Developing an Integrated Pest Management Solution for Pierce’s Disease Spread by

Monitoring of non-agricultural areas demonstrated that glassy-winged sharpshooters occur at significant densities within the landscaped, riparian, and natural areas surrounding and interspersed within the agricultural areas of the Temecula valley. As it will be both politically and biologically impossible to eradicate sharpshooters from these non-agricultural areas associated with the Temecula vineyards, these areas will have to be considered to be permanent and constant sources of glassy-winged sharpshooters. Insecticide treatments of citrus had a dramatic and significant effect on parasitoids of glassy-winged sharpshooters in the Temecula region. With insecticide treatments few sharpshooter eggs were found in treated areas and only a minute proportion of these were parasitized. At the beginning of April, 2000, glassy-winged sharpshooter parasitoids were recorded in 30%of all sharpshooter eggs collected. By June the parasitism rate had risen to 90%. Parasitism rates continued in excess of 90%for the remainder of the year. The primary parasitoid encountered was Gonatocerus ashmeadi (Hymenoptera: Mymaridae). Results from the first six months of sampling have yielded positive detection of X. fastidiosa in grapes at many sites throughout the Temecula valley. Xylella fastidiosa was also consistently detected in almond and oleander in the Temecula region. Three other plant species (coyote brush, mustard, and elderberry) showed weak positive results for the presence of X. fastidiosa when tested by ELISA; however, neither culturing or PCR techniques supported these results. We expect that more plants will yield positive results as we begin to routinely use the more sensitive immunocapture PCR method. Thus far, no obvious source of inoculum of PD, other than grape and almond, was detected in the valley using these methods. To date, five parasitoid species have been investigated as potential biological control agents of GWSS. Two were found to lack the host plant range required of an effective biological control agent and the remaining three were found suitable for further research. Rearing methodologies for sharpshooters and parasitoids have been significantly improved. We are now providing sharpshooter egg masses to two laboratories for biocontrol research, one laboratory for behavior research and three laboratories involved in insecticide research. The work determining host plant preference of sharpshooters was largely abandoned as it was deemed no longer necessary with the successful development of rearing methodologies. The construction of the phenological degree-day model is continuing. To date we have taken the insect through a single generation at several temperatures; this research will require additional funding and at least another 18 months of effort before any significant conclusions are made. Field-trials conducted at U.C. Riverside in 2000 showed that two soil-applied and one foliar-applied neonicotinoid induced high sharpshooter mortality through an 8 week period. Most striking is our observation that imidacloprid applied to grapevines in September 1999 had a substantial and disruptive impact on sharpshooter feeding almost a year later. This may, in fact, be more important to protecting plants from Xylella fastidiosa-carry’mg sharpshooters then inducing mortality as disease acquisition and transmission will likely be disrupted as well. 30 Plant microelements such as zinc, copper, manganese and iron, as well as three antibiotics were tested for inhibition against Xylella fastidiosa in vitro. Tetracycline was the most effective antibiotic and zinc was the most toxic microelement. Prophylactic materials being evaluated included 3 inducers of systemic acquired resistance (SAR) and 4 microelement formulations. Therapeutic materials include several formulations of microelements and 2 antibiotics. All prophylactic field plots were mapped for PD each fall, however no new infections were found in either the treated or control vines in 2000. Bactericides were applied as foliar sprays, as materials packed into hollow nylon (DP) screws or in drilled holes packed with bactericides that were suspended in agarose and the ends of the holes were sealed with DP screws. Several of the drilled through/DP screw treated vines did not show any PD symptoms following treatment, however these vines were also severely pruned following treatment. A final assessment of the efficacy of these bactericide treated vines will be made in the summer of 2001. An injection machine that is widely used for injecting avocado trees was found to work well for injecting vines in the spring but less effective in the fall. Several potted plant experiments were performed using soil drenches of microelements as therapeutic or prophylactic bactericides. Although significant phytotoxicity occurred with some of the materials, managanese and zinc treatments may have some potential. A custom-made pressure bomb was purchased and used to express xylem sap from 1 meter long grapevine canes that were treated with various microelements. Surprisingly high concentrations of zinc and managanese were found in the xylem sap of grapevines treated with amino acid chelates of these elements. Additional experiments are now being done to determine whether the xylem sap is actually toxic to Xylella fastidiosa or if the ions are too tightly bound.

