1 BIOLOGY, POPULATION GROWTH, AND ECOLOGICAL NICHE MOD ELING OF Calophya terebinthifolii (HEMIP TERA: CALOPHYIDAE ), A CANDIDATE FOR BIOLOGICAL CONTROL O F BRAZILIAN PEPPERTR EE, Schinus terebinthifolius ( SAPINDALES: ANACARDIACEAE) By LINDSEY RACHEL CHRIST A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010
2 2010 Lindsey R. Christ
3 To Bob and my parents for their su pport, love and patience
4 ACKNOWLEDGMENTS I would like to thank Dr. Jim Cuda, Dr. Bill Overholt and the School for Natural Resources and Environment for this opportunity. I would also like to thank Dr. Doria Gordon for serving on my committee. In the lab q uarantine, and the greenhouse, I routinely had assistance from Judy Gil lmore, Daniel O kine, and Susan Wright. Abhishek Mukherjee provided assist ance with the modeling and was a great lab mate. I could not have accomplished my research in Brazil without the help of Dr. Marcelo Vitorino, Maria F. Pollnow, Liliam Beal, Andr Buss, and Tatiana Reichert Drs. Veronica Manrique, Rodrigo Diaz, Julio Me dal, and Greg Wheeler assisted my research by bring back psyllids from Brazil. I would also like to thank Dr. Daniel Burckhardt for his assistance in psyllid identification and Dr. Dean Williams for DNA analysis.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURE S ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Litera ture Review ................................ ................................ ................................ .... 14 Classical Biological C ontrol ................................ ................................ .............. 14 Schinus terebinthifolius ................................ ................................ ..................... 20 Taxo nomy ................................ ................................ ................................ .. 20 Morphology ................................ ................................ ................................ 20 Distribution ................................ ................................ ................................ 21 Invasive properties ................................ ................................ ..................... 22 Control methods ................................ ................................ ......................... 23 Calophya terebinthifolii ................................ ................................ ..................... 24 Biology of Gall Formation ................................ ................................ ................. 26 Project Justification ................................ ................................ ................................ 28 Object ives ................................ ................................ ................................ ............... 30 2 COLONY ESTABLISHMENT OF C. TEREBINTHIFOLII ................................ ........ 31 Materials and Methods ................................ ................................ ............................ 31 Results and Discussion ................................ ................................ ........................... 35 3 BIOL OGY OF C. TEREBINTHIFOLII ................................ ................................ ...... 40 Material and Methods ................................ ................................ ............................. 40 A) Location of Pit Gall Formation on Plants in the F ield ................................ ... 41 B) Female F ecundity ................................ ................................ ........................ 41 C) Verification of the N umber of Instars and S izes ................................ ........... 42 D) Development Time and S urvivorship ................................ ........................... 42 Statistical A nalysis ................................ ................................ ............................ 43 Results and Discussion ................................ ................................ ........................... 44 A) Location of Pit Gall F ormati on on Plants in the F ield ................................ ... 44 B) Female F ecundity ................................ ................................ ........................ 45 C) Verification of the Number of Instars and S izes ................................ ........... 47 D) Development Time and S urvivorship ................................ ........................... 51
6 4 BRAZILIAN PEPPERTREE HAPLOTYPE SUITABILIT Y ................................ ....... 55 Materials and Methods ................................ ................................ ............................ 58 Result s and Discussion ................................ ................................ ........................... 58 5 ECOLOGICAL NICHE MOD ELING ................................ ................................ ........ 60 Materials and Methods ................................ ................................ ............................ 61 Results and Discussion ................................ ................................ ........................... 63 Niche Modeling ................................ ................................ ................................ 63 Calophya terebinthifolii Niche P rediction in Florida ................................ .......... 66 Future Survey S ites in South America for C. terebinthifolii ............................... 67 6 BRAZILIAN PEPPERTREE MANAGEMENT USING C. TEREBINTHIFOLII .......... 71 Considerations ................................ ................................ ................................ ........ 71 Recommendations ................................ ................................ ................................ .. 73 7 CONCLUSION ................................ ................................ ................................ ........ 75 APPENDIX A FIELD COLLECTION SIT ES AND DESCRIPTIONS ................................ .............. 78 B FIELD COLLECTION DAT ES AND WEATHER PARAM ETERS ............................ 82 C GPS COORDINATES FOR C. TEREBINTHIFOLII IN SOUTH AMERICA ............. 83 D GPS COORDINATES BRAZ ILIAN PEPPERTREE HAP LOTYPE A IN FLORIDA .. 85 E PSYLLID DNA SEQUENCI NG ................................ ................................ ............... 86 F IMAGES OF PARASITOIDS ................................ ................................ ................... 87 G PSYLLID SHIPMENTS ................................ ................................ ........................... 89 LIST OF REFERENCES ................................ ................................ ............................... 90 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 102
7 LIST OF TABLES Table page 3 1 Body size of all life stages of C. terebinthifolii ................................ ..................... 50 3 2 Life and fertility table ................................ ................................ ........................... 53 4 1 C alophya terebinthifolii success on Brazilian peppertree haplotypes ................. 58 5 1 Minimum training threshold dependent binomial test (Florida) ........................... 65 5 2 Minimum training threshold dependent binomial test (South America) ............... 6 5 A 1 Field collection sites and descriptions ................................ ................................ 78 B 1 Field collection dates and weather parameters ................................ .................. 82 C 1 GPS coordinates for C. terebinthifolii in South America ................................ ..... 83 D 1 GPS coordinates Brazilian peppertree haplotype A in Florida ............................ 85 G 1 C alophya terebinthifolii shipments ................................ ................................ ...... 89
8 LIST OF FIGURES Figure page 1 1 How classical biological control works ................................ ................................ 15 1 2 Photographs of Brazilian peppertree in Brazil ................................ ..................... 21 2 1 Collection sites of C. terebinthifolii ................................ ................................ ..... 32 2 2 Photographs of Brazilian peppertree s eedlings ................................ .................. 34 2 3 Calophya sp cf terebinthifolii images ................................ ................................ .. 39 3 1 LAMPF location and image of pit galls ................................ ............................... 45 3 2 C alopha terebinthifolii eggs and first instar images ................................ ............ 46 3 3 Wing length and egg number regression ................................ ............................ 47 3 4 Frequency distribution of larval lenghts ................................ .............................. 48 3 5 C alophya terebinthifolii images ................................ ................................ ........... 49 3 6 Size comparisons between C. terebinthifolii and C. schini ................................ 51 3 7 Survivorship curve ................................ ................................ .............................. 52 3 8 Rate of development ................................ ................................ .......................... 54 4 1 cpDNA haplotype network of Brazilian peppertree ................................ ............. 57 5 1 Jackknife test in MaxEnt ................................ ................................ ..................... 64 5 2 Climatic suitability prediction (Florida) ................................ ................................ 67 5 3 Climatic suitability prediction (South America) ................................ .................... 69 F 1 Images of parasitoids ................................ ................................ ......................... 87
9 LIST OF ABBREVIATIONS FBCL Florida Biological Control Laboratory GIS Geographic Information System IPM Integrated Pest Management LAMPF Laboratrio de Monitoramento e Proteo Florestal SEM Standard Error Mean
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science BIOLOGY, POPULATION GROWTH, AND ECOLOGICAL NICHE MOD ELING OF Calophya terebinthifolii (HEMIP TERA: CALOPHYIDAE ), A CANDIDATE FOR BIOLOGICAL CONTROL O F BRAZILIAN PEPPERTR EE, Schinus terebinthifolius ( SAPINDALES: ANACARDIACEAE) By Lindsey Rachel Christ August 2010 Chair: James P. Cuda Cochair: William A. Overholt Major: Interdisciplinary Ecology Brazilian peppertree, Schinus terebinthifolius Raddi (Anacardiaceae), a perennial woody plant native to Brazil, Argentina, and Paraguay, has become one of the most invasive weeds in Florida. A leaflet pit galling psyllid, Calophya terebinthifolii Burckhardt & Basset (Hemiptera: Calophyidae) has been identified as a potential biological control agent for Brazilian peppertree. However, b iological information on the psyllid including its life history and rearing procedures, is lacking. This type of information is essential when importing an insect f or biological control purposes. F rom May August 2009 field and laboratory research was conducted at the Laboratrio de Monitoramento e Proteo Florestal (LAMPF) in Gaspar, Santa Catarina, Brazil with psyllids collected from the Atlantic coastal region o f Santa Catarina Rearing was successful in Brazil but a colony was not successfully reared in Florida. The results of f ield studies in Brazil showed that the open pit galls produced by the developing nymphs were locat ed on the adaxial (upper) side of th e lea ves (2.6 1.8 galls/leaflet s range of 0 34). Laboratory studies on the psyllid in Brazil focused on: female fecundity (55.3 8.9 eggs/female)
11 the number and size of the immature stages age specific survivorship, life table construction and mea n generation time (43.7 1.2 days). Preliminary evidence from feeding trials suggest s the psyllids are locally adapted to specific genotypes of Brazilian peppertree. The psyllids from the Atlantic coastal region of Santa Catarina appear to be locally ad apted to Brazilian peppertree haplotype A plants, which occur in Florida. Ecological niche modeling was done in MaxEnt to determine if there was climatic overlap between Florida and the native range of the psyllid in South America. Using collection and survey locations of the psyllids in their native range and point locations for haplotype A plants in Florida, prediction maps were created. The climatic overlap found in Florida includes Volusia, coastal Pasco and Hernando, and a small section of southwes tern Polk counties, which could be used as target locations for release if the psyllid is approved as a biocontrol agent. The map generated for South America could be used to find future survey sites for collecting psyllids. The results of this research suggest C. terebinthifolii may not be an effective biological control agent for Brazilian peppertree. However, issues surrounding rearing and Brazilian peppertree genotypes need to be resolved before being eliminated as a candidate for biocontrol of Brazi lian peppertree.
12 CHAPTER 1 INTRODUCTION The introduction of exotic species in the United States is not a new phenomenon. Exotic species have been imported into the US for hundreds of years intentionally and unintentionally (Schmitz 2007) Not all exot ics become invasive even if they become naturalized in the environment. Some research has suggested there are positive environmental effects from exotics in natural areas For instance, exotic plants increase plant diversity on islands (Sax & Gaines 2008 ) or in som e situations where native species are unavailable can be used for habitat restoration (Ewel & Putz 2004) There are more than 50,000 exotic species in the United States (Pimentel et al. 2005). While a blanket condemnation of the importation of exotic species is un realistic, exotics often become invasive and are considered to be one of the leading causes of species extinction and endangerment. Invasive exotic plants and animals are the second largest threat to biodiversity in the United States behind habitat destruction, with nearl y half of all imperiled species ne gatively impacted by exotics (Wilcove et al. 1998) Individual invasive species have the potential to modify ecosystem properties by shifting resource availability, stand structure and composition, and changing disturbance regimes (Gordon 1998). Control of exotic invasive species results in a large economi c investment of approximately $120 billion/year which is a conservative estimate since some negative consequences are hard to quantify (Pimentel et al. 2005) Florida is particularly susceptible to exotic plant invasions because of its unique geography. The state is surrounded on three sides by water and the fourth side by a freeze line giving it the geography of an island ( Myers & Ewel 1990 ). With its subtropical climate, diverse and disturbed habitats, as well as the numerous lakes,
13 streams, and rive rs linking the various areas of the state, and an international port in Miami, numerous non native plants have become naturalized ( Simberloff et al. 1997 ). In Florida, 25,000 exotic plants species are imported annual ly and more than 900 have escaped to be come naturalized in the surrounding ecosystems (Pimentel et al. 2005, Frank & McCoy 1995, Frank et al. 19 97 ; Simberloff et al. 1997 ). Florida, along with California and Hawaii, is one of the top three states most affected by nonindigen ous species (Center et al. 1997 are nonindigenous (Frank & McCoy 1992). However, not all naturalized plants threaten the states ecology or econom y. P lant s that do become problematic often require substantial r esourc es to control them. From 1980 2006 over $250 million has been spent in Florida by federal, state, and local agencies to control invasive non native species (Schmitz 2007) Classical biological control is one option for controlling undesirable, weedy pl ant species and can be used as part of an Integrated Pest Management (IPM) plan (Cuda et al. 2006). The use of a biological control agent can be an effective and economical way of regulating weedy plant populations as an alternative to o r in conjunction, with chemical, physical, and mechanical control methods. My research focus ed on a leaflet galling psyllid, Calophya terebinthifolii Burckhardt & Basset (Hemip tera: Calophyidae ) a natural enemy of Brazilian peppertree, Schinus terebinthifolius Raddi ( Sapindales: Anacardiaceae). One of the subject of a biological control program since the 1980 s (Cuda et al. 2006).
14 Literature Review Classical biological c ontrol The first use of the term biol ogical control is credited to Smith (1919) and is defined as the use of natural enemies to control an undesired species or pest. Biological control falls u nder the discipline of applied e cology but also relies on ethology, taxonomy, physiology, genetics, biochemistry, and other disciplines (Cock 1986; Center et al. 1997 ) The most successful biological control programs are based upon detailed studies of the pest and its ecology (Cock 1986). Classical biological control is the applied phase of discovery, importation, and establishment range to exert pressure on the invasive species ( DeBach & Rosen 1991 ). Eradication is never the goal of any biological cont rol program (Center et al. 1997 ) Instead, the goals are to reduce the target organism to a leve l that allows an increase in indigenous biodiversity, the restoration of ecosystem processes, or improvement in the human economy (McEvoy & Coombs 1999) One theory of why invasive plants do so well is based on the enemy release hypothesis (Williams 1954 ; Wolfe 2002 ; Liu & Stiling 2006 ) which states that invasive non native plants spread rapidly because they are liberated from their co evolved natural enem ies. The release from natural enemies allows the exotic plant to outcompete other native species Classical biological control is successful because it reunites the host plant with one or more of its coevolved natural enemies. W hen natural enemies are r eleased a competitive advantage of the invasive plant has been removed ( Liu & Stiling 2006 ) One study recently examined the amount of herbivory on native and exotic plants and found significantly less herbivory on highly invasive plants than
15 natives len ding credibility to the enemy release hypothesis (Carpenter & Cappuccino 2005). After the population of released natural enemies increases to a point where it can negatively affect the host plant population, the host plant will decline to low population levels (Figure 1 1) until the bio control agent has difficulty findi ng the host (Center et al. 1997 ). The bio control agent population is reduced more than the population of the host plants at which point a low level steady state results. The biocontrol ag ents are usually more effective a t preventing outbreaks of pests than reducing outbreak populations (Murdoch et al. 1985). Figure 1 1. Graph adapted from the Commonwealth Scientific and Industrial Research Organisation (CSIRO) in Australia demonstratin g theoretically how classical biological control works to suppress an inv asive weed in the environment to a low level steady state ( http://www.csiro.au/org/WeedBiocontrol.html ) The process for conducting biological control research is fairly complex and involves extensive testing of potential host plants and navigating a complex permitting
16 process dealing with multiple levels of government regulation. The process involves the following steps (M essing & Wright 2006 ; Center et al. 1997 ): 1. Select ing the target by conducting an economic and ecological cost benefit analysis 2. (requires the identification o f the native range and surveys for natural enemies) 3. potential to inflict substantial stress on the invasive species 4. Importation and quarantine studies of the biocontrol agent 5. Extensive host range testing (= risk assessment) 6. Obtain ing federal and state permits to release the biocontrol agent 7. Releasing the biocontrol agent in the new environment 8. Monitoring establishment, dispersal, and efficacy 9. Incorporating the biological control agent into existing and new management plans for the invasive species For biological control projects to be successful it is important to understand the potential problems and shortcomings involved in the biocontrol agent evaluation process and be prepared to overcome those challenges. Knowing the right questions to ask in the beginning is important. During the selection process, the following questions need to be considered: will the agent reach required densities, what and how much damage will the agent cause to provide effective control, and will the agent be evaluated using molecular techniques, biogeography, and/or quantitative surveying (Julien et al. 2007).
17 The selection of the biological control agents is vital in ensuring the success of a biological control project. Criteria used during the selection process sho uld include: 1 ) obtaining natural enemies from climatic regions similar to co ntrol of the pest is desired, 2 ) choosing natural enemies that have a higher rate of increase relative to the target pest, and 3 ) have a high degree of host specificity (Hokkanen & Pimentel 1984). Understanding the biology of the weed and the potential agent are critical i n answering whether the agent breeds continuously, has seasonal variations, has multiple genera tions per year, enters diapause or uses alternate hosts (Cock 1986 ). Knowing the agent s biology will also help when trying to incorporate it into an IPM program (Cock 1986). Another area to consider durin g the selection process is the effect of multi species interactions and whether phylo geny can predict host range Increasingly, molecular techniques are able to determine weed origins and identify specific geographic locations for bioco ntrol agent surveys, as well as distinguish species and subspecies of plants and insects ( Sheppard et al. 2005 ; Julien et al. 2007 ; M ound et al. 2010 ). Ecophysiological tolerances (e.g. nutrient requirements moisture, humidity, temperature) need to be determined to prevent a mismatch of introduced biotypes resulting i n poor weed control (Geiger & Gutierrez 2000 ; Manrique et al. 2008 ). Finally, it is advisable that systematic post release evaluations a re conducted using ecological modeli ng and economic evaluations that clearly demonstrate ecological and economic gains (Julien et al. 2007). There are a number of benefits for using classical biological control for restoration in natural areas. It is economically cheaper to release a successful c ontrol
18 agent than spend years controlling species with mechanical, chemical, or physical removal (Cuda et al. 2006 ; Schmitz 2007 ; McEvoy & C oombs 1999) Biological control is an important tool for providing practical and affordable solutions for weed control either on its own or in combination with additional management strategies. It is especially useful for weedy species growing in sensiti ve, conservation, and riparian areas where herbicide and physical controls are undesirable, prohibited or ineffectual (Julien et al. 2007). Natural enemies are self sustaining, self dispersing, and generally adjust their population size in relation to tha t of the target pests (Messing & Wright 2006) If the research and testing before release is done corre ctly, biocontrol agents rarely a ffect non target species ( Pemberton 2000 ; Louda & Stiling 2004) Biocontrol agents are highly regulated and must go thr ough extensive testing and safe guards before a release is granted (Louda & Stiling 2004). Overall the results of intentional releases of biological agents worldwide have been favorable. According to Julien et al. (1984) of 174 worldwide biological cont rol projects, 39% were successful. That same study found of the 101 species of weeds targeted for control, 48% were controlled by the biological agents. Of the 178 species released as biological control agents, 71% established and 28% contributed to succ essful weed control. Biological control is not without its problems. The early history of biological control includes some introductions that le d to negative consequences (Howarth 1991 ; Simberloff & Stiling 1996 ). Some conservationists are against the i ncrease in biological pollution caused by introducing another non native species to an already degraded ecosystem (Messing & Wright 2006) Once the species is released, it is nearly impossible to reverse the process (but see Moeri et al. 2009 ) ( Howarth 1983 ; Louda &
19 Stilling 2004) Also, i t is difficult to predict the true effect the species will have in the field from lab experiments and unintended consequences can arise (Callaway et al. 1999 ; McEvoy & Coombs 1999 ; Louda & ; Louda & Stillin g 2004 ; Pearson & Callaway 2006 ) For instance a number of insects have been released against s potted knapweed Centaurea maculosa Lam., but when they fail ed to control it the small amount of herbivory actually caused an increase in plant growth (Callaw ay et al. 1999) Sometimes, when multiple agents are released competition between the agents can interfere with the effectiveness of their control ( McEvoy & Coombs 1999 ) Another indirect effect not foreseen prior to the release of two tephritid flies ( Urop h ora affinis Frauenfeld and Uroph ora quadrifa s ciata Meigen ) on spotted knapweed was the added food resource for deer mice ( Peromyscus maniculatus ) of the fly larvae. An increase in the deer mice population led to a decline in the population of other small mammals (Pearson et al. 2000) and an increase in Hantavirus (Pearson & Callaway 2006) Risk management is another important aspect of classical biological control research projects. The potential negative impacts on native plants, invertebrates a nd other species needs to assessed before releases occur. Worldwide, governments are adopting attitudes with lower tolerances for acceptable risk with increasing restrictions on importation and release, thus necessitating the importance of sound and effec tive research techniques (Julien et al. 2007). Risks can include economic spillover (destruction to important economic plants used for agricult ure), environmental spillover (e ffects on non target species), and general uncertainty (Tisdell et al. 1984). C lassical biological control can be a practical and useful way to manage problematic invasive weeds as long as it is done intelligently.
