss, T.M., Baxendale, F.P., Baird, L.M., Sarath, G. and L. Highley.2006. Chinch bug (Hemiptera: Blissidae) mouthpart morphology, probing frequencies, and locations on resistant and susceptible germplasm. J. Econ. Entomol. 99: 212–221.
Andres, L. A. and F. D. Bennet. 1975. Biological control of aquatic weeds. Ann. Rev. Entomol. 20:31-46.
Awamack, C. S., and S. R. Leather. 2002. Host plant quality and fecundity in herbivorous insects. Ann. Rev. Entomol. 47: 817-844.
Baldwin, I. T. 1999. Inducible nicotine production in native Nicotiana as an example of adaptive phenotypic plasticity. J. Chem. Ecol. 25: 3-30.
Baranowski, R. M. 1979. Notes on the biology of Ischnodemus oblongus and I. fulvipes with descriptions of the immature stages (Hemiptera: Lygaeidae). Ann. Entomol. Soc. Am. 72: 655-658.
Baranowski, R. M., and J. A. Slater. 2005. The Lygaiedae of the West Indies. Fla. Agr. Exp. Bull. 402.
Baskerville, G. L., and P. Emin. 1969. Rapid estimation of heat accumulation from maximum and minimum temperatures. Ecology 50: 514-517.
Baxendale, F. P., T. M. Heng-Moss, and T. R. Riordan. 1999. Blissus occiduus (Hemiptera: Lygaeidae): a chinch bug pest new to buffalograss turf. J. Econ. Entomol. 92: 1171-1176.
Bayoh, M. N., and S. W. Lindsay. 2004. Temperature-related duration of aquatic stages of the Afrotropical malaria vector mosquito Anopheles gambiae in the laboratory. Med. Vet. Entomol. 18: 174–179.
174
Bernays E. A. 1985. Arthropods for weed control in IPM systems, pp. 373-391. In M.A. Hoy and D.C. Herzog (eds.), Biological Control in Agricultural Integrated Pest Management Systems. Academic Press.
Bernays E., and M. Graham. 1988. On the evolution of host specificity in phytophagous arthropods. Ecology 69: 886-892.
Bernays E. A., and R. F. Chapman. 1994. Host plant selection by phytophagous insects. Chapman & Hall, New York.
Bi, J. L., S. J. Castle, and N. C. Toscano. 2007. Amino acid fluctuations in young and old orange trees and their influence on glassy-winged sharpshooter (Homalodisca vitripennis) population densities. J. Chem. Ecol. 33: 1692-1706.
Blossey B., R. Casagrande, L. Tewksbury, D. A. Landis, R. N. Wiedenmann, and D. R. Ellis. 2001. Non-target feeding of leaf beetles introduced to control purple loosestrife (Lythrum salicaria). Nat. Area. J. 21: 368-377.
Blossey B., M. Schwarzländer, P. Häfliger, R. Casagrande, and L. Tewksbury. 2002. Common Reed. In R. Van Driesche, S. Lyon, B. Blossey, M. Hoddle, and R. Reardon (eds.). Biological Control of Invasive Plant in Eastern United States. USDA Forest Service Publication FHTET-2002-04. (http://www.invasive.org/eastern/biocontrol/9CommonReed.html)
Blossey, B. and R. Notzold 1995. Evolution of Increased Competitive Ability in Invasive Nonindigenous Plants: A Hypothesis. J. Ecol. 83: 887-889.
Blumental, D. M. 2005. Interrelated causes of plant invasion. Science. 310:243-244.
Blumental, D M. 2006. Interactions between resource availability and enemy release in plant invasion. Ecol. Lett. 9: 887-895.
Bommireddy, P. L., M. N. Parajulee, and D. O. Porter. 2004. Influence of constant temperatures on life history of immature Lygus elisus (Hemiptera: Miridae). Environ. Entomol. 33: 1549-1553.
Borror, D. J., Triplehorn, C. A. and N. F. Johnson. 1989. An Introduction to the Study of Insects, 6th ed. Philadelphia: Saunders College Publishing.
Brambila, J. and F. Santana. 2004. First records for Ischnodemus variegatus (Hemiptera: Blissidae) in North America. Fla. Entomol. 87: 585-586.
Brière, J. F., P. Pracos, A. Y. Le Roux, and J. S. Pierre. 1999. A novel model of temperature- development for arthropods. Environ. Entomol. 28: 22-29.
