Oviposition preference of Homalodisca coagulata (Say) and two of its mymarid egg parasitoids, Gonatocerus ashmeadi Girault and G. triguttatus Girault, for two Citrus limon L. cultivars (‘Eureka’ and ‘Lisbon’) was investigated. In laboratory oviposition choice tests, the number of leaves containing H. coagulata egg masses, the number of H. coagulata egg masses, and the total number of H. coagulata eggs were significantly higher at 187.2%, 204.2%, and 181.7%, respectively, on ‘Eureka’ versus ‘Lisbon’. In the field, there was no significant difference in the number of H. coagulata motiles counted in five-minute searches of foliage on ‘Eureka’ and ‘Lisbon’ trees, and numbers of leaves with old (emerged) and new (unemerged) H. coagulata egg masses were equivalent between field-planted cultivars. In the laboratory, parasitism of H. coagulata egg masses by G. ashmeadi and G. triguttatus was 18.6% and 23.2% higher, respectively, for eggs laid on ‘Eureka’ leaves compared to ‘Lisbon’, but these differences were not significant. Leaf surface morphology and thickness of leaf cell layers of both lemon cultivars were compared with scanning electron microscopy (SEM). SEM demonstrated that total leaf thickness and the thickness of the palisade layer was 19.2% and 38.6% higher, respectively, in ‘Eureka’ leaves compared to ‘Lisbon’, and that H. coagulata egg placement was between the lower epidermis and spongy parenchyma layer for both cultivars. Furthermore, ‘Lisbon’ leaves had a smooth underside, whereas ‘Eureka’ leaves had many small ridges. The thickness and rough surface of ‘Eureka’ leaves may be beneficial for H. coagulata oviposition. However, additional research is required to further investigate whether leaf characteristics or xylem chemistry are responsible for H. coagulata oviposition choice. For mass rearing programs with lemons as host plants, it is recommended that the ‘Eureka’ cultivar be used in preference to ‘Lisbon’ because H. coagulata prefers this cultivar for oviposition and parasitoid foraging is not adversely affected.
The glassy-winged sharpshooter, Homalodisca coagulata (Say) (Hemiptera: Cicadellidae), is native to southeastern U.S.A. and northeastern Mexico, and following its establishment in California in the 1980s, it has become a significant threat to agricultural and ornamental industries due to its ability to spread the plant pathogenic bacterium, Xylella fastidiosa Wells et al. (Hopkins & Adlerz 1988; Purcell & Saunders 1999; Purcell & Feil 2001; Hopkins & Purcell 2002; CDFA 2003; Hoddle et al. 2003). A major classical biological control program has been launched in California against H. coagulata, including the mass rearing and release of two mymarid egg parasitoids, Gonatocerus ashmeadi Girault and G. triguttatus Girault. Citrus limon L. (lemon) has been chosen as a host plant for H. coagulata rearing and related experiments because it is one of the preferred hosts for H. coagulata in southern California and research on this pest is predominately conducted in agro-ecosystems where citrus dominates (Perring et al. 2001). However, several lemon cultivars exist, and two, ‘Eureka’ and ‘Lisbon’, are the most common commercial cultivars. It has been well documented that leafhoppers demonstrate significant oviposition preferences between plant species and between cultivars of the same species (McClure 1980; Stiling 1980; Singh & Agarwal 1988; Catindig et al. 1996; Sharma & Singh 2002). However, when this work was conducted it was not known which cultivar of C. limon H. coagulata preferred for oviposition under artificial rearing conditions. Consequently, the research undertaken here sought to determine the oviposition preference of H. coagulata for two C. limon cultivars, ‘Eureka’ and ‘Lisbon’, and whether G. ashmeadi and G. triguttatus exhibited preferences for H. coagulata eggs laid on ‘Eureka’ or ‘Lisbon’ leaves. To assist with interpretation of preference data collected from the laboratory, we conducted field surveys of H. coagulata life stages on ‘Eureka’ and ‘Lisbon’ lemons and compared leaf surface characteristics and measured leaf cell layers of both cultivars with scanning electron microscopy (SEM).