Biological, Cultural, and Chemical Management of Pierce’s Disease

Objective 1: Understand how Xf moves, and the patterns of its movement, in systemic (grape, blackberry) and non-systemic (willow) plant hosts using microscopy. (Purcell) Confocal microscopy of red willow inoculated with A/* showed that Xf can multiply to very high levels within individual xylem elements but does not move to adjacent cells in non-systemic hosts. Thus, the low populations of ^/”recovered from willow by culturing represent many Xf cells within a very few colonized cells rather than fewer numbers of Xf in a larger number of cells. High magnification scanning electron microscope and modified staining methods of Xf’vn. grape revealed “tentacle-like” structures (fimbriae) at the narrow ends and division plane of dividing Xf cells collected from expressed xylem sap of infected grape but not when collected from cultured^ cells from PW medium. .A/aggregated on the surface of xylem vessels in PD-infected grape petioles as well as dormant infected grape tissue (green and woody stem pieces) early or late in colonization of grape xylem. The bacteria were embedded in a matrix of fibers that covered them in a net-like fashion. This matrix of Xf cells and extracellular materials (a biofilm) adhered to the xylem cell walls. The fimbriae of attached A?” connected onry m contact with plant tissues. Biofilm formation appears to be a consistent feature of colonization of grape by Xf. Cells of Xf grown in a new culture provided by Brazilian collaborators produced abundant extracellular fibrils but not fimbriae, and these cells attached strongly to glass, unlike cells grown in traditional Xf media (PW, etc). Our findings suggest that Xf produces attachment structures only under specific environmental cues.

Objective 2: Understand how temperature influences the movement and survival of Xf and the incidence and/or severity of PD. (Purcell) Field inoculations of grape at different times of the year at Oakville (1997), Davis (1997-98), Fresno (1998-99) and Hopland (1999) consistently confirmed that infections during April through May and to a much lesser in June resulted in persistent infections of JSyfthe following year; whereas most June, July or August inoculations resulted in non-persistent infections unless the base of canes were inoculated. At the Hopland site, recovery of early-season infected vines was the first evidence of possible climate-mediated recovery of PD-infected vines that was not explainable by pruning eliminating the Xf infections. Inoculations of vines during April 2000 in San Luis Obispo Co. and Santa Barbara Co. suggest that cool temperatures (not exceeding 16 C) following inoculation may not allow systemic infection by Xf. The highest populations of Xf occurred at the bases of symptomatic canes and decreased towards the tip of sampled canes in experimentally inoculated or naturally infected vines. Experimental freezing of dormant, PD-infected potted vines cured the plants of PD at rates of 0 to 70%after exposures to temperatures ranging from -2 to -10 C for 3 successive freezing exposures. Cane segments treated identically at the same time never recovered, suggesting that some aspect of plant physiology, rather than the direct action of freezing alone, kills Xfby freezing in grape xylem. The survival of Rafter freezing in various liquid media further supports this view. 39

Objective 3: Determine whether vegetation barriers or trap crops can reduce the incidence of PD in riparian areas. (Weber) Two trap crop trials have been established on either side of a large vineyard in Napa. One trial borders the Napa River, the other Milliken Creek. Vines are spaced 9 feet (rows) by 5 feet (in-row). In each trial, St. George rootstock is planted at the ends of adjacent rows to create the trap crop treatments. Trap crop treatments include the first 6 vines in 12 adjacent rows. In each trial, there are three replicate trap crop planting and three control plantings where Chardonnay or Pinot Noir are planted to the end of the row. The vineyard was planted in May 1999. By October 2000, trap crop vines along the Napa River had reached the upper trellis wire and had filled in well. They were pruned in December along with the rest of the vineyard leaving wood on the “fruiting” wire. On the Milliken Creek side, deer did considerable grazing damage throughout the year and the vines did not develop as well. Admire (soil-applied imidacloprid) was applied to trap crop vines in November 1999. Due to concerns about its efficacy, this will be replaced with a spring 2001 treatment of Provado (foliar-applied imidacloprid). BGSS were monitored in 2000 in both trials using yellow sticky cards placed at the ends of rows. BGSS were detected in all treatments, although counts were relatively low. Per-trap catch totals ranged from 2 to 23 BGSS for the trapping period March-October. A PD disease survey was conducted in both trials in Sept. 2000 extending approximately 40 vines into the vineyard. Only 5 vines were found showing PD symptoms. Monitoring and mapping will continue in 2001.