20 Schinus terebinthifolius Taxonomy Brazilian peppertree is in the plant order Sapindales and in the family Anacardiaceae Anacards are trees, shrubs, or lianas c haracterized by vertical resin cana ls throughout the plant tissues. The resin is clear when fresh but turns black when dry and often causes dermatitis (Judd et al. 2008 ; Morton 1978 ). F our g enera in the Anacardiaceae are indigenous to eastern North America ( Rhus Toxicodendron, Metopium and Cotinus ) with a few more introduced as agricultural crops including Mangifera indica L. (mango), Pistacia spp. (pistachio), and Spondias spp. (purple mo mbin) (Cuda et a l. 2006). However, there are no species of the genus Schinus native to North America (Brizicky 1962 ; Gleason & Cronquist 1963). The genus Schinus has approximately 30 species with a native distribution throughout South America (Barkley 19 57). Four species of Schinus have become naturalized in the US and include: S. longifolius (Lindl.) Spreg., S. polygamus (Cav.) Cabrera, S. molle L., and Brazilian peppertree (Cuda et al. 2006). Brazilian peppertree is known by the common names Brazilian pepper, pink pepper, peppertree, Christmasberry, Christmas pepper, and Florida holly (Cuda et al 2006). Morphology Brazilian peppertree is a perennial shrub to tree. It is dioecious producing flowers in dense axillary panicles twice a year in Florida that result in numerous bird dispersed red drupes (Figure 1 2) on the female plants ( Ewel 1978 ). With its shiny, evergreen compound leaves (Figure 1 2) and bright red drupes commonly observed in sery industry until 1990 when the sale was prohibited by the Florida Department of Environmental Protection
21 (Austin & Smith 1998) Capable of growing to a height 10 m or more, Brazilian peppertree rarely grows higher than 3 5 m in Florida (Habe ck et al 1 994 ). A B Figure 1 2. Photographs of Brazilian peppertree in its native range in Santa Catarina, Brazil: A) depicts the compound leaves of Brazilian peppertree and B) shows Brazilian peppertree in fruit. Photo credit L. R. Christ. Distribution Brazilian peppertree is native to Brazil, Argentina, and Paraguay (Barkley 1944, 1957). It was originally introduced to Florida in the 1840s as an ornamental (Mack 1991). It escaped cultivation and was first observed to be naturalized in Florida in the 1950s (Austin & Smith 1998). Since its introduction in Florida, this invasive weed occupies more than 300,000 ha in central and so uth Florida (Habeck et al. 1994 ; Cuda et al. 2006). Until recently, Brazilian peppertree was thought to be limited to south and central Florida due to its intolerance to freezing temperatures but there is now evidence that its range is expanding northward (Anon. 2007). Recent chloroplast DNA (cpDNA) analysis has shown there were two different introductions of Brazilian pepp ertree to Florida resulting in the establishment of two different haplotypes; A, B which have extensively hybridized since introduction (Williams et al. 2005, 2007). Brazilian peppertree has naturalized in Arizona, California,
22 Hawaii and Texas ( Habeck et al. 1994 ; Cuda et al. 2006). In Arizona, it is naturalized but not considered to be a problem weed ( A DA 2006). The distribution of Brazilian peppertree in California is limited (EDDMa ps 2009) and the plant is listed as an invasive species by the C alif ornia Invasive Plant Council (Cal IPC 2006 ). In Hawaii, it is widely distributed in lowland areas (Smith 1985) Brazilian peppertree is located on 6 of the 8 main Hawaiian Islands and on the Midway Atoll (Smithsonian 2010). It is considered to be an inv asive species in Hawaii by the US Forest Service Pacific Island Ecosystems at Risk (PIER) program (USFS 2010). Plants sampled in Hawaii are haplotype A (Williams et al. 2005). In Texas, it was first found on Galveston Island in 2003 and remains a plant Noxious Plant List (Texas Department of Agriculture 2003). Plants sampled in Texas are both haplotypes A and B (Williams et al. 2005). In Florida, Brazilian peppertree is listed as a noxious weed (FLDACS 1999), a prohibited plant (FLDEP 1993), and a Category I inva sive species (FLEPPC 2009 ) Invasive properties Brazilian peppertree is extremely adaptable and able to colonize disturbed sites such as roadsides, canals, p astures, pine woods, hammocks, and abandoned fields (Habeck et al 1994 ) It also can establish in undisturbed communities and environmentally sensitive areas like the Everglades National Park (Ewel et al. 1982, Morton 1978). Characteristics of Brazilian peppertree con tributing to its invasive properties include: fast growth, prolific seed production, tolerance to shade, fire, drought, salinity, and vigorous resprouting (Ewel 1979 ; Nilsen & Muller 1980 ; Doren et al. 1991 ; Ewe & Sternberg 2005) It can for m dense monocultures that displace native vegetation and decrease the biodiversity of native plants and animals (Bennet t et al.
23 1990 ). Brazilian peppertree fruits contain secondary compounds that produce allelopathic effects on red mangrove, Rhizophora ma ngle L., (Acanthaceae), black mangrove, Avicennia germinans L., (Rhizophoraceae), Bidens alba L. (Asteraceae) and Rivina humilis L. (Acanthaceae) plants by reducing their growth and germination ( Habeck et al. 1994 ; Morgan & Overholt 2005 ; Donnelly et al. 2008). The seeds are able to remain viable up to 7 days after being in water allowing seeds to disperse long distance along water pathways (Donnelly & Walters 2008). Control methods Current control methods for Brazilian peppertree incl ude chemical, mec hanical, and physical means. Herbicides are the most common way to control Brazilian peppertree and the most cost effective (Cuda et al. 2006). Mechanical control employs heavy equipment to remove the plants but has to be avoided in envi ronmental sensitive areas (Cuda et al. 2006). Mechanical methods also can promote suckering of the plant if the roots are left intact (Tassin et al. 2007). Physical removal can include soil removal, prescribed burning, or flooding. Complete soil removal was shown to be very effective in eliminating Brazilian peppertree in the Everglades but it is very expensive and creates large spoil hills (Dalrymple et al. 2003). The use of fire to control the species has had mixed results depending on the size of the invasion. Small populations are susceptible to fire and frequent fires can keep it at a lower density (Stevens & Beckage 2009). Larger densities can reduce fire temperature and the spread of the fire facilitating continued invasion converting pine savan nas to a monoculture of Brazilian peppertree (Stevens & Beckage 2009). Due to its invasiveness and spread in the Everglades, Brazilian peppertree was targeted for classical biological control in Florida in the late 1970s (DelFosse 1979 ;
24 Campbell et al. 19 80). The lack of native congeners minimizes the risk of damage to non target plants from introduced natural enemies, making it a good candidate for biological control (Pemberton 2000 ) Biological control of Brazilian peppertree could provide a more cost effective and sustainable approach for managing this invasive weed, especially when it is integrated with conventional control methods (Cuda et al. 2006). Several candidate biol ogical control agents have been studied for release in Florida (Medal et al. 1 999; Martin et al. 2004 ; Cuda et al. 2005, 2008, 2009; Pedrosa et al. 2006 ; Manrique et al. 2008 ; Moeri et al. 2009) One of these is the leaflet galling psyllid C. terebinthifolii (Vitorino unpublished data ) Calophya terebinthifolii The family Calophyidae is in the order Hemip tera, suborder Sternorrhyncha and superfamily Psylloidea. Psyllids feed on plant phloem and are known vectors of plant diseases: feeding damage to the plant is attributed primarily to the nymphal stage They are generall y very host specific and restricted to perennial eudicot plants (Hodkinson 1974). Psyllids reproduce sexually by laying eggs with a basal pedicel that is inserted into the host plant tissue. The pedicel absorbs water from the plant to prevent egg desicca tion (Hodkinson 1974). Psyllids typically pa ss through five instars that are susceptible to desiccation. Ambient temperature is important for determining egg and larval development rates in non diapausing psyllids and number of generations per year (Hodki nson 2009). In tropical climates, psyllids have continuous generations year round but growth rates vary depending on climatic factors and host plant conditions. Adult psyllids have limited flying ability and commonly have wind assist ed
25 dispersal (Hodkins on 1974). Psyllids frequently occur as highly aggregated colonies on their host plants (Hodkinson 2009) The family Calophyid ae contains the genus Calophya Lw which is a large genus of 59 species with a predominantly New World, Oriental and East P al a earctic distribution in a variety of both temperate and tropical habitats (Hodkinson 1989 ; Burckhardt & Basset 2000). Calophya species are associated with various Anacardiaceae and other families (mostly in the order Sapindales) and tend to be highly h ost specific (Burckhardt & Basset 2000). The distribution of C terebinthifoli i, the focus of this research was thought to be limited to Brazil and Paraguay (Burckhardt & Basset 2000). Recently however, it was found during a survey in n ortheastern Argentina and considered to be a common species (found more than 30% of the time) in the area surveyed (McKay et al. 2009) Plants in the family Anacardiaceae produce three types of toxic phenolic compounds: biflavonoids, alkylcatechols, and a lkylresorcinols. Brazilian peppertree co ntains alkylresorcinols which are less common among Anacardiaceae (Aguilar Ortigoza & Sosa 2004). When comparing the phylogenies of Calophya and Anacardiaceae, most Calophya species can synthesize alkylresorcinols and feed on the members of the family producing these chemicals (such as Schinus spp.) further connecting the evolutionary history and close associations (Aguilar Ortigoza & Sosa 2004). C alophya terebinthifoli i is described in Burckhardt and Basset (2000) Adults have a brown to black head and thorax with a green to yellow abdomen. Adults show morphological differences b etween female and male psyllids
26 Natural enemies of C. terebinthifolii in its native range include flies of the family Syrphidae, vari ous species of Neuroptera, and Araneae (spiders) that catch the adults in their webs (Burckhardt & Basset 2000 ; personal obser vation ). H ymenopterous parasitoids also attack the nymphs (personal observation). Parasitoids of the genera Tamarixia (Eulophidae) and Aprostocetus (Eulophidae) small parasitic wasp s have been recorded as attacking the closely related Calophya schini Tuthill (Hemip tera: Psyllidae) in Mexico and California ( Alvarez Zagoya & Cibrian Tovar 1999 ; Evans 2002 ). Biology of Gal l Formation Cecidology is the study of galls. Gall formation arose numerous tim es within the phylum Arthropoda and is considered to have an old evolutionary history as an adaptation to deal with increased pl ant defenses (Raman et al. 2005 ) Galls also m ay protect psyllids from climatic extremes (Ananthakrishnan 1984). Galls are tissues acting as a metabolic sink pulling nutrients from adjacent plant tissues or other parts of the plant (Harris & Shorthouse 1996). The extent of va scularization and ligni fication of the gall determines the amount of nutrients the gall pulls from the plant (Harris & Shorthouse 1996). Most gall forming arthropods display high levels of specificity to particular plant species and plant organs with leaf tissue as the most sus ceptible plant organ for gall development (Raman et al. 2005 ). Cecidogenous psyllids have a preference for specific hosts indicating a strong coevolution ary relationship between psyllids and their host plants Cecidogenous psyllids are well represented among Psylloidea and Calophyidae but their distribution is uneven geographically and t axonomically (Raman et al. 2005 ) Nymphs feed on actively growing tissues and derive their nutrients from plant sap. N ew tissue growth
27 and Calophy a gall initiation is correlated with increasing gall mortality on maturing leaves on Schinus (Hodkinson 2009). Mature leaves are a poor source of soluble nutrients required by sap feeding psyllids (Hodkinson 2009). Psyllids may also inject saliva into th e plant during feeding which may be involved in plant disease tra nsmission (Ananthakrishnan 1984 ). Growth rate and reproduction is determined by the quality and quantity of nutrients available (Ananthakrishnan 1984 ). Nutrient tissue is not differentiate d in the galls of many psyllids; they take their food in or near the vascular tissue of the plant or gall (Rohfritsch 1992 ). T he mode of feeding of gall insects is also correlated with the complexity of their galls. Most hemipteran galls are simple katap lasmic galls that have walls composed of undifferentiated parenchymous cells and are considered the most primitive type of galls (Rohfritsch 1992 ). Calophya terebinthifolii develops in open pit galls on the leaflets of Brazilian peppertree (Burckhardt & Basset 2000). The pit galls that are formed by the de veloping nymphs are kataplasmic galls and are characterized by a slight arching of the leaf blade (Dreg er Jauffret & Shorthouse 1992 ; Roskam 1992 ). Calophyidae induce gall formation during the first instar, forming pit galls that are circular in outline. The nymphs have a sclerotized, shield like dorsal surface and a membranous v entral surface (Burckhardt 2005 ). Calophyid nymphs are flattened dorso ventrally and adapted for their sedentary existence causing a depression in the leaf where their dorsal surface forms an almost smooth cover ove r the pit (Ananthakrishnan 1984 ). In C. terebinthifolii Vitorino et al. (2007) found the average time for gall initiation was 27 days after plants were exposed t o mated females.
28 Project Justification The rationale for using C. terebinthifolii as a biological control agent is based on a similar, although unintentional, introduction of a closely related psyllid species into California (Downer et al. 1988). Calo phya schini which was first discovered in Los Angeles County California in July 1984 dispersed rapidly from San Diego County to the San Francisco Bay ( spanning a distance of about 800 km) in less than 4 years (Downer et al. 1988). This congener causes ex tensive da mage on Peruvian or California p eppertree, Schinus molle In California, Downer et al. (1988) reported that C. schini attacked only S. molle causing damage to these popular ornamental plants. It is believed that the spread of the psyllid was f acilitated by the nursery industry and from the extensive use of California peppertree by the California Department of Transportation (CALTRANS) along freeways ( Hagan & Tassan 1996) Without natural enemies to control the psyllid in California during the initial outbreaks, the pitting and nymphal development by C. schini on infested trees in California caused extensive defoliation, discoloration, and distortion of the leaflets (Downer et al. 1988 ; CDFA 1984 ) In 1987, a biological control program was initiated to control th e psyllid with a parasitic wasp Tamarixia sp. This parasitoid was effective in reduc ing the abundance of the psyllid (Hagan & Tassan 1996). Assuming C. terebinthifolii would respond simi larly if it were introduced into Florida without its natural enemies, this leaflet galling psyllid may be a promising candidate for the biological control of Brazilian peppertree. Additionally, th ree different studies on two species of Calophy a; C. schin i in Chapingo, Mexico (Alvarez Zagoya & Cibrian Tovar 1999 ) C. schini in Ventura County,
29 CA ( Downer et al. 1988 ), and C. terebinthifolii in Santa Catarina, Brazil ( Barbieri 2004, personal observation ) found the insects to have continuous generations The polyvoltinism exhibited by Calophya psyllids also makes them good candidate s for biological control because they can provide year round control. It should be noted that C. schini was reinstated as a separate species from Calophya rubra Blanchard by Burckhardt & Basset (2000) Therefore, the study in Chapingo, Mexico was based on C. schini and not C. rubra Although not gall forming, other psyllids such as Boreioglycaspis melaleucae Moore have been successfully used as biological control agents Boreioglycaspis melaleucae a biological control agent of Melaleuca quinquenervia (Cav.) S.T. Blake (Myrtaceae), caused foliage damage, defoliation, and twig dieback that was visually evident within 1 year after its release which resulted in a 16% annu al mortality rate of m elaleuca trees (Rayamajhi et al. 2007). This psyllid impacts the leaves of Melaleuca by causing the premature abscission of mature leaves and reduced seedling survival by 55% within three generations (Center 2007). In another Florida study where melaleuca was attacked by a well established population of B. melaleucae the foliage biomass decreased disproportionately among smaller trees but remained relatively unchanged among larger trees (Rayamajhi et al. 2008). Apparently, s e edlings of invasive woody species are more vulnerable to damage from biocontrol agents than mature trees due to the reduced leaf area and carbon storag e reser ves (Franks et al. 2006). The m elaleuca psyllid was found to disperse at a rate up to 10 km per y ear (Hodkinson 2009). Another species of psyllid, Heteropsylla spinulosa Muddiman Hodkinson and Hollis has had similar success as a biological control agent on the giant sensitive plant,
30 Mim osa inv i sa Mar t. ex Colla (Fabaceae). At a sugar plantation in Papua, New Guinea where H. spinulosa was released in 1991, the ground cover with 100% infestation of M. invisa declined to less than 5% within 2 years (Kuniata & Korowi 2004). In this same long term study, Kuniata and Korowi (2004) found that the seed production of M. invisa was reduced to less than 20% over a 10 year period following exposure to the psyllid. Objectives The objectives of this study were designed to accelerate testing of C. terebinthifolii as a biological control agent for Brazilian peppertree in Florida. Five components necessary for developing this agent were conducted: 1. Establish a laboratory colony of C. terebinthifolii in quarantine at the University of Florida Gainesville campus 2. Investigate biology and life table parameters of C. terebinthifolii in Brazil 3. Compare the performance of C. terebinthifolii on the genotypes of Brazilian peppertree that occur in Florida 4. Use ecological niche modeling to predict climatic overlap of C. tere binthifolii in Florida and predictions for future surveys in South America for C. terebinthifolii 5. Develop recommendations or considerations for integrating current Brazilian peppertree management practices with biocontrol efforts using C. terebinthifoli i.