Materials and Methods
Homalodisca coagulata Oviposition Preference in the Laboratory
‘Lisbon’ and ‘Eureka’ trees approximately two years of age and grafted to Macrophylla sp. rootstock were obtained from C & M Nurseries, Nipomo, CA. Five trees of each cultivar were pruned to 60 cm in height (mean no. leaves per tree = 43 ± 5.7 for ‘Eureka’ and 37 ± 8.4 for ‘Lisbon’). The trees were potted into 4-litre containers and fertilized every two weeks with Miracle-Gro (20 ml/3.5 liters of water, Scotts Miracle-Gro Products Inc., Marysville, OH). Each of five cages (30 × 60 × 35 cm) received one tree of each cultivar in a randomized design. Cages with trees were held in a greenhouse at 26° ± 2°C and 30-40% RH under natural light. Forty female and ten male H. coagulata were collected from the field and placed into cages holding experimental plants. Every two days, approximately 25 field collected female and five male H. coagulata were added to each cage. The number of leaves with H. coagulata eggs, the number of egg masses, and the number of eggs laid on each cultivar were recorded daily. This study was conducted over the period May 10-24, 2001. Each combination of date and replicates was treated as a block, and a paired comparison t-test at the 0.05 level of significance was applied to the data in SAS (SAS 1990).
Homalodisca coagulata Oviposition Preference in the Field
A five-minute search for H. coagulata eggs on foliage 1-2 m above the ground was conducted on six ‘Eureka’ and six ‘Lisbon’ trees planted as part of a completely randomized block design variety trial at the University of California, Riverside Agricultural Operations Area. Surveyed trees were 17 years of age and cultivar scions were grafted to Macrophylla sp. rootstock. Old (emerged) and new (unemerged) H. coagulata egg masses found during time searches were harvested from trees and placed in labeled plastic bags. Additionally, a one-minute search for adult H. coagulata and nymphs was conducted on foliage within a 1-2 m band above ground around the tree for each replicate.
For collected leaves with emerged eggs, the number of H. coagulata egg masses, H. coagulata eggs per egg mass, emerged H. coagulata nymphs, solitary parasitoid emergence holes, unemerged nymphs and parasitoids, and ‘unemerged unknowns’ (those that could not be identified) were recorded for each lemon cultivar. Leaves with unemerged egg masses were held at 26° ± 2°C and 30-40% RH under a L14:10D photoperiod in 130-ml plastic vials (40-dram Plastic Vial, Thornton Plastics, Salt Lake City, UT) filled with deionized water and 3 ml of antiseptic [(Listerine Antiseptic Mouthwash, Pfizer Inc., New York, NY) (to prevent bacterial rot)] for two weeks to allow H. coagulata nymphs and parasitoids to emerge before recording data. This sampling and rearing protocol was repeated every two weeks over the period July 12-October 18, 2001. The total number of H. coagulata (adults and nymphs) counted on cultivars and the number of leaves with H. coagulata egg masses was log transformed prior to analysis. Data were compared between cultivars by two-way ANOVA in SAS (SAS 1990).
For emerged egg masses, percentage parasitism [(the number of solitary parasitoid emergence holes + the number of unemerged parasitoids)/(number of H. coagulata eggs) × 100], percentage nymphs [(the number of emerged H. coagulata nymphs + unemerged nymphs)/(number of eggs) × 100] and percentage unknowns [(unknowns/number of eggs) × 100] were calculated. The number of H. coagulata eggs was square-root transformed, and egg, H. coagulata nymph, parasitism, and unknowns data were compared between cultivars by two-way ANOVA. For unemerged egg masses, percentage parasitism, percentage nymphs and percentage unknowns were calculated as above. The number of H. coagulata eggs was square-root transformed, and all data were compared between cultivars by Friedman’s χ2 test. All statistical tests were performed in SAS and means presented here are back-transformed.