Objective 4: Develop transformation and transposon mutagenesis for Xylella fastidiosa. (Kirkpatrick) Procedures were developed to successfully introduce the transposon, Tn5, into two different strains of Xylella fastidiosa (Xf). Electroporation conditions that were similar to those that have been used to electroporate DNA into several Xanthomonas species were found to be efficacious for electroporating DNA into Xf. Several attempts to transpose Xf with two different Tn5 and two TnlO suicide constructs failed to produce transposed Xf cells. However, using identical electroporation conditions, we obtained several hundred Tn5 mutants of both the Fetzer and Temecula Xf strains using a “transposome” complex composed of a transposase protein/DNA complex containing Tn5. Southern blot analysis showed these were random, single, Tn5 inserts throughout the Xf genome. Sequence analysis of Xf genomic DNA that flanked the Tn5 insertion identified several genes that were found in the CVC strain of Xf. We now have approximately 500 Tn5 mutants stored at -80; inoculation of grapevines and other analyses of the mutants will begin in a few months after several thousand mutants have been obtained. Three, 1.8 kb Xf plasmids were cloned and sequenced from the UCLA Xf strain. The largest open reading frame (ORF) on these small Xf plasmids had significant homology with another phage replicase gene, suggesting this Xf ORF encodes a plasmid replicase gene. Initial cloning of these plasmids probably interrupted the promoter for this ORF and we have since recloned these plasmids at another location on the plasmid. We are now introducing the KanR gene from the Tn5 construct described above, which we know is expressed in Xf, into the cloned Xf plasmids. In this manner we hope to construct an XGE. coli shuttle vector that will greatly facilitate molecular genetic analyses of Xf.

Objective 5: Isolate and identify endophytic bacteria that systemically colonize grapevine xylem. Identify natural, or genetically engineer, endophytes to be antagonist to Xf. (Kirkpatrick) Several hundred bacterial isolates were obtained from both healthy and Xf-infected grapevines located in Napa and Yolo counties. Specific grapevines of two cultivars were sampled bimonthly during 1999 and 2000. In addition, healthy appearing grapevines that were growing in the middle of vineyards that were decimated by PD were also randomly sampled. Four different media were used to cultivate the bacteria, however no significant differences in the types or numbers of bacteria were observed using the different media. Gram stains were performed on all of the isolates so that the appropriate Biolog plates can be used to initially identify the isolates, at least to the genus level. Beginning in the spring these isolates will be pinprick inoculated into grapevines growing in the greenhouse. After several weeks, attempts will be made to recover the bacteria from sections of xylem that are 2 or 3cm from the point of inoculation. Any bacteria that are found to systemically colonize grapevine xylem in reasonable concentrations will be tested for natural antagonism towards Xf. We are now screening a random peptide library for synthetic peptides that are 40 inhibitory to Xf. If these peptides are identified, attempts will be made to genetically engineer grape endophytes to express these peptides within grapevine xylem.