175
Brodbeck, B. V., P. C. Andersen, R. F. III Mizell, and S. Oden. 2004. Comparative nutrition and developmental biology of xylem-feeding leafhoppers reared on four genotypes of Glycine max. Environ. Entomol. 33: 165-173.
Bryant, R. S., C. D. Thomas, and J. S. Bale. 2002. The influence of thermal ecology on the distribution of three nymphalid butterflies. J. of Appl. Ecol. 39: 43-55.
Buschmand, L. L., and W. H. Whitcomb. 1980. Parasites of Nezara viridula (Hemiptera: Pentatomidae) and other Hemiptera in Florida. Fla. Entomol. 63:154-162.
Campbell, A., B. D. Frazier, N. Gilbert, A. P. Gutierrez, and M. Mackauer. 1974. Temperature requirements of some aphids and their parasites. J. Appl. Ecol. 11:431-438.
Carrol, S. P., and C. W. Fox. 2007. Dissecting the evolutionary impacts of plant invasions: bugs and beetles as native guides. Glob. Chang. Biol. 12: 1644-1657.
Catalogue of New World Grasses. 2008. Missouri Botanical Garden. [Internet]. [cited 2008 February 4]; Available from: http://mobot.mobot.org/W3T/Search/nwgc.html
Chalker-Scott, L. 1999. Environmental significance of anthocyanins in plant stress responses. Photochem. Photobiol. 70: 1-9.
Charleston K. 2006. Hymenachne. Hymenachne amplexicaulis, control methods and case studies. Department of Natural Resources, Mines and Water. Australia.
Charlotte Harbor Environmental Center, Inc. 2002. Water quality conditions and pollutant loads and yields in the Peace and Myakka River Basins. Punta Gorda, Florida.
Ciepiela, A. P., and C. Sempruch. 1999. Effect of L-3,4-dihydroxyphenylalanine, ornithine and γ-aminobutyric acid on winter wheat resistance to grain aphid. J. App. Entomol. 123: 285-288.
Cipollini, D., Mbagwu, J., Barto, K., Hillstrom, C. and S. Enright. 2005. Expression of constitutive and inducible chemical defenses in native and invasive populations of Alliaria petiolata. J. Chem. Ecol. 31:1255–1267
Colautti, R. I., Ricciardi, A., Grigorovih, I. A., H. J. MacIsaac. 2004. Is invasion success explained by the enemy release hypothesis? Ecol. Lett. 7: 721-733.
Cole, L. C. 1949. The measurement of interspecific association. Ecology 30: 411-424.
Coracini, D. L. A. and R. I. Samuels. 2002. Natural enemies of the chinch bug, Blissus antillus Leonard (Hemiptera: Lygaeidae: Blissinae), pasture pest in Rio de Janeiro State, Brazil. Neotrop. Entomol. 31: 165-167.
176
Costa-Arbulu, C., Gianoli, E., Gonzales, W. L. and H. M. Niemeyer. 2001. Feeding by the aphid Sipha flava produces a reddish spot on leaves of Sorghum halapense: A induced defense? J. Chem. Ecol. 27: 273-283.
CLIMOD. 2007. Southeast Regional Climate Center. The University of North Carolina, Chape Fill, NC. [Internet]. [cited 2007 February 4], http://climod.nrcc.cornell.edu/
CSIRO (Commonwealth Scientific and Industrial Research Organisation). 1973. Commonwealth Plant Introductions, Plant Introduction Review 9, No. 2, List No. 94, Nos. 58026 to 60472.
Csurches, S. M., A. P. Mackey, and L. Fitzsimons. 1999. Hymenachne in Queensland. Pest Status Review Series. Queensland Government-Natural Resources and Mines. [Internet]. [cited 2008 February 4] http://www.nrm.qld.gov.au/
Cuda, J.P., H.B. Johnson, and C.R. Tischler. 1995. Trophic interactions between Mozena, mesquite and a microbe: implications for host-specificity testing of insects of leguminous weeds, p. 575. In E. S. Delfosse and R. R. Scott (eds.), Proceedings of the VIII International Symposium on Biological Control of Weeds, 2-7 February 1992, Lincoln University, Canterbury, New Zealand, DSIR/CSIRO, Melbourne.
Dambach, M, and B. Goehlen. 1999. Aggregation density and longevity correlate with humidity in first instar nymphs of the cockroach (Blattella germanica L., Dictyoptera). J. Insect Physiol. 45: 423-429.
