Rearing of D. citri and T. radiata

Diaphorina citri and T. radiata were reared according to a method modified from Skelley & Hoy (2004). The psyllid colony was maintained with potted citrus trees, approximately 30-50 cm tall. The trees were pruned, fertilized, watered, and placed in PVC (Poly Vinyl Chloride) frame cages (60 × 60 × 60 cm) covered with organdy cloth in a greenhouse maintained at 21-42°C, 20-79% RH, under a 16L:8D photoperiod. After 10-14 d, the trees had new tender growths (= flushes) 0.5-1 cm long, and were exposed to ovipositing D. citri females (approximately 5 per tree). Females were aspirated out with a vacuum pump after 72 h.

The parasitoid T. radiata was reared on potted citrus trees as described above for the D. citri colony. Ten trees were placed in wooden-framed cages covered with organdy cloth (114 × 94 × 76 cm) held in a greenhouse at 20-38°C, 22-81% RH, under a 16L:8D photoperiod. Approximately 50 ovipositing D. citri females were released in the cage and allowed to oviposit for 3-5 d, then removed with a vacuum pump. Adults of T. radiata (50 females and 25 males) were collected and released into the cage when the psyllid population reached the third instar. Adults of D. citri that emerged from unparasitized nymphs were collected as a by-product and were used to maintain the psyllid colony.

Effects of Adjuvants With and Without Petroleum Oil and Copper Hydroxide on D. citri

Screenhouse experiments were undertaken to evaluate the acute toxicity of Silwet L-77 (99.5% polyalkyleneoxide modified heptamethyltrisiloxane; Helena Chemical Co., Collierville, TN) and Kinetic (99% proprietary blend of polyalkyleneoxide modified polydimethylsiloxane and nonionic surfactants; Helena Chemical Co., Collierville, TN) alone and in combination with agrochemicals to psyllid eggs, young or mature nymphs, and adults. The screenhouse bioassays described here were conducted under variable environmental conditions between May and Oct 2007 to mimic grove conditions. Fifteen treatments were tested: water; Silwet L-77 at 0.05% (v:v); Kinetic at 0.05% (v:v); petroleum oil 435 (Growers 435, Growers Fertilizer Co., Lake Alfred, FL) at 0.5, 1, and 2% (v:v); copper hydroxide (Kocide 2000, DuPont, Wilmington, DE) at the recommended lowest label rate (LLR) (2.4 g/L of water) for suppression of citrus canker in Florida; petroleum oil 435 at 0.5% + Silwet L-77 at 0.05%; petroleum oil 435 at 1% + Silwet L-77 at 0.05%; petroleum oil 435 at 2% + Silwet L-77 at 0.05%; petroleum oil 435 at 0.5% + Kinetic at 0.05%; petroleum oil 435 at 1% + Kinetic at 0.05%; petroleum oil 435 at 2% + Kinetic at 0.05%; copper hydroxide at the LLR+ Silwet L-77 at 0.05%; and copper hydroxide at the LLR + Kinetic at 0.05%. The recommended LLR was calculated for 935 L per ha (100 gals/acre). Petroleum oil was evaluated at rates lower than 2% (Childers & Rogers 2005) to determine whether the adjuvants improved its toxicity to D. citri. Because an unnamed organosilicone surfactant tested at 0.2% (v:v) was reported to enhance the spread of citrus bacterial spot caused by Xanthomonas axonopodis pv. citrumelo (Gottwald et al. 1997), questions were raised as to whether Silwet L-77 might enhance the rate of infection with citrus canker (Xanthomonas axonopodis pv. citri). Because copper hydroxide is often used to protect citrus trees from citrus canker, it was combined with these adjuvants to evaluate the toxicity of this formulation to D. citri. All products were mixed with purified water (Barnstead Nanopure II system, Dubuque, IA) at a pH of 6.8 at 24.8°C. The experiments were carried out at the Department of Entomology and Nematology, University of Florida, Gainesville.