Objective 6: Genetics of Resistance to Xf. (Walker) A Design II mating design with a set of 6 females by a set of 6 male parents, from which we will select sets of seedlings to study the inheritance of Xf resistance, is completed. The mean expression of resistance within a given female across all males and similar comparisons of males across females will allow us to draw conclusions about the inheritance of Xf resistance. We define Xf resistance as the ability of a genotype to limit the movement of Xf, particularly in a downward direction, and have tested a set of resistant and susceptible individuals to determine whether this definition is valid. The known susceptible genotypes were V. rupestris ‘ A. de Serres’ (the female parent of the 89 population), Chardonnay and the V. rupestris x M rotundifolia genotype 8909-19; potentially resistant genotypes were 8909-04 and 8909-11; and the resistant genotypes were 890915 and 890917. After 4 weeks, Xf was easily detected in the three susceptible genotypes. By 16 weeks, the differences among resistant and susceptible genotypes are very clear in terms of both symptom expression in leaves and unevenly lignified stems and ELISA readings. ELISA is cheaper, faster, quantifiable and more simple than PCR detection of Xf, and can reliably detect 10,000 cfu/ml of Xf in ground plant sap. 150 samples with duplicate readings can be processed in a day at a material cost of $0.28 per plant sample. IC-PCR detection is 10 more sensitive, but is not quantifiable and costs about $1.44 per sample. Nested PCR gave very good results and is 100 times more sensitive than IC-PCR, but few samples can be run per day, and the cost of materials and labor is high ($2.88 per sample). We also examined spot-PCR which would allow several hundred samples to be run per day, but the sensitivity is equivalent to ELISA and cost $2.70 per sample. (All costs exclude labor). We are mapping Xf resistance in the V. rupestris X M. rotundifolia 9621 population (previously used for Xiphinema index resistance). We have AFLP marker data on about 70 of the 150 individuals and are testing 4 replicates from each of these 150. Xf resistance data from 70 individuals are complete and the others are due for analysis in about 4 months. We plan to place Xf resistance on our existing AFLP based map. If the resistance trait does not place on the existing map, we have seeds for a second mapping population based on a cross of the 8909-15 X 8909-19; a resistant by susceptible genotype which will allow greater power in mapping the resistance genes. However, this seedling population will have to be grown out from seed, propagated and screened for resistance and marker information. We also have (8909-08 and 8909-23) to Chardonnay, which should also segregate widely. We bench-grafted Chardonnay on each of the following rootstocks: AXR#1, St. George, 3309C, 101-14Mgt, Schwarzmann, 44-53 Malegue, Riparia Gloire, 1616C, Lenoir, 5BB, 5C, Borner, 110R, 1103P, Harmony, Freedom, and Ramsey, and Chardonnay (control), Riesling, Sylvaner, Chenin blanc, and Colombard. Own-rooted Chardonnay, Riesling, Sylvaner, Chenin blanc and Colombard were also inoculated. We have evaluated all three reps for PD expression and ELISA sampling to determine the extent of downward movement is continuing. The vines were cut back to basal buds, are now re-growing and will once again be evaluated for symptoms and by ELISA.

Objective 7: Determine whether xylem chemical composition is involved with PD resistance or susceptiblity in grape varieties and common Xf plant hosts. (Andersen) In an effort to determine whether xylem fluid chemistry is related to resistance to Xylella we investigated the chemistry of Vitis genotypes covering a wide range of resistance/tolerance. Xylem fluid chemistry of 10 grape genotypes belonging to 5 species has been characterized, although only the amino acid data was presented. Total amino acids varied over 4-fold, and many amino acids only occurred in less than lOuM or trace quantities. A complete correlation with resistance will await compilation of organic acid and sugar data. Total amino acids in xylem fluid collected from Chardonnay grafted on 4 different rootstocks showed an effect of rootstock and of infection with Xylella fastidiosa. Xylem fluid of Chardonnay on 3309 rootstock 41 tended to be most dilute of the rootstocks and XF infection increased total amino acids in xylem fluid of Chardonnay on all rootstocks except S04. Arginine was the amino acid that showed the biggest increase in amino acids with XF infection. A more complete analysis of the effect of xylem chemistry of grape genotypes on resistance to Xylella fastidiosa and the change in chemistry with Xylella fastidiosa infection awaits further research. An investigation is underway to determine whether resistance to Xylella fastidiosa is influenced by xylem fluid chemistry across plant species common to both Florida and California. Xylem fluid chemistry of 32 host plants species/cultivars was analyzed and natural X. fastidiosa (resistant/susceptible) was noted by PCR analysis. About 50%of the plant species/cultivars were Xylella negative. Those that were Xylella-negative were mechanically inoculated with X. fastidiosa during the fall of 1999. After 4 weeks only one species was positive, Vitis rotundifolia (wild grape). Each of the Xylella negative species/cultivars have also been sampled fall 2000 and are currently being analyzed via PCR. Those species that still do not harbor Xylella may be inoculated with different strains for confirmation of resistance to multiple Xylella strains. In vitro nutrient requirements of Xylella have been studied for 3 months without much success. A UCLA strain of X. fastidiosa was subjected to 3 months of experimentation using Chang and Donaldson’s chemically defined media. This UCLA strain grew well on PD or PW+ but did not grow on the chemically defined media. We switched strains to ATCC 35881 and tested this strain on all the media. The result was that it grew well on all the media and it even grew slowly on a media consisting of glutamine as the only amino acid source. Thus, nutritional fastidiousness is extremely strain dependent. Additional experiments will be performed with a different strain (probably the Temecula strain) after consultation with B. Kirkpatrick, S. Purcell and A. Walker. Lytic peptides were extremely effective against Xylella. They act by disrupting cell membrane integrity of bacteria but not eucaryotic cells of higher animals. Cecropins were one to three orders more effective than tetracycline. Xylella fastidiosa incubated with Cecropin A and B at 1 uM resulted in 100%inhibition of growth; incubation with Xylella at 0.1 uM resulted in 95%inhibition. Indolicidin was a fairly strong inhibitor of Xylella fastidiosa with 100%inhibition at 9.5 uM. Magainin II was followed by Magainin I in potency. For tetracycline 100%inhibition was achieved at 112 uM. These compounds may serve as a potent naturally occurring pesticide against Xylella.