31 CHAPTER 2 COLONY ESTABLISHMENT OF C. TEREBINTHIFOLII Objective 1 Establish a laboratory colony of C. terebinthifolii in quarantine at the University of Florida Gainesville campus One of the steps of biological control research is importatio n and quarantine of the potential agent ( Messing & Wright 2006) It is essential be able to rear an insect in sufficient numbers in the l aboratory to allow host range testing. Materials and Methods Adult psyllids and juveniles contained in pit galls on Brazilian peppertree leaves were collected from three states in Brazil; Santa Catarina, Paran and Bahia (see Appendix A for GPS coordinates). Under permit s from the US Department of Agriculture and the Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renovveis (IBAMA), a total of four shipments and five hand carried packages of psyllids were received at the Florida Biological Control Laboratory (FBCL) in Gainesville, FL between February 2009 and May 2010 from various locations in Brazil (Figur e 2 1 ) The contents of the packages varied but al l packages contained one or both of the following: 1) Brazilian peppertree leaves infested with galls containing late stage instars sealed inside a plastic bag or rigid plastic container and 2) Brazilian peppertree seedlings with roots wrapped in damp paper towels sealed in a plastic bag around the stem with the vegetative part (less than 25 cm) in a plastic 1 liter drink container wit h live adult psyllids (Figure 2 2 ) All shipped packages were sent by DHL delivery service in Styrofoam cooler lined cardboard boxes. Hobo data loggers were included in the boxes to monitor the temperature and relative humidity during
32 transportation. H and carried packages were brought back in the personal luggage of individual researchers from collecting trips to Brazil. Figure 2 1 Map of collection sites where C. terebinthifolii has been found The bright green circles indicate a potential new species of Brazilian peppertree psyllid as identified by Dr. Daniel Burckhardt (personal communication).
33 All packages containi ng psyllids were opened in the m aximum security room at the FBCL and the psyllids remained in the maximum security room for the du ration of this study Voucher specimens were submitted to Dr. Susan Halbert of the Florida Department of Agriculture & Consumer Services, Division of Plant Industry. Dr. Halbert is a n Entomologist whose area of expertise is psyllid systematics Shipment s were thoroughly examined for surviving and newly emerged adults, parasitoids, and live nymphs. Leaves were examined under a dissecting microscope for viable nymphs inside pit galls To increase the success of psyllid emergence from shipped leaves vari ous techniques were tested. 1. Individual leaflets with fifth insta rs were placed individually in small plastic 29.6 ml ( 1 oz. ) diet cups with damp filter paper (3.5 cm diameter), a small drop of honey on the side of the cup, a fitted paper lid, and then covered wit h a plastic lid. 2. Infested leaflets (< 5) were placed inside standard 9 cm clear plastic Petri dishes with wet filter paper (6 cm) and sealed with Parafilm wax 3. Leaflets were placed upright in wet, previously autoclaved sand in round plastic insect cages (diameter 25.4 cm, depth 8.89 cm) with 2 4 round holes on the sides covered with fine mesh to allow air circulation. In these containers, a diet cup containing Gat orade with a cotton wick was added to provide a food source for the adults. 4. Leaflets were laid on top of a slightly damp Kimwipes sheet inside square 10x10x10 cm plastic insect containers without ventilation. 5. Occasionally a nymph would be seen crawling out of a gall presumably looking for a source of phloem. These nymphs wer e transferred with a fine bristle paint
34 brush to a Brazilian peppertree plant which was then covered with a mesh bag (30 cm x 60 cm) and secured with a 15 cm vinyl coated twist tie. Live a dults from the shipments were examined under a dissecting microscope to determine their sex. Three different techniques were used for determining the mos t efficient rearing procedure 1) M ale and female adults were released into a large Plexiglas insect cage (61 cm x 61 cm x 61cm) containing 12 15 Brazilian peppertree seedlings less than 45 cm in height in 9 cm square nursery pots (Figure 2 2). Seedlings were obtained from haplotype A (West coast) and haplot ype B (East coast) trees i n Florida A B Figure 2 2 Photographs of Braz ilian peppertree seedlings: A) s eedling prepared for shipping in a plastic 1 liter drink container and B) s eedling in FBCL exposed to adult psyllids obtained from overseas shipments. Photo credit L. R. Christ. 2) Individual Brazilian peppertree plants ( haplo type A) grown in 2 liter nursery pot s were covered with cylindrical Plexiglas insect cages (51 cm height x 15 cm diameter) with mesh tops and mesh ventilation holes on the sides Male and female psyllids were
35 released in to the cylinders. 3) Using larger Brazilian peppertree plants ( ~ 1 m), male and female psyllids were placed on the plants then covered with a 30x60 cm mesh bag and secured with a nylon cable ties. Hobo data loggers were placed inside the cages to recor d the temperature and humidity. The mean temperature inside the cages was 26 1.5 C with a relative humidity of 84% 4.9%. All cages were kept at a 16 L : 8D photoperiod with access to artificial and natural light The plant material from Brazil was che cked daily for adult emergence; water was added as needed to filter paper and sand to maintain humidity. Leaflets were checked for p syllid emerge nce for up to 30 days from the date the shipped samples were collected depending on the condition of the plant material. P arasitoids collected were either frozen (to be pointed) or placed in 95 % EtOH to be identified (Appendix F) Seedlings obtained from Brazil were planted in Fafard 4 Mix Professional Formula potting soil and fertilized weekly with a 5 1 1 (5% total N, 1% P 2 0 5 1% K 2 O) fish emulsion following the dilution recommendations from the manufacturer. Results and Discussion I was unable to establish a colony of psyllids at the FBCL. One reason was the amount of time the shipments spent in transit. Shipments leaving Brazil and going through US Customs and Homeland Security took as long as 10 days to arrive. After 9 days in transit adults and nymphs die d and plant materia l rotted The shipments that contained the most liv e adults and nymphs were in transit < 5 days (see Appendix G)
36 Another reason for the lack of success in establishing a colony was that a few collecting trips occurred when psyllid field populations were low. For instance, a collecting trip in March 2010 coincided with the flowering of Brazilian peppertree ; in Santa Catarina state, the psyllids were scarce during this time Although I was successful in collecting psyllids further n orth in the state of Paran, the psyllids did not survive on Florida ha p lotype plants. Hand carried packages produced the most live adults. Adult females f rom a hand carried shipment in August 2009 laid eggs. However, most nymphs died during t he first instar and none develop ed beyond third instars The rearing technique used that produced the eggs and instars was the third technique using the me sh bag over the seedling plants. Red banded thrips Selenothrips rubrocinctus Giard, also reduced survival of the psyllids at FBCL. The thrips are a pest that damages the leaves of the Brazilian peppertree. When they are on the same plants as the psyllids, pesticides cannot be used to control them and h and removal of the thrips is required. Having the individual plants separated with mesh bags helped with the thrips problem in quarantine Psyllids did not produce eggs when released into the large Plexiglas insect cages with multiple plants. Receiving cut leaves with late inst ars had mixed results. Once the leaves are severed from the plant, the only nymphs that can emerge as adults are 5 th instars that are able to mature within 1 week after the leaves are removed from the plant. After that time, the leaves are too dried out to maintain the developing nymp hs I was unable to test the e ffects of fertilizers on the survival of the developing psyllids because a colony was not established.
37 Based on these findings I would recommend hand carrying psyllid shipments Most live adults and adult emergence occurred in the hand carried shipments. The only gall development on the plants came from eggs laid by adults received via hand carry If shipping were the only option, shipments received within 5 days of being packaged w ould increase survivorship For packaging psyllids for shipment, placing cut leaves in hard plastic containers (as opposed to plastic bags) produced more live adults that emerged in transit. In the plastic bags, adults that emerged were usually dead by t he time the package was received. Placing filter paper or paper towels in t he containers with the leaves was essential to control the humidity and prevent the leaves from molding. The best method for rearing the insects in the lab oratory was using plant s (over 60 cm tall) with multiple new leaf flushes covered with large mesh bags. Timing the collection of psyllids from Brazil was also critic al. During this study, mid May to early August 2009 and early December 2009 were the best times for collecting l arge numbers of adults and nymphs It was important to collect psyllids when the plants were not flow ering and most of the fruits were gone. During the flowering/fruiting season, the plants flush few new leaves and the psyllids are not abundant Gall fo rmers require host tissues to be at a specific developmental stage in order to initiate gall induction (Weis et al. 1988). Preliminary data suggest C. terebinthifolii is locally adapted to the Brazilian peppertree genotypes ( see Chapter 4 ) so another impo rtant aspect of rearing is having the appropriate plant type that corresponds with the collection site in Brazil. During a collection trip in March 2010, p syllids collected o n Brazilian peppertree in Salvador, Bahia, Brazil were shipped to me for rearing After submitting voucher samples to Dr. Halbert, they were determined to be a different, undescribed species in
38 the genus Calophya Samples were sent to Dr. Daniel Burckhardt, a psyllid expert from the Naturhistorisches Museum, Switzerland. His description of the species follows: Morphologically your specimens resemble a lot Calophya terebinthifolii but adults from Bahia are green/yellow compared to the dark C. terebinthifolii The terminal setae on the antenna look slightly shorter and the fema le subgenital plate may be slightly more curved ventrally. In the males I did not find striking differences but the morphology of the male terminalia is relatively homogeneous in the Schinus calophyids. The differences between C. terebinthifolii and your s ample are most marked in the nymph where your sp ecimen has anteriorly more round ed humeral lobes, only 3 sectasetae on the antenna and few, very slender marginal setae on the wing pads. If these differences are constant they indicate that there are two, cl osely related, species At the moment I would call the sample from Bahia Calophya sp. cf. terebinthifolii Specimens were also provided to Dr. Dean Williams at Texas Christian University to perform DNA extraction, amplification, and sequencing and comp ared the results to the psyllids collected from Curitiba and Santa Catarina, Brazil. He found the psyllids from Salvador, Bahia, BR to be different from the psyllids collected in southern Brazil. He found a 13.4% sequence divergence between the Salvador psyllids and the psyllids from southern Brazil using the Kimura Two Parameter model (K2P) (Kimura 1980) possibility indicating the two are different species. He sequenced four psyllids from Salvador and found three different haplotypes The divergence is slight (0.2 0.7%) and most likely indicates that all the Salvador samples are the same species. See Appendix E for the Fasta file with the sequencing of the southern Brazilian psyllid C. terebinthifolii and the 4 psyllids sequenced from Salvador Calophya sp. cf. terebinthifolii Images of late instars and adult females collected in Salvador were obtained using Automontage (Figure 2 3). Measurements (in mm) of a dults (n=28) found : 1) body length 1.409 0.036 SEM, 2) wing length 1.501 0.019 SEM and 3) head capsule width 0.382 0.008 SEM. Using ProcTTest in SAS software, Version 9.0 (SAS 2002) with a significance level of
39 males approached significance ( P =0.0598) which may be explained by the low sample size of males (n=9). Further research is needed on this potential ly new species including a full taxonomic description and classification. A B C D E F Figure 2 3. Automontage images of Calophya sp. cf. terebinthifolii A) Adult female, B) A dult female, C) Adult male, D ) 4 th or 5 th instar dorsal view, E ) 4 th or 5 th instar ventral view, and F ) 5 th instar molting into an adult. Photo credit L.R. Christ.
4 0 CHAPTER 3 BIOLOGY OF C. TEREBINTHIFOLII Objective 2 Investigate biology and life table parameters of C. terebinthifolii in Brazil I spent three months (11 May 8 August 2009) in Gaspar, S anta C atarina Brazil (Figure 3 1) working with Dr. Marcelo Vitorino at the Laboratrio de Monitoramento e Proteo Florestal LAMPF, ( Forest Protection and Surveillance Laboratory located at 26.9110, 48.9362) part of the Fundao Universidade Regional de Blumenau FURB, ( Regi onal University of Blumenau ). By conducting my research in Brazil, I was able to avoid working in quarantine in Florida an d study the psyllid under more natural conditions. Working in Brazil also provided me with a readily accessible supply of psyllids for my research and for shipping to Florida to establish a colony at the FBCL in Gainesville, FL Material and Methods To begin the life cycle tests, I established a colony of psyllids collected in the field. I n total 13 field collection trips were made in Brazil. Geoposit i on sampl e data and s ite descriptions were recorded and are listed in Appendix A for each collection site Collection d ates with corresponding weather conditions (i.e. relative humidity, temperature, wind speed, and precipitation) are provided in Appendix B Adults were collected individually in the field along the eastern coast of Santa Catarina, Brazil using The gel caps containing adult psyllids were stored in coolers and then transported ba ck to LAMPF within 3 5 hours after captur e. Field collected adults were released on Brazilian peppertree seedlings The seedlings used in the studies were grown in the greenhouse at LAMPF prior to my
41 arrival and were all approximately 40 cm in height in 1 liter nursery pots with at least 3 flushes of new leaflets The following parameters were measured: A) l ocation of pit gall formation on plants in the field, B) female fecundity C) v erificati on of the number of instar s and sizes and D) development time and survivorship. A) Location of Pit Gall Formation on Plants in the F ield The location of the pit galls on the Brazilian peppertree leaves could be an important factor affecting psyllid survival. For instance, gall orientation may provide protection from predation or adverse weather conditions I hypothesize d that C. terebinthifolii created pit galls on the adaxial (upper) side of young leaflets of Brazilian peppertree based on personal observations To test this hypothesis, I examine d 10 randomly selected Brazilian peppertrees with C. terebinthifolii galls from four different field sites ( n=867 leaflets ) Sites were s eparated by a minimum of 3 km. The diameter at breast height ( DBH) wa s measured and one branch at breast height (1.37 m) was cut in each cardinal direction (when possible) using a compass on a Garmin GPS map 76S unit to determine the direction. Where I could not obtain branches from the four cardinal directions on all 10 trees the closest branch at breast height was collec ted. Sampled branches were labeled, bagged and transported to LAMPF where the number and location of psyllid pit galls were observed and recorded. B) Female F ecundity To determine fecundity, 1 female and 1 male (newly emerged) were released on a Brazilian peppertree seedling (n=28) covered with a 30x60 cm mesh bag and secured at the b ase of the seedling stem with two vinyl coated twist ties. Leaves were che cked daily for eggs T emperature and relative humidity were also recorded. Female wi ng length (mm) body leng th (mm) and head width (mm) were measured at death.
42 Fecundity was re gressed against female wing length with desiccated females removed from the analysis (Geiger & Gutierrez 2000). C) Verification of the Number of I nstars and S i ze s First instars were observed in the lab but not on the vegetation f rom the field because pit galls do not form until the second instar (personal observation) Using psyllids on the branches cut for the gall orien tation study (n=177) and first instars emerging in the lab (n=18) individual instar body length (mm), body width (mm), and head capsule width (mm) were measured Based on previous studies, this sample size of 195 nymphs was adequate for generating a histogram of ins tar s and size range of each instar for cecidogenous psyllids (Devi & Prabhoo 1995; Alvarez Zagoya & Cibrian Tovar 1999). Length, width, and head capsule width were measured under a dissecting microscope From a literature review on other psyllids, I hypo thesized there would be 5 instar s ( Hodkinson 1974) Measurements were compared to those of C. schini recorded by Alvarez Zagoya and Cibrian Tovar (1999) as a point of reference and comparison between the two similar species. D ) Development Time and S urv ivorship Using the eggs from the female fecundity study, psyllid egg cohort s were followed through their development with survivorship and temperature recorded daily. Psyllids developed on a Brazilian peppertree seedling covered with a 30x60 cm mesh bag a nd secured at the base of the seedling stem with two vinyl coated twist ties. The psyllids were kept at three t emperature regimes: 1) A rearing room located at LAMPF without any temperature or humidity control with access to both natural and artificial light on a 12L:12D cycle with a n average temperature of 18 .0 5.0 C and a relative humidity of 85.9 5.0 % (n=1) 2) Environmental growth chamb er with a 16L:8D light
43 cycle, an average of 22.5 2.6 C, and relative humidity of 66.5 6.8 % (n=3). 3) The last temperature regime was a combination of time spent in the rearing room (first 12 days) and an e nvironmental growth chamber (final 41 days) for an overall average of 20 .0 4.9 C a nd 72 9.8 % relative humidity (n=1). Statistical A nalysis The data were analyzed with SAS software, Version 9.0 (SAS 2002). Sample means SEM were determined for all parameters After confirming the data were normal ly distribut ed a linear regression was performed to correlate fecundity with female size. A f requency distribution was created to determine number of instars A the means of body length of male and female adult C. terebinthifolii and of C. terebinthifolii and C. schini were statistically compared Sample statistics were calculated using standard techniques (Price 1997). A survivorship curve of the population an d a life and fertility table were constructed. The mean generation time and instantaneous rate of increase w as generated from the life tables using the following equations: x = age at beginning of interval l x = age specific survivorship m x = expected daughters Net reproductive rate: Cohort generation time: Instantaneous rate of increase : Mean number of females present during age interval x: T x = cumulative sum of the product of individual and time units Life expectation:
44 Development time and degree days were also generated using the following equations: where y = rate of development and x =temperature Minimum threshold for development: Thermal constant: or Results and Discussion A) Location of Pit Gall Formation on Plants in the F ield The results of field studies confirmed the hypothesis that the open pit galls produced by the developing nymphs were located on the ada xial (upper) side of the leaflets (Figure 3 1) Of the 2239 galls observed, 2235 galls were on the adaxial side (99.82%) with only 4 galls on the abaxial side of the leaflets. The mean number of galls found o n a leaflet per branch was 2.58 1.8 (range 0 34 galls per leaflet) In addition to the location of the galls on the leaf surface, psyllid nymphs tolerated a range of growing conditions in the field. For instance, viable galls were observed on Brazilia n peppertree in direct sunlight, shaded by other branches, windy areas along dusty unpaved roads, and along freeways. From laboratory observations, psyllids tended to form a pit gall after the crawling stage and remain there until adult emergence. Howeve r, on several occasions, nymphs were observed exiting their pit galls when a leaflet had been severed from the plant. The psyllids were presumably looking for another nutrient source.