The protocol to evaluate the acute toxicity of the pesticide solutions on D. citri eggs, young or mature nymphs, and adults was similar to that of Srinivasan et al. (2008), except that each shoot was treated separately, and the untreated canopy was shielded with a plastic bag. The experimental design was modified because they did not test treatments on mature nymphs of D. citri and they evaluated only the residual toxicity of imidacloprid and abamectin to psyllid adults. To investigate the toxicity of treatments to psyllid eggs, potted citrus trees with at least 5 flushes 0.5-1 cm long that were infested with eggs were selected. On each tree, 5-6 shoots were sprayed with a different Preval® gun sprayer (Precision Valve Corp., Yonkers, NY) for each treatment. Shoots were sprayed from a distance of 70-80 cm for 10 s around each shoot, which covered approximately 90% of the leaf surface. Each shoot had 28-369 eggs (mean ± SD = 110 ± 57) and was considered a replicate. The trees were placed into cages inside a covered screenhouse for 9 d at 13-32°C, 27-94% RH, under a natural photoperiod during May 2007. The numbers of live nymphs and the numbers of unhatched eggs were scored with use of a dissecting microscope. Dead first-instars were considered as uneclosed eggs, because hatching larvae might be intoxicated if the treatment penetrated the egg shell or the larvae encountered a lethal concentration on the egg shell.

To assess the acute toxicity of the treatments to first- and second-instars (= young) D. citri, the potted citrus trees were infested as described for eggs. After 72 h, the trees were placed into cages inside the same screenhouse and left undisturbed for 6 d at 11-30°C, 19-96% RH, under a natural photoperiod during May 2007 to allow development to the nymphal stage. Shoots were sprayed using the same method as for the eggs. Each shoot had 33-390 nymphs (mean ± SD = 143 ± 72). After 4 d at 12-31°C and 18-94% RH under a natural photoperiod, live and dead nymphs were scored under a dissecting microscope. Motion, posture, presence of waxy excretions, dehydration of the body were considered to determine whether the nymphs were dead or live. To estimate the percentage distribution of the different instars at the time of the treatment, 6 flushes were sampled separately from shoots used to evaluate the treatments. The nymphs in each instar were scored separately to assess the instar densities across the tree. Each flush had 55-164 nymphs (mean ± SD = 113 ± 45).

Another experiment was conducted with third- to fifth-instars to verify whether the treatments were effective on ‘large nymphs’. The same methodology as for young nymphs was used, except that the trees were left undisturbed for 10 d after the oviposition period at 18-34°C and 27-95% RH under a natural photoperiod during Jun 2007. Each shoot had 20-184 nymphs (mean ± SD = 79 ± 34). The treatments were based on the results of the toxicity bioassay on young nymphs, and chosen on the assumption that ineffective treatments for young nymphs were likely to be ineffective against larger nymphs. The 9 treatments were: water; Silwet L-77 at 0.05%; Kinetic at 0.05%; petroleum oil 435 at 2%; copper hydroxide at the LLR; petroleum oil 435 at 2% + Silwet L-77 at 0.05%; petroleum oil 435 at 2% + Kinetic at 0.05%; copper hydroxide at the LLR+ Silwet L-77 at 0.05%; and copper hydroxide at the LLR + Kinetic at 0.05%. Two days after treatment, alive and dead nymphs were recorded with use of a dissecting microscope, based on the same criteria as described above for young nymphs (post-treatment conditions: 21-36°C, 26-94% RH, natural photoperiod). The shoots sampled to estimate the percentage distribution of nymphs had 63-163 nymphs (mean ± SD = 96 ± 36).

Acute toxicity to adults was assessed on potted citrus trees with only young flushes approx 3-5 cm long that fit inside Plexiglas cylinders (45.5 cm tall, 12.6 cm outside diameter). The top of the cylinder and 4 side holes (5.4 cm in diameter) were covered with mesh for air circulation. The soil surface was lined with plastic wrap and 2 paper coffee filters. The coffee filters were taped to the pot to facilitate the count of psyllid adults. Psyllid adults approximately 3 weeks old were exposed to the trees for 4-6 h within a cage in the greenhouse. Trees were infested with 20-98 adults (mean psyllids per tree ± SD = 51 ± 19). Different sprayers for each treatment were used to spray the trees containing adults; sprayers were held 70-80 cm from the trees and sprays applied for 15 s around the canopy. The trees were set inside the cylinders and placed into a roofed screenhouse. Alive and dead adults inside cylinders were counted after 72 h. Motion and maintenance of their typical posture (with the head touching the surface and the abdomen raised to a 30-45 degree angle) were considered to determine whether the adults were dead or live. The experiment was conducted 8 times on 4 different dates during Jul, Aug, and Oct 2007. During the experiments, the average minimum and maximum temperatures (mean ± SD) were 22 ± 3.1°C and 33 ± 2.9°C, respectively. The mean RH (± SD) averaged 42 ± 8.6% (minimum) and 92 ± 4.5% (maximum). Daily weather data were obtained from the weather station maintained by the Department of Agricultural and Biological Engineering, University of Florida, Gainesville, approximately 800 m away. Temperature and RH inside and outside the cylinders were recorded with a Traceable® Digital Thermometer (Fisher Electronics, Pittsburgh, PA). The mean minimum and maximum temperatures inside the cylinders were higher (0.18°C and 0.48°C, respectively) than the environment temperature. The minimum and maximum RHs inside the cylinder were higher by 7.8 and 0.2%, respectively. Data expressed as percentages were evaluated with one-way analysis of variance with Proc GLIMMIX (SAS Institute 2002), and a Tukey-Kramer test was used to separate means (α< 0.05).