Davis, M. A, Grime, J. P. and K. Thompson. 2000. Fluctuating resources in plant communities: a general theory of invasibility. J. Ecol. 88: 528-534
DEP. 2008. Florida Department of Environmental Protection. Research Programs. Tallahase, FL. [Internet]. [cited 2008 March 4]; Available from: http://www.dep.state.fl.us/lands/invaspec/2ndlevpgs/Research.htm
Degomez, T. and M. Wagner. 2001. Arthropod diversity of exotic vs. native Robinia species in northern Arizona. Agric. Forest Entomol. 3:1-19
Denlinger, D. L. 1986. Dormancy in tropical insects. Ann. Rev. Entomol. 31: 239-264.
DeWalt, S. J., Denslow, J. P. and K. Ickes. 2004. Natural enemy release facilitates habitat expansion on the invasive tropical shrub Clidemia hirta. Ecology 85: 471-483.
Dhileepan, K, M. Treviño, and S. Raghu. 2006. Temporal patterns in incidence and abundance of Aconophora compressa (Hemiptera: Membracidae), a biological control agent for Lantana camara, on target and nontarget plants. Environ. Entomol. 35: 1001-1012.
177
Diaz, R., W. A. Overholt, J. P. Cuda, D. A. Pratt, and A. Fox. 2008a. Temperature-dependent development, survival and potential distribution of Ischnodemus variegatus (Hemiptera: Blissidae), an herbivore of West Indian marsh grass (Hymenachne amplexicaulis). Ann. Entomol. Soc. Am. (in press).
Diaz, R., W. A. Overholt, J. P. Cuda, D. A. Pratt, and A. Fox. 2008b. Host specificity of Ischnodemus variegatus, an herbivore of West Indian marsh grass (Hymenachne amplexicaulis). Submitted to Biocontrol Science and Technology.
DiTomaso, J. M. 2000. Invasive weeds in rangelands. Species, impacts, and management. Weed Science. 48:255-265.
Dye, C.W.1995. Florida's urban lakes: threatened treasures of the Sunshine state. Lakeline 15: 20–21, 38–39
EDD Maps. 2008. Early detection and distribution mapping system. Florida Exotic Pest Plant Council. http://www.fleppc.org/EDDMapS/
Elzinga, C. Salzer, D. W., Willoughby, J, and J. Gibbs. 2001. Monitoring Plant and Animal Populations. Blackwell Science. MA
Enriquez-Quiroz, J. F., A. R. Quero-Carrillo, A. Hernandez-Garay, and E. Garcia-Moya. 2006. Azuche, Hymenachne amplexicaulis (Rudge) Nees, forage genetic resources for floodplains in tropical Mexico. Genet. Resour. Crop Ev. 53: 1405-1412.
FLEPPC 2007. Florida Exotic Pest Plant Council. Gainesville, Florida. [Internet]. [cited 2007 September 22]; Available from: http://www.fleppc.org/list/list.htm
Franks, S. J., Pratt, P. D., Dray, F. A. and E. L. Simms. 2008. No evolution of increased competitive ability or decreased allocation to defense in Melaleuca quinquenervia since release from natural enemies. Biol. Invasions. 10:455-566.
Girousse, C., B. Moulia, W. Silk, and J. Bonnemain. 2005. Aphid infestation causes different changes in carbon and nitrogen allocation in alfalfa stems as well as different inhibitors of longitudinal and radial expansion. Plant Physiol. 137: 1474-1484.
Grevstad, F. S., D. R. Strong, D. Garcia-Rossi, R. W. Switzer, and M. S. Wecker. 2003. Biological control of Spartina alterniflora in Willapa Bay, Washington using the planthopper Prokelisia marginata: agent specificity and early results. Biol. Control 27: 32-42.
Grime, J.P. 2001. Plant Strategies, Vegetation Processes and Ecosystem Properties, 2nd edn. John Wiley & Sons Ltd, Chichester
178
Hager, H. A. and R. D. Vinebrooke. 2004. Positive relationships between invasive purple loosestrife (Lythrum salicaria) and plant species diversity and abundance in Minnesota wetlands. Can. J. Botany. 82:763-773.
Halbert, S. E. 2000. Florida Department of Agriculture and Consumer Services. Entomology section. Trilogy 39 (5). [Internet]. [cited 2008 February 4]; Available from: http://www.doacs.state.fl.us/pi/enpp/00-sep- oct.html#ent.