Gonatocerus ashmeadi and G. triguttatus Oviposition Preference
Parasitoid colonies were maintained at the University of California, Riverside at 26° ± 2°C and 30-40% RH under a L14:10D photoperiod in cages (50 × 40 × 40 cm) on H. coagulata eggs laid on ‘Eureka’ leaves. Colonies were provisioned with honey-water solution (3:1 Natural uncooked honey, Wild Mountain Brand, Oakland CA). Harvested H. coagulata egg masses on excised leaves were removed from colonies and checked daily for parasitoid emergence to assure uniform age for choice experiments. One ‘Eureka’ and ‘Lisbon’ leaf containing approximately 15 H. coagulata eggs (∼24-48 h of age) per leaf were placed through holes in a lid of a 130-ml plastic vial filled with deionized water and 3 ml of antiseptic. A second 130-ml plastic vial with ventilation [three 2-cm holes (one on the bottom, and one on each of two sides) covered with mesh netting (80 μm Jelliff Corporation, Southport, CT)] was inverted and attached to the lid of the vial holding the water and leaves. One newly emerged mated naïve female parasitoid (∼24 h old) was placed inside the inverted vial that covered the test material and left for 1 h to forage and oviposit at 26° ± 2°C and 30-40% RH under 1.2 ± 0.2 log lumens/sqm light before being removed. This set up was replicated 15 times for G. ashmeadi and G. triguttatus. Vials containing leaves with egg masses exposed to parasitoids were held at 26° ± 2°C and 30-40% RH under a L14:10D photoperiod for three weeks to allow parasitoids to emerge. Percentage parasitism [(the total number of emerged and unemerged parasitoids/total number of H. coagulata eggs) × 100] was calculated for each ‘Eureka’ and ‘Lisbon’ leaf. Females that were not mated (producing male only progeny) were excluded from the analysis. Data were transformed by square root and arcsine transformation and compared between ‘Eureka’ and ‘Lisbon’ cultivars by paired t- tests at the 0.05 level of significance in SAS (SAS 1990). Means presented here are back-transformed.
Investigation of Leaf Characteristics with SEM
One ∼2-mm wide leaf section was cut from the middle of each of 10 fully expanded ‘Eureka’ and ‘Lisbon’ leaves (fifth leaf down from the growing tip of branches from containerized trees) with a new razor blade and fixed in 2% gluteraldahyde in 0.1 M phosphate buffer for 2 h. Leaf pieces were washed in buffer and post fixed in 1% osmium tetroxide in 0.1 M phosphate buffer for 2 h. Leaf pieces were then washed in distilled water and dehydrated in an ethanol series and critical point dried. Prepared leaf pieces were mounted on SEM stubs and coated with gold-palladium. Photographs taken at ∼800 × magnification for each of the 10 samples for each cultivar of the vertical distribution of leaf cell layer were used to calculate mean thickness of the waxy cuticle, epidermis, palisade layer, and spongy parenchyma layer. Leaf parameters were compared between cultivars by two sampled t-test in SAS (SAS 1990). In addition, SEM was used to examine and photograph the intact leaf surfaces. Finally, ‘Eureka’ and ‘Lisbon’ leaves with H. coagulata egg masses (sourced from the “H. coagulata oviposition preference laboratory trial” above) were prepared for SEM as previously described, and examined to determine where the placement of eggs under the leaf cuticle on the undersides of leaves occurred.
Results
Homalodisca coagulata Oviposition Preference in the Laboratory
In the paired choice studies, the number of leaves containing H. coagulata egg masses, the number of H. coagulata egg masses, and the total number of H. coagulata eggs were 187.2%, 204.2% and 181.7% greater, respectively, on ‘Eureka’ trees compared with ‘Lisbon’ (leaves: t = 4.95, df = 76, P < 0.005; egg masses: t = 4.99, df = 76, P < 0.005; total eggs: t = 4.38, df = 76, P<0.005) (Fig. 1).
Homalodisca coagulata Oviposition Preference in the Field
There was no significant difference in the total number of H. coagulata nymphs and adults and the number of leaves with H. coagulata egg masses between ‘Eureka’ and ‘Lisbon’ cultivars (Table 1). For emerged egg masses in the field, the number of H. coagulata egg masses, total H. coagulata eggs, percentage parasitism and percentage unknowns were not significantly different between the two cultivars (Table 1). However, the percentage nymphs was significantly higher (5.8% greater) on ‘Lisbon’ trees compared with ‘Eureka’ (0%) (Table 1). For egg masses harvested in the field and returned to the laboratory, there were no significant differences in emergence rates between both cultivars for all parameters measured (Table 1).