Key to Management of Glassy-winged Sharpshooter: Interactions Between Host

The goal of the program is to determine the relationship of host plant xylem chemistry, and leaf morphology on host selection, feeding and ovipositional behavior of GWSS and its parasites. GWSS oviposits in many plant species, yet the majority of GWSS egg masses tend to be concentrated on a few select host species that apparently offer the quality of xylem fluid (food) required for survival of nymphs. Food quality for nymphs appears to be an important factor affecting the population increase of GWSS. We concentrated on developing field methods and data towards determining the host selection behavior and use for oviposition of known host plants preferred for feeding by the adult GWSS. Experimental sites that offered mixed host plants were established in isolated islands in an open field and a large planting of crape myrtle. Plants on the islands and the surrounding crape myrtle were examined for the presence of GWSS life stages weekly. Results from both sites indicated a statistically significant preference of GWSS for ovipositing on holly plants over all other plant species. The second host most frequently used by GWSS for oviposition was Bradford pear. A few egg masses were also found on other hosts. Pyracantha was chosen as oviposition host only very early in the summer and in the fall. Parasites were able to utilize the GWSS eggs on all hosts. The data suggests that proximity of adult and ovipositional hosts may greatly increase exposure of adult hosts plants to GWSS. We also worked on standardizing and quantifying methodology for rearing GWSS. Strict attention must be given to lighting conditions especially from late to early season. When diapause occurs, diapause can be terminated if three weeks of short daylength are followed by long daylength (we employ 16:8 lightdark regime) so long as the proper ovipositional hosts are present. We have now begun the second phase of this experiment, which is to assess the suitability to GWSS based on xylem chemistry of California-relevant host species. We have collected xylem chemistry data on the host plants used in the field plots and these are ready for analyses. We are currently investigating Chardonnay grapes, Navel oranges, Spanish Pink Lemon and Crape Myrtle. Data from these hosts will be compared to rates of development on soybean in order to assess the value of each of these species as developmental hosts for GWSS. Quantitative analysis is needed to prioritize the role of each potential host for implementation of GWSS control measures. By exposing GWSS eggs to parasite adults, we determined the duration of the susceptibility of GWSS eggs to the parasites Gonatocerus ashmeadi and Gonatocerus morrilli. Parasitoids can successfully parasitize 100%of GWSS eggs for at least 7 days after oviposition. We also investigated the overwintering behavior of Gonatocerus sp. and GWSS. We determined for the first time in the U.S. that GWSS as eggs and Gonatocerus sp. within parasitized egg masses can overwinter at north Florida winter temperatures. Laboratory experiments showed that parasitoids fed honeydew provided from excised leaves with live whiteflies lived twice as long as those fed simple honey solution. Perhaps parasitoid abundance could be enhanced by providing alternative food sources.