Harrigton, J. E. 1972. Notes on the biology of Ischnodemus species of America North of Mexico (Hemiptera: Lygaeidae: Blissinae). Occ. Pap. Univ. Conn., Biol. Sci. Ser.2, 6: 47-56.
He, X., Q. Wang, and A. Carpenter. 2003. Thermal requirements for the developmental and reproduction of Nysius hottuni White (Heteroptera: Lygaeidae). J. Econ. Entomol. 96: 1119-1125.
Heard, T. A. 2000. Concepts in host selection behavior and their application to host specificity testing, pp. 1- 10. In R. G. Van Driseche, T. A. Heard, A. McClay, and R. Reardon (eds), Host-Specificity Testing of Exotic Arthropod Biological Control Agents: The Biological Basis for Improvement in Safety. USDA Forest Health Technology Enterprise Team, FHTET-99-1, Morgantown, USA..
Heinrich, B. 1981. Ecological and evolutionary perspectives. Insect thermoregulation. Wiley, New York.
Hemerik, L., C. Busstra, and P. Mols. 2004. Predicting the temperature-dependwent natural population expansion of the western corn rootworm, Diabrotica virgifera. Entomol. Exp. Appl. 111: 59–69.
Henry, T.J. 1997. Phylogenetic analysis of family groups within the infraorder Pentatomorpha (Hemiptera: Heteroptera), with emphasis on the Lygaeoidea. Ann. Entomol. Soc. Am. 90:275-301.
Herrera, A. M. and T. L. Dudley. 2003. Reduction of riparian arthropod abundance and diversity as a consequence of giant reed (Arundo donax) invasion. Biological Invasions. 5:167-177.
Herrera, A. M., D.D. Dahlsten, N. Tomic-Carruthers, and R. I. Carruthers. 2005. Estimating temperature-dependent developmental rates of Diorhabda elongata (Coleoptera: Chrysomelidae), a biological control agent of Saltcedar (Tamarix spp.). Environ. Entomol. 34: 775-784.
Hijmans, R. J., Cruz, M., Rojas, E. and L. Guarino 2001 DIVA-GIS, version 1.4. A geographic information system for the management and analysis of genetic resources data. Manual. International Potato Center and International Plant Genetic Resources Institute, Lima, Peru
179
Holm, L. G., J. V. Pancho, J. P. Herberger, and D. L. Plucknett. 1979. A Geographical Atlas of World Weeds. Krieger Publishing Company, Florida.
Houston, W. A., and L. J. Duivenvoorden. 2002. Replacement of littoral native vegetation with the ponded pasture grass Hymenachne amplexicaulis: effects on plant, macroinvertebrate and fish biodiversity in the Fitzroy River, Central Queensland, Australia. Mar. Freshwater Res. 53:1235-1244.
Huberty, A. F. and R. F. Denno. 2004. Plant stress and its consequences for herbivorous insects: a new synthesis. Ecology 85: 1383-1398.
Irvin, N. A., and M. S. Hoddle. 2007. Evaluation of floral resources for enhancement of fitness of Gonatocerus ashmeadi, an egg parasitoid of the glassy-winged sharpshooter, Homalodisca vitripennis. Biol. Control. 40:80-88.
Johnson, S. R. and A. K. Knapp. 1996. Impact of Ischnodemus falicus (Hemiptera: Lygaeidae) on photosynthesis and production of Spartina pectinata wetlands. Environ. Entomol. 25: 1122-1127.
Joshi, J. and K. Vrieling. 2005. The enemy release and EICA hypothesis revisited: incorporating the fundamental difference between specialist and generalist herbivores. Ecol. Lett. 8: 704–714
Julien, M. H., and M. W. Griffiths. 1998. Biological Control of Weeds. A World Catalogue of Agents and their Target Weeds. Fourth Edition. CAB International, Wallingford, UK.
Kalmbacher, R., J. Mullahey, and K. Hill. 1998. Limpograss and hymenachne grown on flatwoods range pond margins. J. Range Manage. 51: 282–287.
Karley, A. J., A. E. Douglas, and W. E. Parker. 2002. Amino acid composition and nutritional quality of potato leaf phloem sap for aphids. J. Exp. Biol. 205: 3009-3018.
Keane, R. M., and Crawley M. J. 2002. Exotic plant invasions and the enemy release hypothesis. Trends Ecol. Evol. 17: 164–170.
Kibbler, H., and L. M. Bahnisch. 1999a. Physiological adaptations of Hymenachne amplexicaulis to flooding. Aus. J. Exp. Agr. 39: 429-435.