45 A B Figure 3 1. A) L ocation of laboratory studies conducted at LAMPF, Gaspar, SC, Brazil. B) Photograph of open pit galls of C. terebinthifolii on the adaxial side of the leaves. Photo credit L. R. Christ A t LAMPF and FBCL nymph s that had exited their galls were transferred to other Brazilian peppertree leaflets with a fine bristle paintbrush. Although psyllids we re successfully transferred using this method, no ne of the nymphs survived to form galls. Further research is needed to explore whether late insta rs other than the crawling stage are able to mov e successfully to a new leaflet and complete their development in another pit gall. B) Female F ecundity Only 18 of 28 females in cages containing mating pairs of adult psyllids laid eggs. For C. terebint hifolii the mean number o f eggs laid per female was 55.3 8.9 (n=18) with a range of 16 139 eggs (excluding the 10 females that did not lay eggs) These values are similar to those f or C. schini : average number of eggs laid was 50 (no SEM was given) with a range from 25 110 eggs from 10 female psyllids (Alvarez Zagoya &
46 Cibrian Tovar 1999). The oblong shaped eggs of C. terebinthifolii are a milky, translucent cream color when first laid then turn a black, iridescent color (Figure 3 2 ) with a mean le ngth of 0.212 0.002 m m (Table 3 1 ). Similar to C. schini (Downer et al. 1988), the females laid their eggs on the new leaf flush along the leaf margin, the midvein of the leaf leaf petiole, and leaf buds A B Figure 3 2 Photograph s of C. terebint hifolii : A) eggs on S. terebinthifolius along the midvein of a leaf. Cream colored eggs are < 24 hours old B) Shows newly hatched first instar psyllids during their crawling stage. Photo credit L. R. Christ. T he number of eggs laid by a female increased with wing length ( F =5.15; df=9; P =0.0494; Figure 3 3 ). The R 2 is not very high (0.364) due to the high variability in the data Unfortunately, not all female psyl lid s that laid eggs (n=18) were re covered for measurement resulting in a sampl e si ze of 11
47 Figure 3 3 The relationship between female wing length and the number of eggs laid (n= 11 ). C) Verification of the Number of I nstar s and S izes To determine the number of instars, a histogram of body length was constructed and examined for distinct peaks. The histogram (Figure 3 4 ) confirms 5 instars for C. terebinthifolii y = 234.6x 301.38 R = 0.364 0 20 40 60 80 100 120 140 1.35 1.45 1.55 1.65 1.75 No. of eggs laid Female wing length (mm)
48 Figure 3 4 Frequency distribution of larval length measurements of C. terebinthifolii Peaks within groupings indicate five instars. The congener C. schini was reported by Downer et al. (1988) to have only four distinct peaks after measuring 1,000 nymphs in all stages of develop ment over a 10 week period. Alvarez Zagoya and Cibrian Tovar (1999) later studied C. schini and found evidence of 5 instars when instars were measured in a laboratory setting It is possible that Downer et al. (1988) encountered the same problem I did when only field specimens were measured. The first instar crawler stage is hard to find in the field because of the ir small size (Table 3 1) susceptibility to desiccation, and inability to form a pit gall (Figure 3 1 ). Automontage images were taken of the life stages (Figure 3 5 ). The various measurements of all life stages of C. terebinthifol ii are provided in Tabl e 3 1 and are compared with its congener C. schini in Figure ( 3 6 ) The adult female body
49 lengths are significantly larger than adult males ( t =3.65; df=38 ; P =0.008) which is typical for most psyllids (Hodkinson 2009) By comparison, C. schini is significantly larger than C. terebinthifolii in the following stages: egg ( t =10.85; df=52; P < 0.0001) 3 rd instar ( t =3.62; df=42; P =0.0008) 5 th instar ( t =7.63; df=157; P < 0.0001), and adult ( t = 3.54 ; df=68 ; P =0.0007 ) The size of the adult psyllids of the n ew species Calophya sp. cf. terebinthifol ii found in Salvador, Bahia, BR, was also compared to the adults of C. schini and C. terebinthifolii The new species is similar in size to C. schini ( P =0.1399) but is significantly larger than the congeneric C. terebinthifolii ( P =0.0001). A B C D E F
50 G H Figure 3 5. Automontage images of Calophya terebinthifolii A) 1 st instar, B) 2 nd instar, C) 3 rd instar, D) 4 th instar, E) 5 th instar, F) 5 th instar prior to darkening, G) adult male, and H) adult female. Photo credit L.R. Christ. Table 3 1 Body size of all life stages of C. terebinthifolii Stage n (mm) Egg 30 0.212 0.002 length 0.099 0.002 width Instar 1 st 18 0.228 0.005 length 0.183 0 .012 width 0.108 0.004 head capsule width 2 nd 67 0.307 0.002 length 0.298 0.002 width 0.155 0.002 head capsule width 3 rd 14 0.446 0.010 length 0.442 0.0 09 width 0.222 0.0 06 head capsule width 4 th 24 0.674 0.0 15 length 0.66 5 0.015 width 0.320 0.009 head capsule width 5 th 72 0.962 0.006 length 0.942 0.006 width 0.398 0.002 head capsule width Adult 40 1.251 0.020 body length 1.509 0.018 wing length 0.441 0.007 head capsule width Female 28 1.293 0.021 body length 1.542 0.020 wing length 0.453 0.008 head capsule width Male 12 1.154 0.013 body length 1.432 0.025 wing length 0.415 0.009 head capsule width
51 Figure 3 6 Graph comparing the sizes of the life stages of C. terebinthifolii and C. schini Measurements for C. schini obtained from Alvarez Zagoya and Cibrian Tovar (1999). D ) Development Time and S urvivorship Out of the 18 cages where females laid eggs, only five cages produced a full generation from e gg to adult Of those five caged plants, four were haplo type A and one was haplo type O. Due to limited space in the environmental growth chambers, cages had to be split between the rearing room and the chambers The temperature regimes were comparable because my goal was to maximize rearing success from egg to adult rather than to document th e temperature tolerance of the psyllids. With limited 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Egg 1st instar 2nd instar 3rd instar 4th instar 5th instar Adult Length (mm) Psyllid Growth Stage C. terebinthifolii C. schini *
52 time in Brazil and the length of the development time (unknown pr ior to my studies), I was able to get a full generation completed before return ing to Florida. A survivorship curve was developed u sing daily observations on survivorship from the 4 successful cages (n=157 eggs) kept in the environmental growth chambers (Figure 3 7 ) F igure 3 7 Surviv al of C. terebinthifolii in laboratory environmental growth chambers at 21.9 2.6 C ; relative humidity, 67.9 7.6%; with a photoperiod of 16:8 h (L:D). The eggs started to hatch on Day 9. The graph shows a sharp decrease in survival during the first week after hatching when the nymphs are in the vulnerable crawler stage. Survivorship levels off after the crawlers settle into pit galls and then survival gradually decreases during the 3 rd 4 th and 5 th instars unti l adults start to emerge around Day 32. The average time from egg to adult was 43.7 1.2 days.
53 Using the same 4 successful cages used to create the survivorship curve, a life and fertility table was developed (Table 3 2). Table 3 2. Life and fertility table for C. terebinthifolii at 22C (3 females in cohort). Week, x Proportion alive, l x m x Reproductive Expectation ( l x m x ) xl x m x l x+1 L x T x Life Expectation ( e x ) 1 1.00 0.00 0.00 0.00 0.81 0.91 2.86 2.86 2 0.81 0.00 0.00 0.00 0.55 0.68 1.96 2.41 3 0.55 0.00 0.00 0.00 0.52 0.54 1.28 2.32 4 0.52 0.00 0.00 0.00 0.27 0.40 0.74 1.42 5 0.27 0.33 0.09 0.45 0.15 0.21 0.35 1.28 6 0.15 1.67 0.25 1.50 0.05 0.10 0.14 0.90 7 0.05 3.67 0.18 1.28 0.01 0.03 0.04 0.70 8 0.01 0.33 0.00 0.03 0.00 0.01 0.01 0.50 R 0 =0.53 T C =6.19 r = 0.1036 Only 3 females were in the cohort because two of the cages that were successful had the same female lay eggs in both cages. The results from the life and fertility table show a decreasing population since the net reproductive rate, R o =0.53, was less than 1 The instantaneous rate of population increase, r is negative ( 0.1036) With such a low sample size I would not make the assumption that what was found in the laborator y studies regarding rate of increase and net reproductive rate is true for natural populations. The survivorship of the psyllids on the Brazilian peppertree seedling may be due to the size of the plants not being large enough to support more individuals The cohort generation time, T c for C. terebinthifolii was 6.19 weeks at 22C Rates of development of the psyllid at three temperature regimes were plotted to determine degree days required for generation devel opment (Figure 3 8 ).
54 Figure 3 8 Graph showing the linear regression of rate of development for one generation (1/Days) of C. terebinthifolii at three temperature regimes. Using the inverse of the number of days required to complete a generation at the three temperatures, the following equation was generated: y = 0.001531x 0.01285. The minimum threshold for development (z) was 8.39 C and the thermal constant (K) was determined to be 653.02. Using th e parameters of z and K, days required to complete development for various temperatures can be computed. 0.0000 0.0050 0.0100 0.0150 0.0200 0.0250 18 20 22.5 Development Rate (1/D) Temperature ( C)
55 CHAPTER 4 BRAZILIAN PEPPERTREE HAPLOTYPE SUITABILITY Objective 3 Compare the performance of C. terebinthifolii on the genotypes of Brazilian peppertree that occur in Florida. As mentioned earlier, Brazilian peppertree populations in Florida are the result of two separate introductions from different parts of the native range ( Williams et al. 2005). Figure 4 .1 illustrates the relationship between the different cpDNA haplotypes (Williams unpublished data). In Florida, two separate introductions resulted in the establishment of tw o cpDNA haplotypes; A and B. In Florida, haplotype A is concentrated on the west coast whereas haplotype B is mostly found along the east coast. Novel genotypes have resulted from hy bridization between the two types of trees (Williams et al. 2005, 2007). A recent study by Manrique et al. (2008) examined the feeding preferences and performance of two closely related thrips species associated with Brazilian peppertree on the two Florid a genotypes and their hybrids. Only one of the thrips, P. ichini from Ouro Preto, Brazil performed well on the Florida haploid types of Brazilian peppertree. Pseudophilothrips gandolfoi Mound developed poorly on Florida genotypes (Manrique et al. 2008) s uggesting specialization on specific genotypes in the native range. Local adaptation of herbivore species occurs when they become specialized on host genotypes and develop corresponding genetic differentiation (Edmunds & Alstad 1978; Boecklen & Mopper 19 98). Local adaptation can be an important factor when choosing a biocontrol agent. Palmer and Witt (2006) rejected the psyllid, Acizzia melanocephala Burckhardt & Mifsud, for biological control of Acacia nilotica (L.) Delile subsp. i ndica (Benth.) Brenan (Fabaceae) in Australia because it was so specific that it could not establish on the indica subspecies. The psyllid oviposited and developed into first and second instars but third, fourth, and fifth instars were rarely seen. A mismatch
56 of introduced b iotypes adapted to different climates also can result in unsatisfactory control of plants even if the plant is well regulated in its native range (Huffaker & Messenger 1964). Local adaptation tends to occur in insects that are relatively sedentary and sp ecialized (Strauss &Karba n 1998). Peterson & Denno (1998) found lower levels of gene flow in species that exploit woody versus herbaceous hosts. They also found that local adaptation was more likely to occur in natural than agricultural habitats Host plant quality, when measured by insect fitness, can be influenced by genotype (Carr & Eubanks 2002). Calophya terebinthifolii has a number of characteristics that would lead to local adaptation. These characteristics include: 1) being a gall produce r, 2) limited dispersal capabilities (especially as juveniles), and 3) has a woody host plant and occ urs in natural habitats. I hypothesized that C. terebinthifolii is locally adapted to cpDNA haplotypes of Brazilian peppertree. Psyllids tend to be host specific (Hodki n son 1974, 2009) and differences in plant genotypes could lead to C. terebinthifolii only being able to develop on plant types most closely related to those it was collected from in Brazil.
57 Figure 4 1. cpDNA haplotype network of Braz ilian peppertree illustrating relationships between the different haplotypes based on DNA sequencing. Each connecting line indicates one nucleotide differences and unlabeled nodes are inferred intermediates. Florida haplotypes are highlighted. Figure cour tesy of D. A. Williams.
58 Materials and Methods To test the hypothesis that C. terebinthifolii performs better on natal genotypes of Brazilian peppertree, terminal buds were taken from laboratory plants in Brazil and preserved in silica gel for DNA analysis. DNA analyses were conducted by Dr. Dean Williams at Texas Christian University. Sampled plants sent for DNA analyses came from the life cycle tests in Chapter 3 and from plants used for rearing in Chapter 2 (n=22). The plants were chosen based on their health and presence of leaf flush. The sampled plants were categorized as either a success (females laid eggs, instars developed and emerged as adults) or failure (no eggs were laid or psyllids failed to develop into adults). A G test of independence as outlined by Sokal & Rohlf (1995) was used to test the hypothesis with an = 0.05. Results and Discussion The results from the D NA analysis are shown in Table 4 1. The haplotype of the plants were unknown prior to the laboratory rearing experiments in Brazil of the psyllids. Few replicates were used for some of the haplotypes (K, D, and M) and no type B plants were included making it difficult to draw clear conclusions regarding the possible suitability of Florida genotype plants. Table 4 1. Success/failure percentages of Brazilian peppertree haplotype plants used for rearing C. terebinthifolii at LAMPF. Brazilian peppertree cpDNA haplotype N Success Failure % Success A 12 8 4 75% O 5 1 4 20% K 2 0 2 0% D 2 0 2 0% M 1 0 1 0%
59 From this test sample, psyllids only developed to adults on haplotype A and haplotype O plants with the best performance on haplotype A plants. When a G test was conducted with haplotype A plants compared to all other haplotypes (O, K, D, and M), the psyllids perf ormed significantly better on A than other types (G=7.63; P <0.01). It is noteworthy that type O is most closely related to haplotype B (Figure 4 1). If C. terebinthifolii is indeed locally adapted to haplotype A, it may have trouble developing on the nov el genotypes in Florida created by the hybridization of genotypes. In Brazil, these two haplotypes do not occur sympatrically (Williams et al. 2005). However, if C. terebinthifolii from coastal Santa Catarina, Brazil can develop on both haplotypes A and O, it may be possible for it to develop on haplotype B as well. With these preliminary data, it cannot be definitively stated that C. terebinthifolii is locally adapted to specific Brazilian peppertree genotypes, but the results suggest it is a reasonable possibility and needs to be explored further.
60 CHAPTER 5 ECOLOGICAL NICHE MOD ELING Objective 4 Use ecological niche modeling to predict climatic overlap of C. terebinthifolii in Florida and predictions for future surveys in South America for C. te rebinthifolii. Matching climatic conditions required by the biocontrol agent for optimum development will help ensure the establishment and success of the biocontrol agent (Geiger & Gutierrez 2000) Heteropsylla cubana Crawford performed poorly on Leucae na leucocephala (Lam.) De Wit (Fabaceae) in a warm valley in Northern Thailand compared to a cool highland area where it was able to cause significant shoot damage (Geiger & Gutierrez 2000). Ecological niche modeling is a t ool that can predict the fundame ntal niche (where a species could occur) of a species based on climatic realized niche (where a species does occur) (Phillips et al. 2006). Ecological niche modeling can be a ccomplished by using one (or more) of the numerous algorithm available (Wisz et al. 2008) Two models frequently used include the maximum entropy species distribution model (MaxEnt) and genetic algorithm for rule set production (GARP). When MaxEnt was compared to other presence only modeling programs (such as GARP), MaxEnt was less sensitive to sample sizes and had the best predictive power across all sample sizes (Wisz et al. 2008 ; Costa et al. 2010). I chose to use MaxEnt which predicts the ecologic al niche of a species using known occurrence latitude/longitude point locations of the species and environmental climatic parameters (Phillips 2006). Using the input data for the species, MaxEnt can generate a prediction of geographical locations climatic ally suitable for the species.