Impact of the Glassy-Winged Sharpshooter on Pierce’s Disease Spread in

Our second year of studying the epidemic of Pierce’s disease (PD) in Temecula, CA, has revealed a pattern of diseased grapevines that is consistent with vine-to-vine spread. In 1998, the best description of disease incidence was “scattered” with local hot spots, and we documented a strong relationship between PD severity and distance from citrus. Surveys in 1999 revealed vineyards with between 5%and 96%PD-infected vines, and disease incidence was not strongly linked to proximity from citrus. Yellow sticky-card and sweep-net sampling over the past two years indicate that glassy-winged sharpshooters (GWSS) is overwhelmingly the most important PD vector in this area. We also observed low numbers of other PD vectors, including the smoke-tree sharpshooter and the willow sharpshooter, both of which are native to this area. GWSS population densities were lower this year than last year. Anecdotal observations suggest that egg parasitism made a substantial impact on GWSS population density. Consistent between the two years of sharpshooter monitoring was evidence of the importance of citrus to GWSS populations. Also consistent were peaks in GWSS activity in the mid-late summer and in January. In January of 2000, sharpshooters were observed feeding from dormant grapevines and excreting fluid. Our first year of monitoring in the southern Kern County, where the GWSS was observed for the first time in 1998, revealed a small, yet apparently stable population. Studies on GWSS dispersion show that they are strong flyers and disperse deep into vineyards from adjacent citrus groves. We showed no statistical difference between sharpshooters caught in traps placed from 10 to 40 meters from the edges of vineyards. Traps at the edge of vineyards (0 meters) had significantly fewer sharpshooters. Traps in vineyards next to citrus groves had over twice the number of GWSS as traps in vineyards next to native vegetation. Sticky cards set vertically from 1-7 meters showed that GWSS’s do not fly at high altitudes. In fact, a seasonal mean of 95.4%of all the GWSS’s we caught were at altitudes of 5 meters or less. As with last year’s investigation, this year we found insecticides that could be part of a GWSS/PD management program. Our comparison of five neonicotinoids, both foliar and soil-applied, documented GWSS mortality as a function of time of exposure to treated plants, and as a function of time after treatment. The best performers were Admire, a soil applied insecticide that is registered for grapevines, and Acetamiprid, a foliar applied insecticide that is not registered. Admire and Acetamiprid caused over 63%and 81%GWSS mortality, respectively, in 24 hours even 8 weeks after application. Further studies with Admire indicated that GWSS did not feed enough to generate excreta within 3 hours of given access to treated plants. This anti-feedant property may in fact be more important to reducing PD spread than GWSS mortality. In addition to insecticides, we experimented with kaolin, a formulation of aluminum silicate. Kaolin disrupts insect sensory cues which caused a greater than 50%reduction in GWSS in treated verses untreated vineyard plots.

Ornamental Plants as Hosts of the Blue-Green Sharpshooter and Xylella fastidiosa

Ornamental plants within landscapes around wineries and residential areas of Napa Valley near vineyards with incidence of Pierce’s disease were monitored for the presence of the blue-green sharpshooter (BGSS) for two consecutive years. A list of the most commonly used plant species was established and the important hosts for BGSS were identified from the list . For this study, the presence BGSS was determined by using yellow sticky traps while sweep net sampling was used to detect the BGSS within the plant canopy. We found that on the plants that were sampled around wineries, five out of 35 plant species sampled had adults or nymphs of blue-green sharpshooters (BGSS). Liquidambar and basil were found to be breeding host, while rose, butterfly bush, and xylosma were found to be feeding hosts. Twelve plants within residential landscapes were recognized as hosts for the BGSS. The breeding hosts were oak leaf hydrangea, bush sage, liquidambar, and Japanese snowball. Feeding hosts are rose, butterfly bush, oak leaf hydrangea, crape myrtle, bush sage, lilac, African lily, liquidambar, myrtle, Japanese snowball, xylosma, and dogwood. In late May, 1999, mock orange {Philadelphus sp.) and Spirea sp. were determined to be breeding hosts. Plants found to be breeding hosts as well as other plant species from the list of common plants will be used to determine important hosts of the Pierce’s disease bacterium, Xylella fastidiosa (Xf). Plants will be inoculated with either infected BGSS or by using the needle inoculation technique. Basil, one of the identified breeding hosts, was inoculated using infected BGSS and found to not be a host of ^/”according to culture plate isolation. Rose, the most commonly planted ornamental in the study, is currently being evaluated as ^/”host after needle inoculation. Analysis of plant abundance are being determined using measurements of plant coverage along a transect. Transects were made at 20 foot intervals. This survey will allow us to establish a correlation between cover-abundance of landscape species and the activity of BGSS. In conclusion, landscapes have been determined to be important habitats for the BGSS. Residential landscapes appeared to harbor higher BGSS populations than landscapes surrounding wineries. In general, partly shaded landscapes with lush plants are more suitable for the BGSS. During the spring when the weather is cooler, BGSS adults and nymphs can be found on plants in full sun, however. Finally, while a large number of ornamentals were determined to be feeding hosts for the BGSS, fewer plants were found to be breeding hosts.