Gonatocerus ashmeadi and G. triguttatus Oviposition Preference
In paired oviposition preference tests in the laboratory, parasitism by G. ashmeadi and G. triguttatus was respectively 61.7% ± 8.5 and 35.4% ± 12.2 on ‘Eureka, and on ‘Lisbon’ parasitism was 43.1% ± 9.9 and 17.2% ± 7.6, respectively, for G. ashmeadi and G. triguttatus. These differences were not significant (t = 1.3, n = 15, P = 0.22 and t = 1.17, n = 15, P = 0.26).
Investigation of Leaf Characteristics with SEM
Mean total leaf thickness was higher (19.2%) for ‘Eureka’ leaves (156.8μm ± 6.2) compared to ‘Lisbon’ (131.5μm ± 7.3) (t = 2.63, df = 18, P < 0.05) (Fig. 2). The mean thickness of the palisade layer was thicker (t = 2.32, df = 11.1, P < 0.05) in ‘Eureka’ (38.6%) leaves compared to ‘Lisbon’ (Fig. 2). There was no difference in cell layer thickness between the cultivars for cuticle (t = 0.39, df = 18, P = 0.70), epidermis (t = 0.60, df = 18, P = 0.56), and spongy parenchyma layers (t = 1.31, df = 18, P = 0.21) (Fig. 2). The underside of ‘Lisbon’ leaves was smooth, whereas ‘Eureka’ had many small ridges (Fig. 3). Homalodisca coagulata egg placement was between the lower epidermis and spongy parenchyma layer for both cultivars (Fig. 4).
Discussion
Results from this study showed that in the laboratory, the number of H. coagulata egg masses laid on ‘Eureka’ was 204.2% higher than on ‘Lisbon’. This indicates that when given a choice under artificial rearing conditions, H. coagulata preferred ‘Eureka’ over ‘Lisbon’ for oviposition, and suggests that this cultivar should be used for research projects that require large numbers of H. coagulata eggs laid on lemon leaves that can be harvested from young potted plants.
The reasons for this demonstrated H. coagulata oviposition preference are unknown. Scanning electron microscopy (SEM) showed that the underside of ‘Lisbon’ leaves was smooth, whereas ‘Eureka’ had many small ridges on the cuticle that may provide female H. coagulata with enhanced tarsal grip during oviposition. However, the depth of these ridges was not quantified. Results also showed that the thickness of the palisade layer was significantly higher (38.6%) for ‘Eureka’ leaves compared to ‘Lisbon’. A thicker leaf structure may be more favorable for H. coagulata oviposition.
Female H. coagulata preferentially oviposit on lower leaf surfaces, and SEM showed that egg placement occurs between the lower epidermis and spongy parenchyma layer for both lemon cultivars. A thick palisade layer, which is positioned at the top of the leaf, may afford some protection of H. coagulata eggs from adverse conditions. For example, solar radiation falls onto the upper leaf surfaces so a thicker palisade layer may insulate eggs from excessive sun exposure. However, the experimental design used in the current study did not standardize ‘Eureka’ and ‘Lisbon’ trees to possess equivalent numbers of leaves, or surface areas, which may have influenced oviposition preference by H. coagulata because of varying light intensity. Other factors that influence leafhopper oviposition preference, such as plant chemistry, leaf vein characteristics, number of branches, and trichome densities (Singh & Agarwal 1988; Lit & Bernardo 1990; Denno & Roderick 1991; Andersen et al. 1992; Sharma & Sharma 1997; Sharma et al. 1999; Sharma & Singh 2002) were not investigated in this study and may warrant further research if it becomes necessary to explain in detail factors influencing H. coagulata oviposition preferences.