Kibbler, H., and L. M. Bahnisch. 1999b. Distribution of Hymenachne acutigluma (Steudel) Guilliland in ponded pasture is limited by photosynthetic response to temperature. Aus. J. Exp. Agr. 39: 437-443.
180
Konica-Minolta 2008. Minolta Corporation. Chlorophyll meter SPAD-502. User Manual. New Jersey. [Internet]. [cited 2008 April 15];. Available from: http://www.konicaminolta.com/instruments/products/color/chlorophyll- meter/spad502/index.html
Krueger, S. R. Nechols, J. R. and W. A. Ramoska. 1992. Habitat distribution and infection rates of the fungal pathogen, Beauveria bassiana (Balsamo) Vuillemin in endemic populations of the chinch bug Blissus leucopterus (Say) (Hemiptera: Lygaeidae) in Kansas. J. Kansas Entomol. Soc. 65: 115-124.
Kushlan, J. A. 1990. Freshwater marshes. Pages 324-363 In R. L. Myers and J. J. Ewel (eds.) Ecosystems of Florida. University of Central Florida Press, Orlando.
Larson, K. C., and T. G. Whitham. 1991. Manipulation of food resources by a gall-forming aphid: the physiology of sink-source interactions. Oecologia 88: 15-21.
Lockwood, J., and R. Story. 1986. Adaptive functions of nymphal aggregation in the southern stink bug, Nezara viridula (L.) (Hemiptera: Pentamoidae). Environ. Entomol. 15: 739- 749.
Lodge DM, Williams S, MacIsaac HJ, Hayes KR, Leung B, et al. 2006. Biological invasions: Recommendations for U.S. policy and management. Ecol. Appl.: 16: 2035–2054.
Louda, S. M., A. E. Arnett, T. A. Rand, F. L. Russell. 2003. Invasiveness of some biological control insects and adequacy of their ecological risk assessment and regulation. Conserv. Biol. 17: 73-82.
Lukacs, G. P. 1996. A review of the environmental effects of ponded pasture systems. Report No.93/09, Australian Centre for Tropical Freshwater Research. James Cook University of North Queensland, Townsville.
Lyons, J. 1996. Developing ponded pastures at “Wambiana”, Charters Towers. In P. A. Pittaway, J. H. Wildin, and C. K. McDonald (eds.), Beef Production from Ponded Pastures. Proceedings of a workshop held at Yeppoon, Queensland, April 1993. Tropical Grassland Occasional Publication No. 7. Tropical Grassland Society of Australia Inc., St. Lucia, Queensland.
Massey, F. P., A. R. Ennos, and S. E. Hartley. 2006. Silica in grasses as a defence against insect herbivores: contrasting effects on folivores and a phloem feeder. J. Anim. Ecol. 75: 595–603.
Mazzei, K. C., R.M. Newman, A. Loos, and D. W. Ragsdale. 1999. Developmental rates of the native milfoil weevil, Euhrychiopsis lecontei and damage to Eurasian watermilfoil at constant temperatures. Biol. Control. 16: 139-143.
181
Medina, E. and N. Motta. 1990. Metabolism and distribution of grasses in the tropical flooded savannas in Venezuela. J. Trop. Ecol. 6: 77-89.
Meyerson, L. A. and H. A. Mooney. 2007. Invasive alien species in an era of globalization. Front. Ecol. Environ. 5: 199-208.
McColloch, J. W., and H. Yuasa. 1914. A parasite of the chinch bug. J. Econ. Entomol. 7: 219- 227.
Miles, P. W. 1968. Insect secretions in plants. Ann. Rev. Phytopathol. 6: 137–164.
Myers, J. H. and D. R. Bazely. 2003. Ecology and control of introduced plants. Cambridge University Press. United Kingdom.
National Invasive Species Council. 2001. Meeting the Invasive Species Challenge: National Invasive Species Management Plan. Washington, D. C.: US Government, 89 pp.
Negron, J. F., and T. J. Riley. 1990. Long-term effects of chinch bug (Hemiptera: Heteroptera: Lygaeidae) feeding on corn. J. Econ. Entomol. 83: 618-620.