61 Materials and Methods The rationale for developing an e cological n iche model was two fold : 1) to generate a predictive map using known C. terebinthifolii locations in Brazil projected onto Florida to predict the geographi c range of the psyllid if it were to be introduced into Florida and 2) to generate a predictive map to use for future collection surveys in Brazil for C. terebinthifolii adapted to Florida genotypes of Brazilian peppertrees. Two datasets were used for th e ecological niche modeling. The first data set included the geo locations of C. terebinthifolii in its native South American range (Figure 2 1, Appendix C ) which consisted of 47 points obtained from my su rveys, locations pr ovided by Dr. Greg Wheeler and M r. Fernando Mc Kay (personal communication), and from Burckhardt & Basset (2000) Google Earth TM (http://earth.google.com/) was used to obtain coordinate s for the locations provided by M r. McKay and from Burckhardt & B asset (2000). The second data set included 50 point locations o f Brazilian peppertree ( haplo type A) found in Florida prov ided by Dr. William Overholt (Appendix D) Two bio climate (bioclim) variables obtained from the WORLDCLIM database ( http://www.worldclim.org ) were used to construct niche The climate layers were generated using average climate data collected from weather stations worldwide (Hijmans et al. 2005) ESRI 2.5 arc minute resolution grids of cu rren t climate data were used for the model provided by Environmental Systems Research Institute, Inc. (ESRI, Redlands, CA) The two bioclim variables were included, bio14 ( precip itation during the driest month ) and bio 4 ( temper ature seasonality standard deviation of monthly temperature ) Data partitioning is a technique used to provide test points for model verification and accura cy ( Phillips et al. 2006) Both niche models generated in MaxEnt were
62 statistically verified by running 10 random partit ions of the dataset with 8 0% of the points as training data to generate the models and the remaining 2 0% as independent test data for extrinsic verification of predictive accuracy. Each training set was run in MaxEnt with the random seeded sub sampling pro cedure using 10 replications with 10 % as rand om test percentage for intrinsic model testing. The averages generated from the replicates were used for statistical analysis MaxEnt uses presence data and area to generate the model. The study area can be restricted by adding a mask layer. Two predictive maps were generated using the following mask layers : 1) predictive map for Florida using a mask layer restricted to the state boundaries of Florida and 2) predictive map for South America using a mask layer restricted to the countries C. terebinthifolii has been found (Argentina, Brazil and Paraguay) The maps were generated by adding the probabilities from the 10 training datasets. If a pixel in th e study area received a value lower than 20% of the highest predicted value then it was classified as absent. The threshold used is subjective. Values were reclas sified into 4 categories: 1 ) not suitable (absent), 2) suitable based on C. terebinthifolii occurrence data from South America 3) suitable based on Florida distribution of B razilian peppertree type A plants, and 4) overlap between 2 and 3 A minimum t raining threshold dependent one tailed binomial test was conducted to verify the models gener ated by MaxEnt to determine if the predictions perform ed significantly better than random (Phillips et al. 2006 ). The test was based on omission rate (test points falling in pixels predicted as not suitable) and Fractional Predictive Area (FPA, fraction o f pixels predicted suitable) (Phillips et al. 2006). MaxEnt generates
63 continuous predictions T he minimum training threshold (minimum value received by any training data) was used to implement the threshold dependent binomial test because the binomial te st requires the data to be binary, not continuous T he proportion of test points (n test ) predicted not suitable (out test ) was used as the extrinsic omission rate (out test /n test ) (Anderson et al. 2003) The binomial test was performed using Proc Freq in SAS 9.0 (SAS 2002) Results and Discussion Niche Modeling The two climate variables (bio14 and bio4) were selected based on a jackknife test run in MaxEnt version 3.3.2 which found that these variables contributed most (58.64%) t o the model prediction (Figure 5 1). Bio14 contributed the most to the model prediction (37.49 0.13%) with bio4 contributing the second highest percent (21.15 0.39%). Tropical psyllids are especially vulnerable to cold temperatures and drought (Hodkinson 2009). Waring and Cobb (1992) rep orted that in 73% of published studies, gall forming species reacted negatively to drought Young nymphs of several species of gall inducing psyllids often suffer high mortality before gall formation due to low humidity (Ferreira et al. 1990; personal observation). Because C. terebinthifolii reproduces continuously in Brazil, precipitation du ring the driest month (bio14) is a vital parameter to include in the model, as is reflected by high contribution to model prediction. Weins et al. (2006) found that temperature seasonality (bio4) was the most important and highly significant climate varia ble for predicting the distribution of tropical treefrogs in temperate North America.
64 Figure 5 1. Results of a jackknife test performed in MaxEnt to determine the percent contribution of bioclim variables to the model prediction using C. terebinthifo lii points from its native range. The minimum training threshold dependent binomial test results for Florida (validated using Florida type A Brazilian peppertree points) were highly significant for all the data partitions (Table 5 1; z = 3.1623 ; P = 0.000 8 ). The average FPA was 0. 448 The extrinsic omission rates were zero for all partitions. The results for the South American tests were similar (Table 5 2; z =2.5 3.2; P < 0.0001) where all data parti tions were highly significant The average FPA for the South American tests was 0.173. The extrinsic omission rates were slightly higher with an average of 4%. Overall, the models 0 5 10 15 20 25 30 35 40 Percent Contribution Bioclim Variable
65 generated by MaxEnt were accurate with predicts being significantly better than random. Table 5 1. Mini mum training th reshold dependent binomial test results of omiss ion for Florida range predictions Data Partition Fractional Predictive Area Points Not Suitable (out test ) Test Points (n test ) Extrinsic Omission Rate (out test /n test ) Z Part 1 0.406 0 10 0 3.1623** Part 2 0.410 0 10 0 3.1623** Part 3 0.356 0 10 0 3.1623** Part 4 0.475 0 10 0 3.1623** Part 5 0.463 0 10 0 3.1623** Part 6 0.517 0 10 0 3.1623** Part 7 0.419 0 10 0 3.1623** Part 8 0.506 0 10 0 3.1623** Part 9 0.413 0 10 0 3.1623** Part 10 0.510 0 10 0 3.1623** Average 0.448 0 0 ** P= 0.0008 Table 5 2. Minimum training th reshold dependent binomial test results of omiss ion for South American range predictions Data P artition Fractional Predictive Area Points Not Suitable (out test ) Test Points (n test ) Extrinsic Omission Rat e (out test /n test ) Z Part 1 0.222 0 10 0.0 3.1623** Part 2 0.058 1 10 0.1 2.5298** Part 3 0.078 1 10 0.1 2.5298** Part 4 0.232 0 10 0.0 3.1623** Part 5 0.237 0 10 0.0 3.1623** Part 6 0.255 0 10 0.0 3.1623** Part 7 0.260 0 10 0.0 3.1623** Part 8 0.077 1 10 0.1 2.5298** Part 9 0.233 0 10 0.0 3.1623** Part 10 0.075 1 10 0.1 2.5298** Average 0.173 0.400 0.04 ** P< 0.0001
66 Calophya terebinthifolii N ich e P rediction in Florida The prediction of the potential distribution of C. terebinthifolii in Florida is shown in Figure 5 2. The prediction has C. terebinthifolii occurring in a few coastal regions along the panhandle, west coast, and along the east coast in Florida. Areas of overlap where Brazilian peppertree haplo type A plants were known occur and predi cted to be suitable for establishment of C. terebinthifolii include the following counties: Volusia coastal Pasco and Herna n do, and a small section of southwestern Polk These counties should be target ed counties for planned releases of C. terebinthifolii if it is approved for use as a biological control agent.
67 Figure 5 2. Map of predicted climatic suitabili ty for C. terebinthifolii Brazilian peppertree haplo type A, and their predicted overlap in Florida Yellow squares (n=50 ) indicate locations of Brazilian peppertrees identified as haplotype A through DNA analysis. Future Survey S ites in South America for C. terebinthifolii Using the known locations of C. terebinthifolii a map (Figure 5 3) was generated in MaxEnt to predict areas suitable for the psyllid This area can be used to better target futur e surveys. The ma p for South America predicts C. terebinthifolii to occur in the following countries : 1) southern Chile, 2) central Bolivia, 3 ) southeaster n Paraguay, 4) Uruguay, and 5 ) southern Brazil in the states of Rio Grande do Sul, Santa Catarina,
68 P aran and coastal S o Paulo. These areas (with the exception of Chile and Bolivia ) would be good locations to s urvey for C. terebinthifolii because the full native range for the psyllid is not known. If the psyllid is locally adapted to different Brazilian peppertree genotypes psyllid s used for biological control in Florida need to come from areas that have similar genotypes of those found in Florida Using the haplotype A points from Florida, a climate prediction for the potential occurrence of haplo type A plants (with the same climatic conditions as Florida) in South America was overlaid with the habitat prediction for C. terebinthifolii The locations that fit both the climatic requirements for Florida types of Brazilian peppertree haplo type A and C. terebinthifoli i are scattered and their distributions are disjunct. The areas include : 1) southern Chile, 2) coastal Uruguay, 3) northern Rio Grande do Sul, BR 4) coastal Santa Catarina and Paran BR, and 4) scattered throughout Santa Catarina an d Paran extending into Paraguay. However, Brazilian peppertree has not been reported to occur in Chile or Bolivia
69 Figure 5 3. Map of predicted climatic suitability for C. terebinthifolii Florida Brazilian peppertree haplo type A, and their predicted overlap in South America Yellow squares (n= 47 ) indicate locations of C. terebinthifolii The two prediction maps provided here are severely constrained by the lack of data points for C. terebinthifolii Consequently, the maps generated should be considered preliminary One of the assumptions made when generating niche models is that the native range of the species is known and is accounted for in the point locations (Phillips et al. 2006). The full native range of C. terebinthifolii i s unknown As
70 more information about the native distribution of the psyllid becomes available, a more accurate model can be developed. However, these preliminary models are reasonable for predicting locations for future surveys in South America and poten tial habitats in Florida where the psyllid could be released if it were determined to be an effective biocontrol agent.
71 CHAPTER 6 BRAZILIAN PEPPERTREE MANAGEMENT USING C. TEREBINTHIFOLII Objective 5 Develop recommendations or considerations for in tegrating current Brazilian peppertree management practices with biocontrol e fforts using C. terebinthifolii. Many factors are involved in the successful introduction of a biological control agent. Insects may behave differently in a natural setting vers us a laboratory environment. When controlling an invasive plant species, the overall management plan also needs to be considered to determine how the control agent will perform in conjunction with other control methods. Considerations In addition to t he climatic differences between the native range of the psyllid and the introduced range of Brazilian peppertree in Florida, other factors need to be considered before the psyllid should be recommended as a biological control agent. Synchronous phenology is an important factor in plant herbivore interactions. Brazilian peppertree in Florida produces new foliage more or less continuously most of the year except for a cycle of dormancy from October to December when it is flowering (Ewel et al. 1982). In Br azil, growth of new foliage is limited during the flowering period, making it difficult to find adult or juvenile psyllids (personal observation). Because the phenology of Brazilian peppertree in Florida appears to be similar to that in its native range o f Brazil ( although the season are reversed), the psyllid should be able to complete its life cycle in Florida. In the field, psyllids were observed in a range of different habitats. For instance, psyllids were found along freeways with high winds and ai r pollution and also were
72 collected near the coast exposed to full sun, wind, salt spray, and dust from unpaved roads. In Brazil, Brazilian peppertrees were observed growing along roadsides and waterways similar to what is seen in Florida except that in B razil it does not grow as a monoculture (personal observation). The Brazilian peppertrees examined in the field in Brazil were adapted to various growing conditions and were effectively colonized by C. terebinthifolii One of the growing conditions that Brazilian peppertree may be able to tolerate better than the psyllid is drought (Nilsen & Muller 1980). It is important to consider how the psyllid will interact with other biocontrol agents being researched or slate d for release on Brazilian peppertree. It is possible that competition between agents may result in reduced suppression of a target weed ( McEvoy & Coombs 1999). Three agents are currently being researched at the University of Florida; a thrips Pseudophil othrips ichini Hood (Thysanoptera: Phlaeothripidae), a weevil Apocnemidophorus pipitzi Faust (Coleoptera: Curculionidae), and a leaf rolling moth Episimus unguiculus Clarke (Lepidoptera: Tortricidae). The larvae and adult thrips are usually found clustere d around the stem of new flush feeding on the plant phloem, which frequently results in death of the growing tip (Garcia 1977; Cuda et al. 2006 ; Cuda et al. 2009). The thrips feeds on the same part of the plant that is the preferred oviposition site of th e psyllid. Pseudophilothrips ichini and E. utilis are compatible with B razilian peppertree haplotypes (Manrique et al. 2008). The weevil, as an adult, is a leaf feeder and as a larva is a stem borer (McKay et al. 2009). The moth attacks the foliage of B razilian peppertree as a larva and pupates inside a rolled up leaf (Martin et al. 2004). There is potentially less overlap between the psyllid and the weevil or leaf rolling moth for preferred ovipositing and feeding sites.
73 A synergistic effect was rec ently observed between insect herbivory and plant parasitism negatively affecting the performance of Brazilian peppertree (Manrique et al. 2009 ). The native love vine Cassytha filiformis L. is widely distributed in south central Florida and can limit the growth and reproduction of Brazilian peppertree (Burch 1992). When the vine was combined with E. unguiculus in a laboratory experiment the performance of Brazilian peppertree was significantly decreased Combining control techniques such as this can lead to better control and suppression of Brazilian peppertree and should be considered when releasing biological control agents. Recommendations Recommendations can be used by researchers and land manag ers when considering the use of a biological control agent. The recommendations are based on field observations, literature review, and prediction models. The following are a few recommendations for using C. terebinthifolii as a b iological control agent : 1. Collect psyllids from coastal Santa Catarina, Brazil as they may be most compatible with Florida haplotype A plants. 2. Choose release points in Florida that are climatically similar to Brazil and where haplotype A plants occur. DNA of initial release plan ts should be tested to ensure the correct haplotype. If the insect is approved for field release, priority should be given to the following counties based on the MaxEnt model: Volusia, coastal Pasco and Hernando, and a small section of southwestern Polk. 3. C oordinate releases with appropriate agencies. If the planned release is in a managed natural area, it is important to coordinate with the land managers regarding prescribed burns, vegetation removal, or herbicide treatments. This is
74 especially important if a roadside release is planned because roadside maintenance crews routinely trim vegetation. It would be counterproductive to release psyllids that require over 30 days to complete their life cycle if the area is scheduled to be burned or cleared in the short term. It would be more appropriate to release the psyllids after trimming or burning when new shoots begin to develop and suckers are abundant 4. Release the psyllids during times when the Brazilian peppertrees have a large amount of new leaf flush. This tends to occur from May to August during the period that trees do not have flowers or fruits (Ewel et al. 1982). 5. If other biological control agents are approved for release, avoid releasing the psyllid in clos e proximity to them until it can be confirmed that competition will not negatively affect the establishment of the psyllid.
75 CHAPTER 7 CONCLUSION The results of my research provide baseline data for evaluating C. tereb inthifolii as a candidate for biological control of Brazilian peppertree Although the research is preliminary and numerous areas need further investigation th ere are several indicators that suggest the psyllid may not be a good biological control agent. As of 2005, no gall inducing hemipterans had been used as biocontrol agents and overall cecidogenous arthropods (with a few exceptions) have not been successful agents despite their high specificity and potential to cause severe plant tissue damage (Muniappan & McFadyen 2005). In th eir review of gall inducing biocontrol agents, Muniappan and McFadyen (2005) propose three reasons for the failure of this group t o produce significant results. First, the often extreme host specificity sometimes to the level of local adaptation to plant genotypes, may result in their inability to develop on subspecies or strains of plants which occur in the invaded range (Muniappan & McFadyen 2005) The results of my research showed C. terebinthifolii is potentially locally adapted (see Chapter 4 ) If this is indeed the case C. terebinthifolii may be too selective to survive on the unique genotypes of Brazilian peppertree found in Florida A second reason cecidogenous arthropods have not been used extensively in biological control is because they are often difficult to rear i n quarantine laboratories (Muniappan & McFadyen 2005) Although I succeeded in rearing the psyllid on pott ed plants in Brazil, I experienced several problems when attempting to rear t he insects in Florida. F actors that may ha ve contributed to this failure included space constraints, plant health, and the low number of live insects received from Brazil Long transit times between Brazil and reception of the insects in Florida resulted in extremely high
76 mortality. As gall form ers, it is also difficult to maintain healthy plants throughout the life cycle of the psyllid. Unlike other insects, if the plant health deteriorates, the psyllids cannot be moved to another plant. The final and most common reason gall forming insects o ften fail as biocontrol agents is attack by parasitoids indigenous to the area of introduction ( Muniappan & McFadyen 2005) Gall inducing arthropods are typically attacked by generalist parasitoids, which can transfer to exotic s pecies. However, when par asitism is below 30% for most of the reproductive season, gall inducing arthropods have been effecti ve (Muniappan & McFadyen 2005 ; Impson et al. 2009). It is unknown at this point whether C. terebinthifolii would be attack ed by native or naturalized parasitoids in Florida. However, there are two parasitoids that could potent ially parasitize the psyllid. Tamarixia sp. (Eulophidae) was introduced to control the Peruvian peppertree psyllid C. schini in California (Hagen & Ta ssan 1993) and another, Tamarixia radiat a Waterston was introduced in Florida to control the Asian citrus psyllid (Michaud 2002) Although gall forming psyllids have not been used as biological control agents, four non gall forming species of psyllids have been released in the field (Center 2007) with a fifth one recently approved for release in the United Kingdom (CABI 2010) with positive results (see Chapter 1). It is unclear why psyllids are not used more often in classical biological control progra ms. Before using the psyllid, C. terebinthifolii further research is needed in a number of areas: 1) Reasons for inability to rear the psyllids in a laboratory setting in Florida 2) Further testing to determine if the psyllid exhibits differences in performanc e on the various Brazilian peppertree genotypes
77 3) Expand exploration in native range. More source locations of C. terebinthifolii would lead to a better prediction about its climatic limitations and ability to survive in Florida. 4) Effect of C. terebinthifo lii on the performance of Brazilian peppertree. Preliminary data (Vitorino unpublished data) suggest that psyllids reduced the biomass of seedlings but it is unknown what density of galls would be required to negatively affect the growth of medium to larg e sized Brazilian peppertrees 5) Determine f ecundity, survivorship, rate of increase, lower and upper temperature development thresholds with larger sample sizes. 6) Host range testing of C. terebinthifolii on ecologically and economically relevant plants in Florida the southeastern US, and the Caribbean. 7) Ability of C. terebinthifolii to disperse or coloni ze. In a preliminary study conducted in Brazil, the psyllids took 6 months to colonize 10 plants placed along a 50 m transect downwind from a source popula tion of the psyllids (Barbieri 2004). More information is required for optimization of a release strategy. 8) Exp lore competitive interactions among candidate biocontrol Overall, the results of this research suggest C. terebinthifolii may not be an effect ive biological control agent for Brazilian peppertree However, issues surrounding rearing and Brazilian peppertree genotypes need to be resolved before being eliminated as a candidate for biocontrol of Brazilian peppertree.