Field results showed that there were no significant differences in the total number of H. coagulata nymphs and adults, and the number of leaves with H. coagulata egg masses between ‘Eureka’ and ‘Lisbon’ cultivars. Consequently the oviposition preference for ‘Eureka’ demonstrated in the laboratory was not confirmed in the field. In fact, field results showed that significantly more (5.9%) H. coagulata nymphs successfully emerged from ‘Lisbon’ leaves compared with ‘Eureka’ (leaf thickness of field-planted trees was not investigated in this study). The lack of congruence between laboratory and field studies may be due to differences in environmental factors and tree age between field-planted (∼17 years of age) and small containerized trees (∼2 years of age). Environmental aspects, such as sunlight, water, soil and nutrients, may affect leaf characteristics and influence host oviposition and utilization (DeReffye et al. 1995), while xylem chemistry varies drastically with plant age and phenology (Andersen et al. 1992, Andersen et al. 1995a, Andersen et al. 1995b). It has been demonstrated that H. coagulata is very sensitive to changes in nutritional quality (Brodbeck et al. 1990, Brodbeck 1999), and in California Toscano et al. (2003) showed that H. coagulata numbers were over 6-fold higher on young citrus trees compared to older trees. It is also possible that given the very high densities of ovipositing females in citrus at the time this field survey was conducted less preferred lemon cultivars were being used due to shortages of oviposition sites on the most preferred cultivars.
Parasitism by G. ashmeadi and G. triguttatus of H. coagulata eggs masses on ‘Eureka’ and ‘Lisbon’, were not significantly different. The lack of statistical significance suggests that using ‘Eureka’ trees for future experiments with these parasitoids should not detrimentally influence parasitism rate data or fecundity estimates, and demonstrates that parasitoids were not conditioned to oviposit preferentially on ‘Eureka’ over ‘Lisbon’ even though they had been reared on ‘Eureka’.
Summary
Based on the results of oviposition preference studies conducted in the laboratory we conclude that the lemon cultivar ‘Eureka’ is preferred over ‘Lisbon’ for oviposition by field-collected H. coagulata females. This cultivar preference by H. coagulata, however, was not observed in the field. Furthermore, the mymarid parasitoids G. ashmeadi and G. triguttatus do not exhibit cultivar preferences indicating that the use of ‘Eureka’ will not adversely affect studies on the reproductive and developmental biology of Gonatocerus spp. It is recommended that for mass rearing programs that require the use of small young containerized lemons to maximize H. coagulata egg mass or Gonatocerus spp. production the ‘Eureka’ cultivar should be used in preference to ‘Lisbon’.
Acknowledgments
This work was supported in part by the California Department of Food and Agriculture (CDFA) Pierce’s Disease-Glassy-Winged Sharpshooter Management Research Program. We thank David Morgan, CDFA, Mt Rubidoux Field Station, Riverside, California for supplying parasitoids to initiate our colonies, and Liangwei Wang for assistance with statistical analysis. We thank Betsy Boyd, Bryan Carey, and Ruth Vega for assistance in the field and laboratory.
References Cited
Fig. 1.
The mean number of leaves containing H. coagulata egg masses, the number of H. coagulata egg masses and the total number of H. coagulata eggs laid on ‘Eureka’ and ‘Lisbon’ lemon trees in the laboratory [different letters indicate significant (P < 0.05) differences between cultivars; error bars indicate ± SEM]
Fig. 2.
Mean thickness (μm ± SEM) of the cuticle, epidermis, palisade and spongy parenchyma leaf cell layers within ‘Eureka’ and ‘Lisbon’ lemon cultivars [different letters indicate significant (P < 0.05) differences between cultivars]
Fig. 3.
SEM photographs of the lower leaf surface of ‘Lisbon’ and ‘Eureka’ Citrus limon cultivars: (A) ‘Lisbon’ with smooth leaf surface; (B) ‘Eureka’ leaf surface with many small ridges on the cuticle that may enhance H. coagulata tarsal grip during egg laying; (C) Close up of small ridged on ‘Eureka’ leaves that were not present on ‘Lisbon’
Fig. 4.
SEM photographs showing placement of H. coagulata egg mass within ‘Eureka’ and ‘Lisbon’ lemon leaves: (A) ‘Eureka’ leaf—egg sliced latitudinally, shows placement of egg between epidermis and spongy parenchyma; (B) ‘Eureka’ leaf—egg sliced longitudinally, shows placement directly beneath epidermis; (C) ‘Lisbon’ leaf—egg sliced latitudinally, shows placement of egg between epidermis and spongy parenchyma (note: egg protein filled leaf cells)