New York Botanical Garden. 2008. International Plant Science Center. The C. V. Starr Virtual Herbarium. New York, New York. [Internet]. [cited 2008 March 20]; Available from: http://sciweb.nybg.org/science2/vii2.asp
Overholt, W. A., S. M. L. Ewe, R. Diaz, E. C. Morgan, and O. E. Moeri. 2004. Feeding effects of Ischnodemus variegatus (Hemiptera: Blissidae) on photosynthesis and growth of Hymenachne amplexicaulis (Poaceae). Fla. Entomol. 87: 312–316.
Painter, R. H. 1928. Notes on the injury to plant cells by chinch bug feeding. Ann. Entomol. Soc. Am. 21: 232-242.
Paynter, Q., N. Martin, J. Berry, S. Hona, P. Peterson, A. H. Gourlay, J. Wilson-Davey, L. Smith, C. Winks, and S. V. Fowler. 2008. Non-target impacts of Phytomyza vitalbae a biological control agent of the European weed Clematis vitalba in New Zealand. Biol. Control 44: 248-258.
Pemberton, R. W. 2000a. Predictable risk to native plants in weed biological control. Oecologia 125: 489-494.
Pemberton, R. W. 2000b. Naturalization pattern of horticultural plants in Florida. In: Prooceedings of the X International Symposium on Biological Control of Weeds. Ed: Spencer, N. R. Bozeman, MT: Montana State University. pp 881.
182
Pemberton RW. 2002. Selection of appropriate future target weeds for biological control. In R. Van Driesche, S. Lyon, B. Blossey, M. Hoddle, and R. Reardon (eds), Biological Control of Invasive Plant in Eastern United States. USDA Forest Service Publication FHTET- 2002-04. [cited 2008 January 01]. (http://www.invasive.org/eastern/biocontrol/30FutureTargets.html).
Perry, J. N., and I. P. Woiwood. 1992. Fitting Taylor’s power law. Oikos 65: 538-542.
Pilkington, L. J., and M. S. Hoddle. 2006. Use of life table statistics and degree-day values to predict the invasion success of Gonatocerus ashmeadi (Hymenoptera: Mymaridae), an egg parasitoid of Homoladisca coagulate (Hemiptera: Cicadellidae), in California. Biol. Control. 37: 276-283.
Pimentel, D. Zuniga R. and D. Morrison. 2005. Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecol. Econ. 52: 273-288.
Pinder, L. and S. Ross. 1998. Classification and ordination of plant formations in the Pantanal of Brazil. Plant Ecol.136: 151-165.
Ponsonby, D. J., and M. J. W. Copland. 1996. Effect of temperature on development and immature survival in the scale insect predator, Chilocorus nigritus (F.) (Coleoptera: Coccinellidae). Biocontrol Sci. Techn. 6: 101-109.
Posey, M. H. 1988. Community changes associated with the spread of an introduced seagrass, Zostera japonica. Ecology. 69: 974-983.
Prado, E., and W. F. Tjallingii. 1997. Effects of previous plant infestation on sieve element acceptance by two aphids. Entomol. Exp. Appl. 82: 189-200.
Reinert, J. A. 1978. Natural enemy complex of the southern chinch bug in Florida. Ann. Entomol. Soc. Am. 71: 728- 731.
Sandström, J. M. and J. Pettersson. 1994. Amino acid composition of phloem sap and the relation to intraspecific variation in pea aphid (Acyrthosiphon pisum) performance. J. Insect Physiol. 40: 947 -955.
SAS Institute. 1999. SAS/STAT User’s guide. SAS Institute, Cary, NC.
Sandstrom, J., Telang, A. and N. A. Moran. 2000. Nutritional enhancement of host plants by aphids: a comparison of three aphid species on grasses. J. Insect Physiol. 46:33-40.
Scott, J. K., and P. B. Yeoh. 1999. Bionomics and the predicted distribution of the aphid Brachycaudus rumexicolens (Hemiptera: Aphididae). B. Entomol. Res. 89: 97-106.
183 Schooler, S. S., E. M. Coombs, and P. B. McEvoy. 2003. Nontarget effects on cripple myrtle by Galerucella pusilla and G. calmariensis (Chyrsomelidae), used for biological control of purple loosestrife (Lythrum salicaria). Weed Sci. 51: 449-455.
Schoonhoven, L. M., J. J. A. van Loon, and M. Dicke. 2005. Insect-plant biology. Oxford Univ. Press., Oxford, UK
Scriber, J. M. and F. Slansky Jr. 1981. The nutritional ecology of immature insects. Ann. Rev. Entomol. 26:182-211.