78 APPENDIX A FIELD COLLECTION SITES AND DESCRIPTIONS Site # Coordinates BP Haplotype Location Description 2009 01 I, L, S 26.7800 48.6001 A Address: Rua Joao De Souza Costa 331, Praia Grande, Penha, SC, BR This location has two large trees and it is at uninhabited residential house on a residential street. One tree is located near the road; the other is in the middle of the yard. The one in the middle of the yard has been trimmed producing lots of new growth near the main trunk where the psyl lids flourish. Insects collected from all directions. No seedlings present. 2009 002 I, L, S 26.9972 48.5900 Praia de Laranjeiras, Balnerio Cambori, SC, BR The location is on the beach. It is a large tree in front of a restaurant. When you enter the beach from the main road, the tree is south. No seedlings present. Very windy from the ocean breeze with high amounts of salt. Insects only collected from the north, east, and south of the tree (no access to the west). No seedlings present. 2009 003 I 26.94739 48.63035 Intersection of Rua Delfin Mario and Rua Dulo Furlan, Praia Brava, Itaja, SC, BR Residential area the trees line the outside of the property and are medium sized trees. This location is across the street from the constructio n of a new apartment complex. A few streets away from the beach. Insects collected from the west (no access to the east). No seedlings present. Trees are subject to trimming. 2009 004 I, L, S 26.9447 48.6297 A Intersection of Rua Renato Melim Cunha and Rua Eliziario da Rosa, Praia Brava, Itaja, SC, BR Residential area the trees line the outside of the property and are medium sized trees. These trees line an undeveloped piece of property. Lots of new development happening on the beach, not sure how long these empty lots will remain empty. Insects collected from the west and north (no access to the south and east). No seedlings present. Trees are subject to trimming. 2009 005 L 26.9350 48.6275 A Dirt road that runs right along the beach. There is a fence row of the medium sized BP plants. Insects collected from the west. No seedlings present. Trees are subject to trimming. 2009 006 L, S 26.9210 48.6403 A Rua Dep. Francisco Evaristo Canziani, Itaja, SC, BR Two lane paved road is high up in the rocky cliffs overlooking the ocean. Very windy, the BP plants are large trees. Insects collected from the east (no access to the west). No seedlings present. Trees are subject to trimming. 2009 007 L, S 27.0323 48.5827 Rua Napoleo Vieira, Praia do Estaleiro, SC, BR (next to Restaurante Estrela do Atlntco) Dirt road with houses near the beach with two large BP trees in an abandoned lot. One tree is close to the street, the other tree is furth er off the road. No seedlings present. 2009 008 L, S 27.0324 48.5879 A Near Rua Domingos Fonseca, Praia do Estaleiro, SC, BR Abandoned property where a large BP tree fell over a while ago
79 but is still growing. No seedlings present. No seedlings present. 2009 009 L 27.0345 48.5842 A Rua Domingos Fonseca, Praia do Estaleiro, SC, BR Abandoned property (corner lot), the BP trees are medium sized and follow the fence row around the property. Insects collected from the east and south (no access from the west or north). No seedlings present. 2009 010 L, S 26.9110 48.9362 LAMPF, Gaspar, SC, BR Location at the FURB lab station in Gaspar of the coastal variety with potted seedlings underneath. 2009 011 I, L 26.9978 48.5904 A Praia de Laranjeiras, Balnerio Cambori, SC, BR On t he dirt road as you enter the main entrance to the beach the plants with galls are located on the south side of the street near the fence line. There are 2 medium sized BP trees. There are other BP trees along the road but adults/galls were rarely seen o n these other trees. No seedling present. Trees subject to trimming. 2009 012 I, L 27.0528 48.5904 A Intersection of Rua Anaor Romarioda Silva and L.A.P. Rodesindo Pavan, Praia do Estaleirinho, SC, BR Long fence row of BP trees on the south side of the street across from the school. Along the fence of the school are two medium sized trees that always contained adults. Dirt road at that ends at the beach. Trees are subject to trimming. 2009 013 I, L, S 27.0529 48.5894 Intersection of Rua Anaor Romarioda Silva and L.A.P. Rodesindo Pavan, Praia do Estaleirinho, SC, BR Long fence row of BP trees on the south side of the street across from the school Dirt road at that ends at the beach. Trees are subject to trimming. Needs to be considered as part of 2009 012. 2009 014 I 27.0529 48.5894 Intersection of Rua Anaor Romarioda Silva and L.A.P. Rodesindo Pavan, Praia do Estaleirinho, SC, BR Long fence row of BP trees on the south side of the st reet across from the school. Dirt road at that ends at the beach. No BP seedlings present. Trees are subject to trimming. Needs to be considered as part of 2009 012. 2009 015 I 27.0353 48.5839 Intersection of L.A.P. Rodesindo Pavan and Rua Virglio R. Pereira, Praia do Estal eiro, SC, BR Residential area close to the beach on an abandoned corner lot. Has two large BP trees; one on the north end of the lot, one on the south. No BP seedlings present. Trees are subject to trimming. 2009 016 L, S 27.0261 48.5901 A Rua Daniel Aastcio Fraga So Judas, Praia do Estaleiro, SC, BR next to the road. Had to climb to get to the trees (about 2.5 meters high from the road). Medium sized BP trees with very lo ng branches. Windy road that gains elevation as it leaves the beach and heads towards BR 101. No BP seedlings present. Not much development along the roadside. 2009 017 L 26.9188 48.6427 Rua Dep. Francisco Evaristo Canziani, Praia do Atalaia, Itaja, SC, BR Two lane paved road is high up in the rocky cliffs overlooking the ocean. Very windy. Very close to 2009 006 site.
80 2009 018 I, L, S 27.1497 48.6107 A BR 101 Sul (south) between km 155 156 Large BP trees along the freeway in a small isolated group. Vegetation around the trees is mowed. Pollution from vehicles and very windy. 2009 019 I, L 27.1606 48.6031 A SC 412, along the roadside towards Porto Belo, SC, BR Windy with pollution. 2009 020 I, L, S 27.1577 48.5652 Rua Pedro Reig off SC 412, Porto Belo, SC, BR Residential road, private property, containing one large BP tree with a few smaller trees. 2009 021 L 27.03269 48.58463 A Intersection of Rua Ver. Domingos Fonseca and L.A.P. Rodesindo Pavan, Praia do Estalerio, SC, BR Dirt road with 3 to 4 large BP trees in a residential area. Vacant lot. No seedlings present. 2009 022 L 26.94898 48.63001 A Intersection of Rua Caudio des. Ferreira and Rua Delfim Mrio de Pdua Peixoto, Praia Brava, Balneario Cambriou, SC, BR Location is located along a fence row along the roadside. Did not collect adults but old galls were present on the leaves. 2010 001 I, L 25.4454 49.5011 C Just off BR 277 on a road running parallel to BR 277, west of Curitiba, PR, B R Very windy roadside location with med large BP trees. Collected adults 10 Mar 2010. 2010 002 I, L 25.4548 49.4644 D BR 277 on the roadside, west of Curitiba, PR, BR The area was mowed and about 20 ft from the highway. The BP is regularly cut facilitating new growth. High wind and pollution. Adult psyllids were collected 10 Mar 2010. 2010 004 I 27.1884 48.5077 Intersection of R. Girassol and R. Coneira in Bombinhas, SC, BR Very close to the coast, very windy. Adults collected on 11Mar 2010. 2010 005 L 26.6945 48.6863 A Intersection of Rua Pirabeiraba and Av. Nereu Ramos in Piarras, SC, BR No sign of galls or psyllids 2010 006 L 26.6945 48.6863 A R. Joo F da Costa off R. Itajab, Barra Velha, SC, BR BP trees located just off a brackish water stream (mangue). Galls were present but no adults were found. 2010 007 L, I 26.6317 48.6882 A Side street off Av. Paran, Barra Velha, SC, BR In a residential area on private property. Found insects and young instars, the trees were very young and weedy without flowers at enough to spare. 2010 008 L 26.3722 48.7115 D Off BR 280, Araquari, SC, BR On a side dirt road off BR 280 near an industrial area. The trees were covered with dust from the road, no sign of galls or adults. 2010 009 L 26.3644 48.6645 D On BR 280, Ilha do Linguado, SC, BR On the side of BR 280 in roadside pull off area. Medium side BP present but no sign of gall s or adult psyllids. 2010 010 L 26.3276 48.6386 A Estr. da Gamboa, So Francisco do Sul, SC, BR Road runs along a railroad track. No sign of galls or adult psyllids. 2010 011 L 26.3314 48.6315 D Rod. SFS 330, So Francisco do Sul, SC, BR Unpaved road, vegetation very dusty, but old galls were present.
81 2010 012 L, I 26.2555 48.6431 A Alameda Dr. Nereu Ramos, So Francisco do Sul, SC, Brazil In a residential area near the road growing with other weedy shrubs. Found adults and galls but not enough to spare for preservation. 2010 013 L, I 25.5366 49.2260 C Zoolgico do Parque Iguau Along the road going into the zoo. Found adults and galls. 2010 014 L, I 25.4286 49.2315 P Just west of Curitiba, PA, BR along BR 277 Located on a road running parallel to BR 277 on the north side of the highway. The plants were not very large, smaller trees up on hill. Found adults and galls. Collected the seedlings from this location. 2010 015 L, I 25.5576 49.2315 C Zoolgico do Parque Iguau Inside the park near the parking lot to the entrance. Found adults and galls from medium size trees. 2010 016 L 26.905395 49.079549 A FURB main campus, Blumenau, SC, Brazil Found some BP trees on campus and some showed signs of galls however, no adults or live nymphs were found at the time of the survey (3/12/2010) I=insects (psyllids collected from location), L=leaf sample collected from a tree at location, S=seeds collected from the same tree as leaf sample
82 APPENDIX B FIELD COLLECTION DATES AND WEATHER PARAMETERS Weather Station: Itaja Santa Catarina, BR Latitude: 2657'01'' Longitude: 4845' 41'' Altitude: 5 m Dat e Min RH% Max RH% Avg. RH% Min Temp Max Temp Avg. Temp Min Vel ocity km/h Max Velocity km/h Avg. Velocity km/h Precipitation 24h (mm) 5/12/2009 72.00 93.00 82.50 12.00 29.80 20.90 0.00 10.80 5.40 0.00 5/15/2009 44.00 86.00 65.00 9.50 21.40 15.45 3.60 10.80 7.20 4.60 5/16/2009 52.00 95.00 73.50 4.00 22.40 13.20 7.20 10.80 9.00 0.00 5/22/2009 58.00 96.00 77.00 15.50 27.40 21.45 0.00 7.20 3.60 0.00 5/29/2009 66.00 96.00 81.00 14.00 20.20 17.10 0.00 3.60 1.80 0.00 6/2/2009 36.00 75.00 55.50 3.00 19.00 11.00 3.60 14.40 9.00 0.00 6/8/2009 65.00 97.00 81.00 6.50 22.60 14.55 3.60 10.80 7.20 0.00 6/29/2009 79.00 92.00 85.50 16.00 23.00 19.50 3.60 14.40 9.00 0.00 7/7/2009 72.00 98.00 85.00 13.00 24.80 18.90 3.60 14.40 9.00 10.30 7/13/2009 63.00 100.00 81.50 7.00 21.40 14.20 0.00 7.20 3.60 0.00 7/19/2009 75.00 100.00 87.50 12.00 21.40 16.70 0.00 10.80 5.40 0.00 8/3/2009 69.00 89.00 79.00 15.20 22.60 18.90 7.20 14.40 10.80 0.00 8/5/2009 69.00 98.00 83.50 13.00 26.00 19.50 0.00 7.20 3.60 0.00
83 APPENDIX C GPS COORDINATES FOR C. TEREBINTHIFOLII IN SOUTH AMERICA Latitude Longitude Location Source 23.6382 46.4615 Maua, So Paulo, Brazil Burckhardt & Basset 2000 24.7902 50.0116 Castro, Paran, Brazil Burckhardt & Basset 2000 25.1254 54.9591 Alto Paran, Paraguay Burckhardt & Basset 2000 25.4286 49.3796 Curitiba, Paran, Brazil L. Christ 25.4454 49.5011 Paran, Brazil L. Christ 25.4548 49.4644 Paran, Brazil L. Christ 25.5366 49.226 Curitiba, Paran, Brazil L. Christ 25.5576 49.2315 Curitiba, Paran, Brazil L. Christ 25.8951 50.2953 Paran, Brazil G. Wheeler 26.2556 48.6432 So Francisco do Sul, Santa Catarina, Brazil L. Christ 26.3314 48.6315 So Francisco do Sul, Santa Catarina, Brazil L. Christ 26.3999 54.6268 Eldorado, Misiones, Argentina F. McKay 26.5668 54.766 Montecarlo, Misiones, Argentina F. McKay 26.6309 54.1132 San Pedro, Misiones, Argentina F. McKay 26.6317 48.6882 Barra Velha, Santa Catarina, Brazil L. Christ 26.6946 48.6863 Barra Velha, Santa Catarina, Brazil L. Christ 26.78 48.6001 Penha, Santa Catarina, Brazil L. Christ 26.8101 55.0252 Puerto Rico, Misiones, Argentina F. McKay 26.911 48.9362 Gaspar, Santa Catarina, Brazil L. Christ 26.9188 48.6427 Itaja, Santa Catarina, Brazil L. Christ 26.921 48.6403 Itaja, Santa Catarina, Brazil L. Christ 26.9333 55.0667 Capiov, Misiones, Argentina F. McKay 26.935 48.6275 Itaja, Santa Catarina, Brazil L. Christ 26.9447 48.6297 Itaja, Santa Catarina, Brazil L. Christ 26.9474 48.6304 Itaja, Santa Catarina, Brazil L. Christ 26.949 48.63 Itaja, Santa Catarina, Brazil L. Christ 26.9972 48.59 Balnerio Cambori, Santa Catarina, Brazil L. Christ 26.9978 48.5904 Balnerio Cambori, Santa Catarina, Brazil L. Christ 27.0261 48.5901 Balnerio Cambori, Santa Catarina, Brazil L. Christ 27.0323 48.5827 Balnerio Cambori, Santa Catarina, Brazil L. Christ 27.0324 48.5879 Balnerio Cambori, Santa Catarina, Brazil L. Christ 27.0327 48.5846 Balnerio Cambori, Santa Catarina, Brazil L. Christ 27.0345 48.5842 Balnerio Cambori, Santa Catarina, Brazil L. Christ 27.0353 48.5839 Balnerio Cambori, Santa Catarina, Brazil L. Christ 27.0528 48.5904 Balnerio Cambori, Santa Catarina, Brazil L. Christ 27.0529 48.5894 Balnerio Cambori, Santa Catarina, Brazil L. Christ 27.1497 48.6107 Prto Belo, Santa Catarina, Brazil L. Christ 27.1577 48.5652 Prto Belo, Santa Catarina, Brazil L. Christ
84 27.1606 48.6031 Prto Belo, Santa Catarina, Brazil L. Christ 27.1884 48.5077 Bombinhas, Santa Catarina, Brazil L. Christ 27.3621 55.9009 Posadas, Misiones, Argentina F. McKay 27.3693 55.5819 Santa Ana, Misiones, Argentina F. McKay 27.4066 55.9094 Libertad, Misiones, Argentina F. McKay 27.4815 55.1235 Ober, Misiones, Argentina F. McKay 27.6337 55.4976 Cerro Azul, Misiones, Argentina F. McKay 27.9769 49.5818 Santa Catarina, Brazil G. Wheeler 28.0505 56.0167 Gobernador Virasoro, Corrientes, Argentina F. McKay 31.6653 60.7654 Santo Tom, Corrientes, Argentina F. McKay
85 APPENDIX D GPS COORDINATES BRAZILIAN PEPPERTREE HAPLOTYPE A IN FLORI DA Longitude Latitude Longitude Latitude 83.03007507 29.17666626 81.73705292 28.80274963 83.02752686 29.17676163 81.68782806 26.94575691 83.02483368 29.17625809 81.65468597 28.83793259 82.66589355 28.44402504 81.60637665 26.2699604 82.65820313 28.2062397 81.59760284 28.57048035 82.6518631 28.53580093 81.58760071 28.53269958 82.6505661 28.48753929 81.53755951 27.93568039 82.56598663 27.5781498 81.43653107 26.57951736 82.33322906 27.59872437 81.36858368 27.43455696 82.32174683 27.51235199 81.19249725 28.7881031 82.15074158 27.01672173 81.13694763 26.95058441 82.1484375 27.45969391 81.13098145 28.64427757 82.13552094 27.33484268 80.97051239 27.365242 82.04100037 26.92603302 80.96334076 29.08193016 82.04086304 26.93893242 82.03811646 26.92234993 81.95938873 27.06791687 81.95582581 27.79106522 81.94255829 27.99681091 81.91894531 27.48036766 81.91622162 27.2309494 81.88580322 27.28050041 81.87229919 28.82208252 81.85630035 27.52980042 81.85134888 27.59501648 81.83693695 27.86992073 81.83364868 26.72613335 81.79216766 27.20858383 81.79090118 26.17124939 81.79090118 26.17130089 81.79066467 27.49846649 81.78918457 26.17203331 81.77607727 26.33956909 81.77401733 26.33939934 81.76054382 26.97284698 81.75966644 26.86582184