Acute and Residual Toxicity of Adjuvants and Selected Pesticides to T. radiata Adults

The acute and residual toxicities of adjuvants alone and in combination with petroleum oil, copper hydroxide, abamectin, and imidacloprid were tested in the laboratory with clip cage bioassays (Villanueva-Jimenez & Hoy 1998).

Residual Toxicity. Potted citrus trees with leaves 1 month old were sprayed to cover approximately 90% of the leaf surface. Silwet L-77 and Kinetic were tested alone (0.05%) and in combination with petroleum oil 435 at 2%; copper hydroxide at the LLR; abamectin (Agrimek 0.15 EC; Syngenta Crop Protection, Greensboro, NC) at the LLR; and imidacloprid (Provado 1.6 Flowable; Bayer Cropscience, Research Triangle Park, NC) at the LLR. Water was used as control. Trees were allowed to dry for about 30 min. Two treated leaves were pruned randomly from each treatment and placed at the top (with the abaxial surface downward) and the bottom (with the adaxial surface upward) of an acrylic ring (38 mm outside diameter, 10 mm tall). Foam tape (M-D Building Products, Oklahoma City, OK) 6 mm wide was placed on both edges of the ring to seal the clip cage. The treated surface of the clip cage chamber represented ca. 64.3% of the internal area. The rings had 2 mesh-covered windows 5 mm in diameter to reduce RH inside the chamber and provided a method for feeding T. radiata by placing a water-soaked rolled paper outside the mesh window and a honey-soaked strip inside the chamber. The clip cage was held together by 2 pieces of circular cardboard on the top and bottom (40 mm diam) with 4 hair clips (Villanueva-Jimenez & Hoy 1998). To place T. radiata into the clip cage, 10 adult females 3-4 d old were chilled inside a 50-mL conical centrifuge tube in crushed ice for 5 min and gently tapped inside the clip cage. The clip cage was placed inside a plastic bag with a wet paper towel to ensure survival of the parasitoids throughout the experiment (McFarland & Hoy 2001). The plastic bags were placed inside a growth chamber at 24 ± 1°C under a 16L:8D photoperiod. Adult mortality was evaluated after 24 h. The insects were touched gently with an insect pin and considered alive if they walked, jumped, or flew away, and dead if they did not move or only moved antennae, wings, or legs. When the mortality in the water control was >15%, the entire replicate was eliminated. The bioassay was replicated 6 times on 3 different dates. The percentage of mortality data were arcsine √P transformed before statistical analysis to approximate a normal distribution. Data were subjected to one-way analysis of variance (ANOVA) with Proc GLM (SAS Institute 2002), and significant differences among means were evaluated by Fisher’s Least Significance Difference (LSD) test (α= 0.05%). When water-control mortality was detected, data were corrected by Abbott’s formula for control response (Abbott 1925).

The position of T. radiata adults inside the clip cage chamber was observed to determine whether the treatments were repellent, and the parasitoid preferred to move or settle on the untreated surfaces. The relative position (inner surface of the ring or treated leaves) of the parasitoid was recorded after 2 and 6 h, in 3 replicates per treatment. Data expressed as number of T. radiata on the leaves over total parasitoids inside the clip cage were evaluated with analysis of variance (Proc GLIMMIX) (SAS Institute 2002).

Acute (Direct) Toxicity. The bioassays were performed as for the residual toxicity test, except that untreated foliage was used and T. radiata were treated directly by placing 10 females 3-4 d old into a Petri dish upon crushed ice for 5 min (Villanueva-Jimenez & Hoy 1998). The adult females then were sprayed with a gun sprayer from a distance of 1 m for 5 s and placed inside the clip cage chamber. Twelve treatments was tested: water, oil 435 at 2%, abamectin at the LLR, copper hydroxide at the LLR, Silwet L-77 (0.05%) and Kinetic (0.05%) alone and in combination with petroleum oil 435 at 2%, abamectin at the LLR, and copper hydroxide at the LLR. Because imidacloprid alone and in combination with both adjuvants killed almost all parasitoids in the residual toxicity bioassay (see Results), acute toxicity of these products was not tested. Data were analyzed as for the residual toxicity test.