Slater, J. A. 1976. Monocots and chinch bugs: A study of host plant relationships in the lygaeid subfamily Blissinae (Hempitera: Lygaeidae). Biotropica. 8: 143-165.
Slater, J. A. 1987. The taxonomic status of Ischnodemus oblongus (Fabricius) and Ischnodemus variegatus (Signoret) (Hemiptera: Lygaeidae: Blissinae). J. New. York Entomol. Soc. 95: 294-297.
Slater, J. A. and D. B. Wilcox. 1969. A revision of the genus Ischnodemus in the neotropical region (Hemiptera: Lygaeidae: Blissinae). Misc. Publ. Entomol. Soc. Amer. 6: 199-238.
Slater, J. A., and D. B. Wilcox. 1973. The chinch bugs or Blissinae of South Africa (Hemiptera: Lygaeidae). Memoirs of the Entomological Society of Southern Africa 12: 1-135.
Scriber, J. M., and F. Slansky Jr. 1981. The nutritional ecology of immature insect. Annu. Rev. Entomol. 26: 183-211.
Slater. J. A., and R. M. Baranowski. 1990. Lygaeidae of Florida. (Hemiptera: Heteroptera). Vol 14. Arthropods of Florida and neighboring land areas. Florida Department of Agriculture and Consumer Services. Gainesville, Florida.
Soares, D. J., R. W. Barreto. 2006. Additions to the Brazilian mycobiota of the grassy weed, Hymenachne amplexicaulis, with a discussion on the taxonomic status of Paraphaeosphaeria recurvifoliae. Australas. Plant Path. 35: 347–353.
Sodoyama, Y. 2007. Effects of temperature on the development of Eumicrosoma blissae (Hymenoptera: Scelionidae), an egg parasitoid of the oriental chinch bug, Cavelerius saccharivorus Okajima (Heteroptera: Lygaeidae). Appl. Entomol. Zool. 42: 613–617
Southwood, T. R. E. 1978. Ecological Methods. Chapman and Hall, London.
Strong, D. R., Lawton J. H., and T. R. E. Southwood. 1984. Insects on plants. Community patterns and mechanisms. Blackwell Scientific Oxford, UK.
184 Sweet, M. H. 2000. Seed and Chinch Bugs (Lygaeoidea), pp. 143–264. In C.W. Schaefer and A.R. Panizzi (eds.), Heteroptera of Economic Importance, CRC Press, Boca Raton, London, New York, Washington, DC.
Talley, T. S. and L. A. Levin. 2001. Modification of sediments and macrofauna by an invasive marsh plant. Biol. Invasions. 3: 51-68.
Talley, T. S., and L. A. Levin. 2001. Modification of sediments and macrofauna by an invasive marsh plant. Biol. Invasions 3: 51-68.
Taylor, L. R. 1961. Aggregation, variance and the mean. Nature 189: 732-735.
Taylor, L. R., J. N. Perry, I. P. Woiwood, and R. A. J. Taylor. 1988. Specificity of the spatial power-law exponent in ecology and agriculture. Nature 332: 721-722.
Tejos, M. R. 1978. Effect of age on the productivity of paja de agua grass (Hymenachne amplexicaulis (Rudge) Nees) in a controlled flooding savannah. Agron. Trop. 28: 613- 626.
Thorp, J. R., and R. Lynch. 2000. The Determination of Weeds of National Significance (National Weeds Strategy Executive Committee, Launceston). Australia.
Tropical Weeds Research Centre. 2007. Hymenachne (Hymenachne amplexicaulis) control and management in Queensland. Natural Resources and Water. Queensland Government. Queensland, Australia. [Internet]. [cited 2007 September 27]; Available from: http://www.nrw.qld.gov.au/tropical_weeds/projects/hymenachnecontrol.html
Tscharntke, T., and H. J. Greiler. 1995. Insect communities, grasses, and grasslands. Ann. Rev. Entomol. 40: 535-558.
USDA. 2000. United States Department of Agriculture (USDA), Animal and Plant Inspection Service (APHIS), Plant Protection and Quarantine (PPQ). Reviewer’s manual for the technical advisory group of biological control agents of weeds: Guidelines for evaluating the safety of candidate biological control agents. p 224.