86 APPENDIX E PSYLLID DNA SEQUENCI NG Fasta fi le provide by Dr. Dean Williams, Texas Christian University >Ct18 A ( C. terebinthifolii from southern Brazil) AAAGTATTGTGATAGCCCCAGCGAGTACT GGTAGCGCAAATAGAAGAAGTAGTGCTGTGA TAAGTAC TGATCAAACAAAAAGAGGTATT TTTTCCAGCTTATATAAGCAACTTCGTATAT TTATAATTGTAGAAATA AAGTTAATAGCT CCTAGAATGGAAGAGATCCCAGCGAGGTGGA GAGAGAAGATTGACATATCAACTG AGTAT CCTCTATGAAAGATAGAATTAGACAGTGGGG GGTAAACTGTTCAACCTGTTCCTACTCCT TGA TCAATTATTCTTCTCATAAGTAGTAGGT AAAGAGATGGAATAAGTAGCCAAAAACTTAAATTATTAAGT CGTGGGAACGCTATATCAGGAGCTCCAATTATTAGTGGTACGAGTCAATTACCGA ATCCCCCAATAA TAATTGGTATTG TTATAAAGAAAATTATAATGAAAGCGTGTGCTGTTACAATG > Ctsal001 (Psyllid 1 from Salvador, Bahia, BR) AAAGTATTGTGATAGCCCCAGCAAGTACG GGTAGAGCAAACAAGAGGAGTAGGGCTGTAA TAAGGA CTGACCAAACAAAAAGAGGTATT TTTTCTAATTTATACAAGCAACTCCGCATAT TTATAATTGTGGAAA TAAAATTAATGGCT CCTAAAATAGATGAAATACCAGCTAGATGGA GAGAGAAAATTGACATGTCTACT GAGTAT C CTCTATGAAAAATAGAGTTAGATAGTGGGG GGTAAACTGTTCATCCTGTCCCCACTCCT TG GTCAATTATTCTTCTTATAAGAAGTAGGT AAAGTGATGGAATAAGTAGTCAAAAACTTAAAT TATTAAG TCGTGGGAAGGCTATATCAG GGGCTCCAATTATTAGGGGGACAAGCCAA TTACCAAATCCTCCAATA ATGATTGGTATAG TTATAAAAAAAATTATAATGAAGGCGTGAGCTGTTACAATT >Ct sal002 (Psyllid 2 from Salvador, Bahia, BR) AAAGTATTGTGATAGCCCCAGCAAGTACG GGTAGAGCAAACAAGAGGAGTAGGGCTGTAA TAAGGA CTGACCAAACAAAAAGAGGTATTTTTTCTAATTTATACAAGCAACTCC GCATAT TTATAATTGTGGAAA TAAAATTAATGGCT CCTAAAATAGATGAAATACCAGCTAGATGGA GAGAGAAAATTGACATGTCTACT GAGTAT CCT CTATGAAAAATAGAGTTAGATAGTGGGG GGTAAACTGTTCATCCTGTCCCCACTCCTTG GTCAATTATTCTTCTTATAAGAAGTAGGTAAAGTGATGGAATAAGTAGTCAAAAACTT AAATTATTAAG TCGTGGGAAGGCTATATCAG GGGCTCCAATTATTAGGGGGACAAGCCAA TTACCAAACCCTCCAATA ATGATTGGTATAG TTATAAAAAAAATTATAATGAAGGCGTGAGCTGTTACAATT >Ctsa l003 (Psyllid 3 from Salvador, Bahia, BR) AAAGTATTGTGATAGCCCCAGCAAGTACG GGTAGAGCAAACAAGAGGAGTAGTGCTGTAA TAAGGAC TGACCAAACAAAAAGAGGTATT TTTTCTAATTTATACAAGCAACTCCGCATAT TTATAATTGTGGAAAT AAAATTAATGGCT CCTAAAATAGATGAAATACCAGCTAGATGGA GAGAGAAAATTGACATGTCTACTG AGTAT CCTCT ATGAAAAATAGAGTTAGATAGTGGGG GGTAAACTGTTCATCCTGTCCCCACTCCTTGG TCAATTATTCTTCTTATAAGAAGTAGGTAAAGTGATGGGATAAGTAGTCAAAAACTTAAATTATTAAGT CGTGGGAAGGCTATATCAGGGGCTCCAATTATTAGGGGGACAAGCCAATTACCAAAC CCTCCAATAA TGATTGGTATAG TTATAAAAAAAATTATAATGAAGGCGTGAGCTGTTACAATT >Ctsal0 04 (Psyllid 4 from Salvador, Bahia, BR) AAAGTATTGTGATAGCCCCAGCAAGTACG GGTAGAGCAAACAAGAGGAGTAGGGCTGTAA TAAGGA CTGACCAAACAAAAAGAGGTATT TTTTCTAATTTATACAAGCAACTCCGCATAT TTATAATTGTGGAAA TAAAATTAATGGCT CCTAAAATAGATGAAATACCAGCTAGATGGA GAGAGAAAATTGACATGTCTACT GAGTAT CCTCTAT GAAAAATAGAGTTAGATAGTGGGG GGTAAACTGTTCATCCTGTCCCCACTCCT TG GTCAATTATTCTTCTTATAAGAAGTAGGT AAAGTGATGGAATAAGTAGTCAAAAACTT AAATTATTAAG TCGTGGGAAGGCTATATCAG GGGCTCCAATTATTAGGGGGACAAGCCAA TTACCAAACCCTCCAATA ATGATTGGTATAG TTATAAAAAAAATTATAATGAAGGCGTGAGCTGTTACAATT
87 APPENDIX F IMAGES OF PARASITOID S Automontage images of hymenopteran parasitoids that emerged at FBCL from psyllid instars shipped from Brazil 2009 2010. Preliminary identification was done by Dr. Greg Evans, a systematic entomologist with USDA. Images A and B are from the same individual. Images A, B, and D are Eulophidae: Tetrastichinae. Images A and B are potentially Aprostocetus sp., which most species tend to be hyperparasitoids but some are primary. Image D is potentially Tamarixia sp., which tend to be primary parasitoids. Image C is in the family Signiphoridae and are hyperparasitoids. Images E and F are in the family Encyrtidae and potentially Metaphycus sp., which are primary parasitoids. A B C D
88 E F G
89 APPENDIX G PSYLLID SHIPMENTS Psyllids received at FBCL from Brazil from shipments or individuals hand carry ing the insects back in their luggage from February 2009 May 2010. Date(s) Material Collected Date Se nt Transit Time (days) Delivery Type Collection Location(s) Material Included Status at Reception 3 Feb 09 (SC) 12 Feb 09 (PR) 15 Feb 09 3 Hand carried to Miami/shipped FedEx Santa Catarina BR (interior) Paran, BR (interior) Leaf material with galls Leaf material (fair) 11 July 09 13 July 09 4 Shipped DHL Santa Catarina, BR (coastal) 18 adults Leaf material with galls 5 live adults Leaf material (good) 19 July 09 20 July 09 7 Shipped DHL Santa Catarina, BR (coastal) 40 adults Leaf material with galls 3 live adults Leaf material (poor) 3 Aug 09 3 Aug 09 10 Shipped DHL Santa Catarina, BR (coastal) 50 adults Leaf material with galls 0 live adults Leaf material (poor) 5 Aug 09 6 Aug 09 1 Hand carried Santa Catarina, BR (coastal) 50 adults Leaf material with galls 50 live adults (many instars emerged en route) Leaf material (good) 6 Dec 09 8 Dec 09 9 Shipped DHL Santa Catarina, BR (coastal) 100 adults Leaf material with galls 0 live adults Leaf material (poor) 10 14 Mar 10 14 Mar 10 2 Hand carried Santa Catarina, BR (coastal) Paran, BR (interior) 40 adults Leaf material with galls 15 live adults Leaf material (good) 24 28 Mar 10 31 Mar 10 1 Hand carried to Miami/shipped FedEx Bahia, BR (coastal) 30 adults Leaf material with galls 12 live adults Leaf material (good) 2 May 10 2 May 10 1 Hand carried Santa Catarina, BR (coastal) 4 adults 2 live adults
90 LIST OF REFERENCES [ADA] Arizona Department of A griculuture 2006. Prohibited, regulated and restricted noxious weeds. Plant Services Division. [ http://www.azda.gov/PSD/quarantine5.htm accessed 6 June 2010] ANANTHAKRISHNAN, T. N. 1984. Biology of Gall Insects. London: E. Arnold. ANDERSON, R. P., LEW D., AND PETERSON, A. T. 2003. Evaluating predictive models of species' distributions: criteria for selecting optima l models. Ecological Modelling 162 : 211 232. ANONYMOUS. 2007. Flo species. Wildland Weeds. 10:13 16. AGUILAR ORTIGOZA, C. J. AND SOSA V. 2004. The evolution of toxic phenolic compounds in a group of Anacardiaceae genera. Taxon 53:357 364. ALVAREZ ZAGOYA, R. AND CIBRIAN TOVAR, D. 1999. Biology of the peppertree psyllid Calophya rubra (Blanchard) (Homoptera: Psyllidae). Revista Chapingo Serie Ciencias Forestales y Del Ambiente. 5:51 57. AUSTIN, D. F., AND S MITH, E. 1998. Pine rockland plant g uide: A field guide to the plants Management, Miami, Florida. BARBIERI, G. 2004. Testes de potencial de dano e de especificidade com Calophya terebinthifolii Burckhardt & Basset, 2000 (Hemiptera: Psyllidae) para o controle biologic da aroeira Schinus terebinthifolius Raddi (Anacardiaceae) no estado da Flrida EUA. Masters Thesis. Universidade Regional de Blumenau. 72 p. BARKLEY, F. A. 1944. Schinus L. Brittonia 5: 160 198. BARKLEY, F. A. 1957. A study of Schinus L. Lilloa Revista de Botonica. Tomo 28. Universidad Nacional del Tucumen, Argentina. 110p. BENNETT, F. D., CRESTANA, L., HABECK, D. H., AND BERTI FILHO, E. 1990. Brazilian peppertree prospects for biological control. Rome Italy: Istituto Sperimentale per la Patologia Vegetale, Ministero dell'Agricoltura e delle Foreste. BOECKLEN, W. J. AND MOPPER, S. 1998. Local adaptation in specialist herbivores: Theory and evidence. In S. Mopper and S. Y. Strauss [eds.] Genetic Structure and Local Adaptation in Natural Insect Populations. New York: International Thomson Publishing. BRIZICKY, G. K. 1962. The genera of Anacardiaceae in the Southeastern United States. J Arnold Arbor 43:359 375.
91 BURCH, J. N. 1992. Cassytha filiformis and limits to growth and reproduction of Schinus terebinthifolius in southern Florida. Florida Scientist 55:28 34. BURCKHARDT, D. 2005. Biology, ecology, and evolution of gall inducing psyllids (Hemiptera: Psylloidea). Enfield: Science Publi shers, Inc. BURCKHARDT, D. AND BASSET, Y. 2000. The jumping plant lice (Hemiptera: Psylloidea) associated with Schinus (Anacardiaceae): systematics, biogeography and host plant relationships. J Nat Hist 34:57 155. [CABI ] Centre for Agricultural Bioscience I nternational. 2010. Japanese Knotweed Alliance. http://www.cabi.org/japaneseknotweedalliance/ [CAL IPC] California Invasive Plant Council. 2006. California Invasive Plant Inventory. [ http: //www.cal ipc.org/ip/inventory/pdf/Inventory2006.pdf accessed 22 June 2010]. CALLAWAY, M. R. DELUCA, T. H. AND BELLIVEAU, W. M. 1999. Biological control herbivores may increase competitive ability of the noxious weed Centaurea maculosa. Ecology 80:1196 1201. CAMPBELL, G. R, CAMPBELL, J. W., AND WINTERBOTHAM, A. L. 1980. The First Fund of Animals, Inc. Schinus terebinthifolius Brazil Expedition, July 1980 Interim Report. [Unpublished]. CARR, D. E., AND EUBANKS, M. D. 2002 Inbreeding alters res istance to insect herbivore and host plant quality in Mimulus guttatus (Scrophulariaceae) Evolution 56 : 22 30 CARPENTER, D. AND CAPPUCCINO, N. 2005. Herbivory, time since introduction and the invasiveness of exotic plants. Journal of Ecology 93:315 321. [CDFA] California Department of Food and Agriculture. 1984. California Plant Pest and Disease Report 3:119 121. CENTER, T. D., FRANK, J. H., AND DRAY, F. A. 1997. Biological control [in] Strangers in Paradise: Impact and Management of Nonindigenous Species in Florida. [eds.] D. Simberloff, D. C. Schmitz, and T. C. Brown. Washington, D.C.: Island Press. CENTER, T. D., PRATT, P. D., TIPPING, P. W., RAYAMAJHI, M. B., VAN, T. K., WINERITER, S. A., AND DRAY, F. A., JR. 2007. Initial impacts and field validation of host range for Boreioglycaspis melaleucae Moore (Hemiptera : Psyllidae), a biological control agent of the invasive tree Melaleuca quinquenerv ia (Cav.) Blake (Myrtales : Myrtaceae : Leptospermoideae). Environ Entomol 36:569 576. COCK, M. J. W. 1986. Requirements for biological control: an ecological perspective. Biocontrol News and Information 7:7 16.
92 COSTA, G.C., NOGUEIRA, C., MACHADO, R.B., AND COLLI, G.R. 2010. Sampling bias and the use of ecological niche modeling in conservation planning: a field evaluation in a biodiversity hotspot. Biodiversity and Conservation 19:883 899. CUDA, J.P., MEDAL, J.C. VITORINO, M.D. AND HABECK, D.H. 2005. Supplementary hos t specificity testing of the sawfly Heteroperreyia hubrichi a candidate for classical biological control of Brazilian peppertree, Schinus terebinthifolius BioControl 50: 195 201. CUDA, J. P., FERRITER, A. P, MANRIQUE, V., AND MEDAL, J. C. [eds.] 2006. Brazilian Peppertree Management plan, 2nd edition: Recommendations from the Brazilian Peppertree Task Force, Florida Exotic Pest Plant Council, April 2006. http://ipm.ifas.ufl.edu/pdf/BPmanagPlan.pdf CUDA, J.P., GILMORE, J.L. MEDAL, J.C. AND PEDROSA MACEDO J.H 2008. Mass rearing of Pseudophilothrips ichini (Thysanoptera: Phlaeothripidae), an approved biological control agent for Brazilian peppertree, Sc hinus terebinthifolius (Sapindales: Anacardiaceae). Florida Entomol. 91: 338 340. CUDA, J.P., MEDAL, J.C., GILLMORE, J.L., HABECK, D.H., AND PEDROSA MACEDO, J.H. 2009. Fundamental host range of Pseudophilothrips ichini sensu lato (Thysanoptera: Phlaeothri pidae), a candidate biological control agent of Schinus terebinthifolius (Sapindales: Anacardiaceae) in the USA. Environ. Entomol. 38: 1642 1652. ARMENTANO, T. V. 2003. Plant colonization after complete and partial removal of disturbed soils for wetland restoration of former agricultural fields in Everglades National Park. Wetlands 23:1015 1029. DEBACH, P. AND ROSEN, D. 1991. Biological Control by Natural Enemies. Cambridge: Cambridge U niversity Press. DELFOSSE, E. S. 1979. Biological control: A strategy for plant management. pp. 83 86 [in] R. Workman, [ed.] Schinus technical proceedings of techniques for control of Schinus in South Florida: A workshop for natural area managers, 2 December 1978. The Sanibel Captiva Conservation Foundation, Inc., Sanibel, FL. DEVI, S. B., AND PRABHOO, N. R. 1995. Biology of leaf gall forming psyllid Paurosylla turberculata (Homoptera). J of Ecobiology 7:75 77. DONNELLY, M. J., GRE EN, D. M., AND WALTE RS, L. J. 2008. Allelopathic effects of fruits of the Brazilian pepper Schinus terebinthifolius on growth, leaf production and biomass of seedlings of the red mangrove Rhizophora mangle and the bl ack mangrove Avicennia germinans J Exp Mar Biol Ecol 357:149 156.
93 DONNELLY, M. J. AND WALTERS, L. J. 2008. Water and boating activity as dispersal vectors for Schinus terebinthifolius (Brazilian pepper) seeds in freshwater and estuarine habitats. Estuari es and Coasts 31:960 968. DOREN, R. F., WHITEA KER, L. D., LAROSA, A. M. 1991. Evaluation of fire as a management tool for controlling Schinus terebinthifolius as secondary successional growth on abandoned agricultural land. Environ Manage 15:121 129. DOW NER, J. A., SVIHRA, P., MOLINAR, R. H., FRASER, J. B., AND K OEHLER, C.S. 1988. New psyllid pest of California USA pepper tree. Calif Agric 42:30 32. DREGER JAUFFRET, F. AND SHO RTHOUSE, J. D. 1992. Diversity of gall inducing insects and their galls. Oxford University Press, Oxford, New York etc. EDDMaps. 2009. Early detection and distribution mapping system for Brazilian peppertree in the United States [ http://www.eddmaps.org/distri bution/uscounty.cfm?sub=3521 accessed 22 June 2010 ] EDMUNDS JR., G.F. AN D ALSTAD, D.N. 1978. Coevolution in insect herbivores and conifers. Science 199:941 945 ELITH, J., GRAHAM, C. H., ANDERSON, R. P., DUDIK, M., FERRIER, S., GUISAN, A., HIJMANS, R. J., HUETTMANN, F., LEATHWICK, J. R., LEHMANN, A., LI, J., LOHMANN, L. G., LOISELLE, B. A., MANION, G., MORITZ, C., NAKAMURA, M., NAKAZAWA, Y., OVERTON, J. M., PETERSON, A. T., PHILLIPS, S. J., RICHARDSON, K., SCACHETTI PEREIRA, R., SCHAPIRE, R. E., SOBERON J., WILLIAMS, S., WISZ, M. S., AND ZIMMERMANN, N. E 2006. Novel methods 29:129 151 EVANS, G. A. 2002. Entomology section. Tri ology 41(3). http://www.doacs.state.fl.us/pi/enpp/triology/archive/02 may june.html EWE, S., AND STERNBE RG, L. 2005. Growth and gas exchange responses of Brazilian pepper ( Schinus terebinthifolius ) and native south G lorida species to salinity. Trees Struct Funct 19:119 128. EWEL, J. J. 1979. Ecology of Schinus. pp. 7 21 IN R. Workman [ed.] Schinus technical proceedings of techniques for control of Schinus in South Florida: A workshop for natural area managers, 2 December 1978. The Sanibel Captiva Conservation Foundation, Inc., Sanibel, FL. EWEL, J., OJIMA, D., KARL, D., AND DEBUSK W. 1982. Schinus in Successional Ecosystems of Everglades National Park. So uth Florida Res. Cent. Rep. T 676. Everglades National Park, National Park Service, Homestead, FL. 141 pp.