Effects of Adjuvants With and Without Petroleum Oil on D. citri

Eggs. Silwet L-77 and Kinetic alone caused only 20.4 and 16.5% mortality, respectively, to D. citri eggs (Table 1). Mortality of psyllid eggs sprayed with petroleum oil at 0.5, 1, and 2% ranged from 61.5 to 66.7%, but the differences were not significant. When petroleum oil at 1 and 2% was combined with 0.05% Silwet L-77, the toxicity increased from 66.3 to 78.7%, and from 61.5 to 81.3%, respectively, (F = 122.37; df = 11, 57; P < 0.0001) (Table 1). Petroleum oil at 0.5, 1, and 2% in combination with Kinetic was less effective (52.6, 57.7, and 52.3%, respectively) in killing psyllid eggs than petroleum oil alone at the same rates. Mortality rates of eggs sprayed with petroleum oil at 1 and 2% plus Silwet L-77 were significantly higher than when petroleum oil in combination with Kinetic was applied.

Young Nymphs. The bioassay on young nymphs was conducted when nymphs were predominantly first or second instars (20.9 and 64.7%, respectively). Silwet L-77 sprayed on trees infested with D. citri nymphs killed 69.7%, while Kinetic alone was significantly less effective (46.2% mortality) (F = 101.09; df = 11, 60; P < 0.0001) (Table 1). Toxicity of petroleum oil to young nymphs increased with the rate applied. Petroleum oil at 0.5% caused 34.5% mortality, which was significantly lower than at 1% (56.5%) or at 2% (84.7%). Silwet L-77 increased significantly the toxicity of petroleum oil to young D. citri nymphs when applied at 0.5% from 34.5 to 57.2%, and at 1% from 56.5 to 68.2%. However, adding Silwet L-77 or Kinetic did not improve the efficacy of petroleum oil at 2% because it was effective by itself to young nymphs (84.7% mortality). When Kinetic was combined with 0.5% petroleum oil, mortality of nymphs increased significantly from 34.5 to 56.2%, but not at the other rates tested. Petroleum oil at 2% was the most effective treatment in suppressing young nymphs (84.7%).

Mature Nymphs. Infested trees used to test the toxicity of petroleum oil and adjuvants to mature nymphs were sprayed when 94.5% of the nymphs were either third, fourth or fifth instars. Mortality of these larger D. citri nymphs on Silwet L-77-treated trees was 64.2%, which was higher than on Kinetic-treated trees (35.2%) (F = 56.95; df = 5, 28; P < 0.0001) (Table 1). When petroleum oil at 2% was applied in combination with an adjuvant, only Silwet L-77 significantly increased the mortality of mature nymphs from 44.8 to 74.3%. These tests indicate that Silwet L-77 is better than Kinetic in suppressing young and mature nymphs, and that Silwet L-77 significantly improves the efficacy of petroleum oil at 2% against large nymphs.

Adults. Silwet L-77 applied to D. citri adults caused 30.6% mortality, which was more effective than Kinetic (5.9% mortality) (F = 58.35; df = 11, 84; P < 0.0001) (Table 1). Mortality of adults sprayed with petroleum oil at 0.5, 1, and 2% was 2.7, 4.9, and 17.5%, respectively, indicating that petroleum oil alone is not effective in suppressing D. citri adults. Combining Silwet L-77 with petroleum oil at 0.5, 1, and 2% significantly increased the mortality of adults to 38.4, 35.2, and 54.7%, respectively. Kinetic applied in combination with petroleum oil improved the toxicity of that insecticide only at the rate of 2%, from 17.5 to 34.5%. These data suggest that Silwet L-77 is better than Kinetic at increasing the efficacy of petroleum oil to D. citri adults.

Effects of Adjuvants With and Without Copper Hydroxide on D. citri

Mortality of D. citri eggs on trees sprayed with copper hydroxide at the LLR alone and in combination with Kinetic was not different, ranging from 14.1 to 14.7% (Table 2). However, when Silwet L-77 was added to copper hydroxide at the LLR, mortality increased from 14.1 to 37.3% (F = 31.45; df = 5, 29; P < 0.0001). Copper hydroxide at the LLR caused 18.8 and 15.1% mortality of young and mature nymphs, respectively. When Kinetic was added to copper hydroxide, mortality rates of young (F = 172.52; df = 5, 30; P < 0.0001) and mature nymphs (F = 69.70; df = 5, 29; P < 0.0001) increased significantly to 30.1 and 22.6%, respectively. However, these rates were significantly lower than mortality of nymphs exposed to copper hydroxide at the LLR plus Silwet L-77, which were 64.9 and 46.7%, respectively (Table 2).