University of Florida, Herbarium, 2007. Collections catalogue. Gainesville, Florida. [Internet]. [cited 2007 February 4]; Available from: http://www.flmnh.ufl.edu/scripts/dbs/herbs_project/herbsproject/herbs_pub_proc.asp van Klinken, R. D. 2000. Host specificity testing: why we do it and how can we do it better?, pp. 54-68. In R. van Driesche, T. Heard, A. McClay and R. Reardon (eds.), Proceedings: Host Specificity of Exotic Arthropod Biological Control Agents: the Biological Basis for Improvement in Safety, Forest Service, Morgantown, West Virginia. van Klinken, R. D., and O. R. Edwards. 2002. Is host specificity of weed biological control agents likely to evolve once released? Ecol. Lett. 5: 590-596.
185 van Loan, A. N., J. R. Meeker, and M. C. Minno. 2002. Congon Grass. In R. Van Driesche, S. Lyon, B. Blossey, M. Hoddle, R. Reardon (eds.), Biological Control of Invasive Plant in Eastern United States. USDA Forest Service Publication FHTET-2002-04. [cited 2007 September 22]. Available from: http://www.invasive.org/eastern/biocontrol/28CogonGrass.html
Vitousek, P. M., Dantonio, C. M., Loope L. L. and R. Westbrooks. 1996. Biological invasions as global environmental change. American Scientist 84: 468-478.
Wapshere, A. J. 1989. A testing sequence for reducing rejection of potential biological control agents for weeds. Ann. Appl. Biol. 114: 515-526
Watson, L., and M. J. Dallwitz. 1992. Grass Genera of the World: Descriptions, Illustrations, Identification, and Information Retrieval; including Synonyms, Morphology, Anatomy, Physiology, Phytochemistry, Cytology, Classification, Pathogens, World and Local Distribution, and References. Version: 18th August 1999. [Internet]. [cited 2008 February 4]; Available from: http://biodiversity.uno.edu/delta/
Wheeler, A. G. 1996. Ischnodemus falicus (Heteroptera: Lygaeidae): first record from ornamental grasses, and seasonality on prairie cordgrass in Pennsylvania. Proc. Entomol. Soc. Wash. 98: 195-198.
Wheeler, G. S. and M. D. Halpern. 1999. Compensatory responses of Samea multiplicalis larvae when fed leaves of different fertilization levels of the aquatic weed Pistia stratiotes. Entomologia Experimentalis et Applicata. 92: 205-216.
Wheeler, G.S. and T.D. Center. 2001. Impact of the biological control agent Hydrellia pakistanae (Diptera: Ephydridae) on the submersed aquatic weed Hydrilla verticillata (Hydrocharitaceae). Biol. Control 21: 168–181
Wilson, L. T. and P. M. Room. 1983. Clumping patters of fruit and arthropods in cotton with implication to binomial sampling. Environ. Entomol. 12: 50-54.
Wolfe, L. M. 2002. Why alien invaders succeed: support fo the escape-from-enemy hypothesis. Am. Nat. 160: 705-711.
Worner, S. P. 1992. Performance of phenological models under variable temperature regimes: Consequences of the Kaufmann or rate summation effect. Environ. Entomol. 21:689-699.
Wunderlin, R. P., and B. F. Hansen. 2008. Atlas of Florida Vascular Plants. [S. M. Landry and K. N. Campbell (application development), Florida Center for Community Design and Research.] Institute for Systematic Botany, University of South Florida, Tampa. [Internet]. [cited 2008 February 4]; Available from: http://www.plantatlas.usf.edu/
Zar, J. H. 1999. Biostatistical analysis. 4th ed. Prentice Hall. Inglewood Cliffs, NJ.
186
BIOGRAPHICAL SKETCH
Rodrigo was born in Quito, Ecuador in 1977. Upon graduation from high school he moved to Honduras to study Agronomy at the Escuela Agricola Panamericana, ‘El Zamorano’. During his undergrad studies he specialized on crop protection and carried out a thesis on biological control of corn earworm using commercially available parasitoids. He then moved in 2000 to the
University of Illinois at Urbana-Champaign to start a 1-year internship in the weed biological control laboratory. In 2001, he started the master’s program in the Department of Entomology at
Texas A&M University in College Station. He studied the role of red imported fire ants as biological control agents of cotton pests. Then, he enrolled in the PhD program in the
Department of Entomology and Nematology at the University of Florida. During his PhD research he studied the biology, host specificity and impacts of a sap-feeding bug on an exotic grass in Florida. He is a current member of the Entomological Society of America, Florida
Entomological Society and the International Organization for Biological Control. Rodrigo’s plans for the future include continuing in biological control as research scientist, strengthening international cooperation on classical biological control and training young scientists from developing countries on invasive species management.
187