94 EWEL, J. J. AND PUTZ F. E. 2004. A place for alien species in ecosystem restoration. Frontiers in Ecology and the Environment 2:354 360. FERREIRA, S. A, FERNANDES. W. G., AND CARVALHO, L. G. 1990. Biology and natural history of Euphaleurus ostreoides (Homoptera: Psyllidae) a gall former on Lonchocarpus guilleminiaus Rev Brasil Biol. 50:417 424. [FLDACS] Florida Department of Agriculture and Consume r Services. 1999. Noxious Weed List [in] Introduction or release of plant pests, noxious weeds, arthropods, and biological control agents, Chapter 5B 57.007. Department of Agriculture and Consumer Services, Division of Plant Industry. Gainesville, Florida http://www.doacs.state.fl.us/pi/enpp/botany/noxweed.html [FLDEP] Florida Department of Environmental Protection. 1993. Chapter 62C 52.001 Prohibited aquatic plants. Pp. 1584 1585. In Florida Statutes 62C 52 Aquatic Plant Importation, Transportation, Non nursery Cultivation, Possession and Collection. http ://www.dep.state.fl.us/lands/invaspec/2ndlevpgs/perrules.htm#62C 52.011%20Prohibited%20Aquatic%20Plants 2007 list of invasive species. http://www.fleppc.org/list/07list.htm FRANK, J. H. AND MCCOY, E. D. 1992. Introduction to the behavioral ecology of immigration The immigration of insects to Florida, with a tabulation of records published since 1970. Florida Entomol ogist 75:1 28. FRANK, J. H. AND MCCOY, E. D. 1995. Introduction to insect behavioral ecology: the good, the bad and the beautiful: nonindigenous species in Florida. Florida Entomologist 78 : 1 15. TSCHINKEL, W. R. 1997. Immigration and introduction of insects [in] Strangers in Paradise: Impact and Management of Nonindigenous Species in Florida. [eds.] D. Simberloff, D. C. Schmitz, and T. C. Brown. Washington, D.C.: Island Press. FRANKS, S. J., KR AL, A. M., AND PRATT, P. D 2006. Herbivory by introduced insects reduces growth and survival of Melaleuca quinquenervia seedlings. Environ Entomol 35:366 372. GARCIA, C.A. 1977. Biologia e aspectos da ecologia e do comportamento defensive comparada de Lio thrips ichini Hood 1949 (Thysanoptera Tubulifera). M.S. Thesis, Universidade Federal do Parana.
95 GEIGER, C. A, AND GUTIERREZ, A.R. 2000. Ecology of Heteropsylla cubana (Homoptera : Psyllidae): Psyllid damage, tree phenology, thermal relations, and parasitism in the field. Environ Entomol 29:76 86. GORDON, D. R. 1998. Effects of invasive, non indigenous plant species on ecosystem processes: lessons from Florida. Ecological Applicat ions 8:975 989. HABECK, D. H., BENNETT, F. D., AND BALCIUNAS, J. K. 1994. Biological control of terrestrial and wetland weeds. In D. Rosen, F. D. Bennett, and J. L. Capinera [eds.]. Pest Management in the Subtropics: Biological Control a Florida Perspect ive. Andover, England: Intercept Ltd. HAGAN, K. S. AND TASSAN, R. L. 1996. Biological control of pepper tree psyllid. Final report. California Department of Transportation. 24p. HARRIS, P. AND J. D. SHORTHOUSE. 1996. Effectiveness of gall inducers in weed biological control. Can Ent 128:1021 1055. HIJMANS, R. J., CAMERON, S. E., PARRA, J. L., JONES, P. G., AND JARVIS, A. 2005. Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25:1965 197 8. HODKINSON, I. D. 1974. Biology of Psylloidea (Homoptera) review. Bull Entomol Res 64:325 338. HODKINSON, I. D. 1989. The biogeography of the neotropical jumping plant lice (Insecta, Homoptera, Psylloidea). J Biogeogr 16:203 217. HODKINSON, I.D. 2009. Life cycle variation and adaptation in jumping plant lice (Insecta: Hemiptera: Psylloidea): a global synthesis. Journal of Natural History 43:65 179. HOKKANEN, H. AND PIMENTEL, D. 1984. New approach for selecting biological control agents. Can Ent 116:1 109 1121. HOWARTH, F.G. 1991. Environmental impacts of classical biological control. Annual Review of Entomology 36:485 509. HUFFAKER, C. B., AND P. S. MESSENGER. 1964. The concept and significance of natural control. In P. DeBach [eds.] Biological control of insect pests and weeds. Reinhold, New York. JUDD, W. S., CAMPBELL, C. S., KELLOGG, E. A., STEVENS, P. F., AND DONOGHUE, M. J. 2008. Plant Systematics: A Phylogenetic Appoach. Massachusetts USA, Sinauer Associates, Inc. 611p. JULIEN, M. H., KERR, J. D., AND CHAN, R R. 1984. Biological control of weeds an evaluation. Protection Ecology 7:3 25.
96 JULIEN, M. H., SCOTT, J. K., ORAPA, W., AND PAYNTER, Q. 2007. History, opportunities and challenges for biological control in Australia, New Zealand and th e pacific islands. Crop Prot 26:255 265. KARBAN, R., AND STRAUSS, S. Y., 1994. Colonization of new host plant individuals by locally adapted thrips. Ecography 17:82 87. KIMURA, M. 1980 A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16 : 111 120. KUNIATA, L. S. AND KOROWI, K. T. 2004. Bugs offer sustainable control of Mimosa invisa and Sida spp. in the Markham Valley, Papua New Guinea. Canberra Australia: CSIRO Entomology. LIU, H. AND STILING, P. 2006. Testing the enemy release hypothesis: A review and meta analysis. Biological Invasions 8:1535 1545. C. W. 2002 Unexpected ecological effects of distributing the exotic weevil, Larinus planus (E), for the biological control of Canada thistle. Conservation Biology 16:717 727. LOUDA, S. M. AND STILING, P. 2004. The double edged sword of biological control in conservation and restoration. Conservation Biology 18:50 53. MACK, R. N. 1991. The commercial seed trade: And early disperser of weeds in the United States. Econ Bot 45:257 273. MANRIQUE, V., CUDA, J. P., OVERHOL T, W. A., WILLIAMS, D. A., AND WHEELER, G. S. 2008. Effect of host plant genotypes on the performance of three candidate biological control agents of Schinus terebinthifolius in Florida. Biol Control 47:167 171. MANRIQUE, V., CUDA, J. P., OVERHOLT, W. A., AND EWE S. M. L. 2009. Synergistic effect of insect herbivory and plant parasitism on the performace of the invasive tree Schinus terebinthifolius Entomologia Experimentalis et Applicata 132:118 125. MARTIN, C.G., CUDA, J.P. AWADZI, K.D. MEDAL J. C. HABECK, D.H. AND PEDROSA MACEDO, J.H. 2004. Biology and laboratory rearing of Episimus utilis (Lepidoptera: Tortricidae), a candidate for classical biological control of Brazilian peppertree, Schinus terebinthifolius (Anacardiaceae). Florida Entomol. 33: 1351 1361. MCEVOY, P. B. AND COOMBS, E. M. 1999. Biological control of plant invaders: regional patterns, field experiments, and structured population models. Ecological Applications 9:387 401.
97 MCKAY F., OLEIRO, M., WALSH, G. C., GANDOLFO, D., CUDA, J. P., AND WHEELER, G. S. 2009. Natural enemies of Brazilian peppertree (Sapindales: Anacardiaceae) from Argentina: Their possible use for biological control in the USA. Florida Entomologist 92:292 303. MEDAL, J. C, VITORINO, M. D., HABECK, D. H., GILL MORE, J. L., PEDROSA, J. H., AND DESOUSA, L. D. 1999. Host specificity of Heteroperreyia hubrichi Malaise (Hymenoptera: Pergidae), a potential biological control agent of Brazilian Peppertree ( Schinus terebinthifolius Raddi). Biol Control. 14: 60 65. MESS ING, R. H. AND WRIGH T. M. G. 2006. Biological control of invasive species: solution or pollution?. Frontiers in Ecology and the Environment 4:132 140 MOERI, O.E., CUDA, J .P., OVERHOLT, W. A. BLOEM, S., AND CAR PENTER, J. E. 2009. F1 sterile i nsect techniq ue: A novel approach for risk a ssessment of Episimus unguiculus (Lepidoptera: Tortricidae), a candidate biological control a gent of Schinus terebinthifolius in the Continental USA BioControl Sci. & Tech. 19: 303 315. MORGAN, E. C., AND O VERHOLT, W. A 2005. Potential allelopathic effects of Brazilian pepper ( Schinus terebinthifolius raddi, Anacardiaceae) aqueous extract on germination and growth of selected Florida native plants. J Torrey Bot Soc 132:11 15. MORTON, J. F. 1978. Brazilian pepper its imp act on people, animals and the environment. Econ Bot 32:353 359. MOUND, L.A., WHEELER G.S., AND WILLIAMS D.A. 2010. Resolving cryptic species with morphology and DNA; thrips as a potential biocontrol agent of Brazilian peppertree, with a new species and overview of Pseudophilothrips (Thysanoptera). Zootaxa 2432:59 68. MUNIAPPAN, R. AND MC FADYEN, R E. 2005. Gall inducing arthropods used in the biological control of weeds. Enfield: Science Publishers, Inc. MURDOCH, W. W., CHES SON, J., AND CHESSON P. L. 19 85. Biological control in theory and practice. American Naturalist 125:344 366 MYERS, R. L. AND EWE L J. J. 1990. Ecosystems of Florida. Orlando, University of Central Florida Press. 765p. NILSEN, E.T. AND MUL LER W.H. 1980. A comparison of the relativ e naturalizing ability of two Schinus species (Anacardiaceae) in southern California. II Seedling establishment. Bulletin of the Torrey Botanical Club 107:232 237.
98 PALMER, W. A. AND WITT, A. B. R 2006. On the host range and biology of Acizzia melanocephal a (Hemiptera : Psyllidae), an insect rejected for the biological control of Acacia nilotica subsp indica (Mimosaceae) in Australia. Afr Entomol 14:387 390. PEARSON, D. E., MCKELVEY, K. S., AND RUGGIERO, L. F. 2000. Non target effects of an introduced bio logical control agent on deer mouse ecology. Oecologia 122:121 128. PEARSON, D. E. AND CALLAWAY, R. M. 2006. Biological control agents elevate hantavirus by subsidizing deer mouse populations. Ecology Letters 9:443 450. PEDROSA MACEDO, J.H., POULMA NN, D., STOLLE, L. U KAN, D., CUDA, J.P. AND MEDAL J.C 2006. Greenhouse mass rearing of a defoliating sawfly for biological control of Brazilian peppertree. Floresta 36: 371 378. PEMBERTON, R. W. 2000. Predictable risk to native plants in weed biological control. Oecologia 125:489 494. PETERSON, M.A. AND DENNO, R. F. The influence of dispersal and diet breadth on patterns of genetic isolation by distance in phytophagous insect s. The American Naturalist 152:428 446. PHILLIPS, S.J., ANDERSON, R.P., AND SCHAPIRE R.E. 2006. Maximum entropy modeling of species geographic distributions. Ecological Modeling 190:231 259. PIMENTEL, D., ZUNIGA, R., AND MORRISON, D. 2005. Update on the environmental and economic costs associated with alien invasive species in the United. Ecological Economics 52: 273 288. PRICE, P. W. 1997. Insect Ecology. 3rd ed. New York: Wiley. RAMAN, A., SCHAEFER, C. W., WITHERS, T. M. 2005. Biology, ecology, and evolution of gall inducing arthropods. Enfield, NH: Science Publishers. RAYAMAJHI, M. B., PRATT, P. D., CENTER, T. D., TIPPING, P. W., VAN, T. K 2008. Aboveground biomass of an invasiv e tree melaleuca ( Melaleuca quinquenervia ) before and after herbivory by adventive and introduced natural enemies: A temporal case study in Florida. Weed Sci 56:451 456. RAYAMAJHI, M. B., VAN, T. K., PRATT, P. D., CENTER, T. D., AND TIPPING, P W. 2007. Me laleuca quinquenervia dominated forests in Florida: Analyses of natural enemy impacts on stand dynamics. Plant Ecol 192:119 132. REBELO, H. AND JONES G. 2010. Ground validation of presence only modeling with rare species: a case study on barbastelles Barb astella barbastellus (Chiroptera: Vespertilionidae). Journal of Applied Ecology 47:410 420.
99 RHAINDS, M., AND SHIPP, L ., 2003. Dispersal of adult western flower thrips (Thysanoptera: Thripidae) on chrysanthemum plants: impact of feeding induced senescence o f inflorescences. Environmental Entomology 32:1056 1065. RHAINDS, M., SHIPP, L., WOODDROW, L., AND ANDERSON, D., 2005. Density, dispersal, and feeding impact of western flower thrips (Thysanoptera: Thripidae) on flowering chrysanthemum at different spatial scales. Ecological Entomology 30:96 104. ROHFRITSCH, O. 1992. Patterns in gall development [in] Biology of insect ind uced galls. [eds.] O. Rohfritsch and J. D. Shorthouse. New York: Oxford University Press. ROSKAM JC. 1992. Evolution of the gall inducing guild[in] Biology of insect induced galls. [eds.] O. Rohfritsch and J. D. Shorthouse. New York: Oxford University Pres s. SAX, D.F. AND GAINES, S. D. 2008. Species invasions and extinctions: The future of native biodiversity on islands. Proceedings of the National Academy of Sciences 105:11490 11497. SCHMITZ, D. C. 2007. Florida's invasive plant research: Historical perspec tive and the present research program. Nat Areas J 27:251 253. SHEPPARD, A.W., VAN KLINKEN, R.D., AND HEARD, T. A. 2005. Scientific advances in the analysis of direct risks of weed biological control agents to nontarget plants. Biol. Control. 35: 215 226. SIMBERLOFF, D., SCHMITZ, D. C., AND BROWN T. C. 1997. Strangers in Paradise: Impact and Management of Nonindigenous Species in Florida. Washington, D.C.: Island Press. SMITH, A. C 1985. Flora Vitiensis nova: a new flora of Fiji. National Tropical Botanic al Garden, Lawai, Kauai, Hawaii. Volume 3. 758 p. SOKAL, R.R. AND F.J. ROHLF. 1995. Biometry: The principles and practice of statistics in biological research. New York, W.H. Freeman. STEVENS, J. T. AND BECKAGE B. 2009. Fire feedbacks facilitate invasion of pine savannas by Brazilian pepper ( Schinus terebinthifolius ). New Phytologist 184:365 375. STRAUSS, S. Y. AND KARBAN R. 1998. The strength of selection: Intraspecific variation in host plant quality and th e fitness of herbivores. In S. Mopper and S. Y. Strauss [eds.] Genetic Structure and Local Adaptation in Natural Insect Populations. New York: International Thomson Publishing.
100 TASSIN, J., RIVIERE, J., AND CLERGEAU P. 2007. Reproductive versus vegetativ e recruitment of the invasive tree Schinus terebinthifolius : Implications for restoration on Reunion Island. Restoration Ecology 15:412 419. TISDELL, C. A., AULD, B. A., AND MENZ, K M. 1984. On assessing the value of biological control of weeds. Protection Ecology 6:169 179. TSOAR, A., ALLOUCHE, O., STEINITZ, O., ROTEM, D., AND KADMON R. 2007. A comparative evaluation of presence only methods for modeling species distribution. Diversity and Distributions 13:397 405. [USFS] United States Forest Service. 2010. Pacific Island Ecosystems at Risk (PIER). http://www.hear.org/pier. VITORINO, M. D., MARAN, R. D., CUDA, J. P., AND MEDAL, J. C. 2008. Mass rearing of Calophya terebinthifolii (Hemiptera: Psyllidae) on Brazilian peppertree. Proceedings o f the XII international symposium on biological control of weeds, La Grande Motte, France, 22 27 April, 2007. Wallingford UK: CAB International. 589 p. WARING G. L. AND COBB N. S 1992. The impact of plant stress on herbivore population dynamics [in] P lant Insect Interactions [ed.] E. A. Bernays. Boca Raton, Florida:CRC Press. WEIS, A. E., WALTON, R., AND CREGO, C. L. 1988 Reactive plant tissue sites and the population biology of gall makers Annual Review of Entomology 33 : 467 486 WIENS, J.J., GRAHAM. C.H., MOEN, D.S., SMITH, S.A., AND REEDER, T.W. 2006. Evolutionary and ecological causes of the latitudinal diversity gradient in Hylid frogs: treefrog trees unearth the roots of high tropical diversity. The American Naturalist 168:579 596. WILCOVE, D.S., ROTHSTEIN, D., DUBOW, J., PHILLIPS, A., AND LOSOS E. 1998. Quantifying threats to imperiled species in the United States. BioScience 48:607 615. WILLIAMS, D. A, MUCHUGU, E., OVERHOLT, W. A., AND CUDA J. P. 2007. Colonization patterns of the invasive Brazilian peppertree, Schinus terebinthifolius in Florida. Heredity 98:284 293. WILLIAMS, D. A., OVERHOLT, W. A., CUDA, J. P., AND HUGHES, C. R. 2005. Chloroplast and microsatellite DNA diversities revea l the introduction history of Brazilian peppertree ( Schinus terebinthifolius ) in Florida. Mol Ecol 14:3643 3656. WILLIAMS, JR 1954. The biological control of weeds. In: Report of the Sixth Commonwealth Entomological Congress. London, UK, pp. 95 98.
101 WISZ, M.S., HIJMANS, R.J., LI, J., PETERSON, A.T., GRAHAM, C.H., GUISAN, A., AND NCEAS PREDICTING SPECIES DISTRIBUTIONS WORKING GROUP 2008. Effects of sample size on the performance of species distribution models. Diversity and Distributions 14:763 773. WOLFE, L. M. 2002. Why alien invaders succeed: Support for the escape from enemy hypothesis. The American Naturalist 160:705 711.
102 BIOGRAPHICAL SKETCH Lindsey Christ is an Ohio native having spent most of her childhood in Columbus, Ohio. She attended college at Northland College in northern Wisc onsin but finished her Bachelor of Science at the University of North Carolina at Asheville in environmental s tud ies in 2000. Lindsey work ed as an environ mental educator, naturalist, and grassroots organizer for various non profit organizations in Ohio, North Carolina, and Washington, D.C. In 2002, she was a field assistant to Dr. Greg Dwyer from the University of Chicago working on nucleopolyhedrovirus (N PV) transmission in gypsy She became the Executive Director of Keep Franklin County Beautiful, Inc. (KFCB) a non profit dedicated to environmental education, waste reduction, and commun ity organizing. While at KFCB, the organization earned national and state awards for their programs. In 2004, Lindsey entered into a business venture with a partner and bought a retail video game store in Columbus, Ohio called Games Underground After a prosperous 4 year run with the business, she closed the store t o pursue a Master of Science degree from the Universit y of Florida in Gainesville in i nterdisciplinar y e cology. L indsey will be attending The Ohio State University in pursuit of a Ph.D. in re storation e cology starting in the fall of 2010.