Copper hydroxide at the LLR killed very few D. citri adults (0.4%). When it was combined with Kinetic the mortality rate increased significantly to 6.6%, which was significantly lower than the mortality caused by copper hydroxide at the LLR in combination with Silwet L-77 (34.5%) (F = 45.69; df = 5, 42; P < 0.0001) (Table 2). However, the 34.5% mortality observed with the Silwet L-77 plus copper hydroxide combination was caused primarily by Silwet L-77 because there were no significant differences in mortality between copper hydroxide at the LLR plus Silwet L-77 and Silwet L-77 alone. Adding Silwet L-77 to copper hydroxide or petroleum oil caused these pesticides to uniformly cover the leaf surface rather than forming small separated droplets.

Acute and Residual Toxicity of Adjuvants and Selected Pesticides to T. radiata Adults

Residual Toxicity. Tamarixia radiata adult females did not appear to show any preference between treated and untreated surfaces inside the clip cages sprayed with water, Silwet L-77, and Kinetic alone and in combination with copper hydroxide at the LLR, petroleum oil at 2%, and abamectin at the LLR (data not shown). There was no significant difference in the proportion of T. radiata settling or walking on the leaf surface (treated with water, Silwet L-77, and Kinetic alone and in combination with petroleum oil, copper hydroxide, and abamectin) or on the inner surface of the untreated ring (F = 1.07; df = 11, 59; P = 0.4039). However, after 2 h, parasitoids exposed to leaves previously treated with imidacloprid alone and in combination with Silwet L-77 or Kinetic had died; thus these data were not evaluated by analysis of variance to determine the preferred position of T. radiata within the clip cages.

Mortality of T. radiata exposed to treated foliage was different among treatments (F = 55.05; df = 14, 75; P < 0.0001) (Table 3). Silwet L-77 and water did not affect the survival of the parasitoid, but Kinetic was significantly more toxic, resulting in the death of 13.3% of adults after 24 h. Leaves treated with 2% petroleum oil in combination with Silwet L-77 or Kinetic caused 8.3 and 15% mortality, respectively, which were not different from the mortality (11.7%) caused by 2% petroleum oil alone (Table 3). Mortality of T. radiata adults exposed to leaves treated with abamectin at the LLR (23.3%) or abamectin plus Silwet L-77 (15%) was not significantly different. When Kinetic was added to abamectin at the LLR, the mortality rate increased from 23.3 to 91.7%, suggesting that Kinetic increases significantly the toxicity of abamectin to T. radiata adults (Table 3). Tamarixia radiata adults exposed to that treatment appeared intoxicated; when touched with an insect pin, parasitoids moved wings, legs, or antennae, but could not walk, jump or fly.

Imidacloprid at the LLR alone and in combination with Silwet L-77 or Kinetic caused mortality of T. radiata adults ranging from 95 to 100% (Table 3). Imidacloprid alone or in combination with Silwet L-77 or Kinetic was significantly more toxic to the parasitoid than abamectin alone or with these adjuvants. Mortality of parasitoids exposed to foliage treated with copper hydroxide at the LLR was 5% (Table 3), and when Silwet L-77 or Kinetic was added to copper hydroxide, the mortality of T. radiata did not increase significantly (1.7 and 6.7%, respectively). Toxicity of copper hydroxide alone or in combination with Silwet L-77 or Kinetic to adults of T. radiata was not significantly different from that caused by water.

Acute (Direct) Toxicity. Mortality of T. radiata adult females sprayed with Silwet L-77 or Kinetic alone was 0 and 1.9%, respectively, which was not different from mortality of adults treated with water (1.7%) (Table 3). Petroleum oil at 2% and copper hydroxide at the LLR alone and in combination with Silwet L-77 or Kinetic caused 0 to 3.3% mortality of T. radiata, which were not significantly different from the water control. Abamectin at the LLR alone and in combination with Silwet L-77 or Kinetic killed 91.7 to 100% of T. radiata adults (F = 102.39; df = 11, 60; P < 0.0001), suggesting that abamectin is very toxic to this parasitoid when applied directly, but that petroleum oil at 2% and copper hydroxide alone and in combination with Silwet L-77 or Kinetic do not affect its survival.