Laboratory and field suppression of cowpea thrips by Chrysopa formosa (Neuroptera: Chrysopidae) and predator viability at high temperature Authors: Haihang Zhou, Yinling Xia, Lisheng Zhang, Mengqing Wang, Yuyan Li, Zhongjian Shen, Liangzhou Xu, Junming Dong, Yun Zhao, Zhengmei Li, Hui Dong, Jianjun Mao, Norman Leppla Published: 2026 ABSTRACT AbstractThe recent outbreaks of thrips in subtropical and tropical regions of the world caused serious economic losses. This has increased interest in releasing predatory insects to manage thrips in cowpea and other important crops. Therefore, we assessed the potential of the green lacewing, Chrysopa formosa Rambur (Neuroptera: Chrysopidae), a generalist predator, for controlling Megalurothrips usitatus Bagnall (Thysanoptera: Thripidae) and Frankliniella intonsa Trybom (Thysanoptera: Thripidae), the 2 major pests of cowpea in South China. Chrysopa formosa larvae displayed a type III functional response to M. usitatus. The theoretical maximum predation of a C. formosa third instar larvae on M. usitatus second instar nymphs and adults was 1054 and 570, respectively. The C. formosa second instar larvae consumed more than 71% of a mixed population of M. usitatus and F. intonsa in a cowpea field by 14 days after release. The predation of C. formosa on M. usitatus increased as the temperature increased from 15 °C to 35 °C, whereas the predator developed slower above 25 °C. There was no significant difference in the hatch of C. formosa eggs kept at 25 °C and 30 °C, and the survival of first to third instar C. formosa larvae maintained at 25 °C was more than 80%. Thus, C. formosa can effectively suppress M. usitatus and F. intonsa in cowpea.Keywords: thrips, predation, functional response, control efficiency, field trial Introduction Cowpea (Vigna unguiculata L. [Walp.]), a dicotyledonous vegetable belonging to the family Fabaceae, grows in most tropical and subtropical countries and is one of the most important pulse crops in the world (Muhammad et al. 2022). Cowpea contains various nutrients, including rich plant-based protein, potassium, calcium, iron, and phenolic compounds (Apea-Bah et al. 2021, Gerrano et al. 2022). The bean flower thrips Megalurothrips usitatus Bagnall (Thysanoptera: Thripidae) is a small-sized (3 to 4 mm) insect. It is an important pest of leguminous crops and extensively distributed in South China, Japan, India, Bangladesh, Srilanka, Pakistan, Fiji, Australia, and the United States. Recently, M. usitatus has become one of the most serious pests of cowpea in many provinces of China, particularly Hainan (Liu et al. 2020). The bioecological characteristics of M. usitatus contribute to its adaptation to a wide range of temperatures and humidity levels and result in its geographic distribution (Tang et al. 2015, Soto-Adames 2020). Megalurothrips usitatus infests 28 species of host plants, and 16 of which are legumes. Various Leguminosae crops in the genera Glycine, Arachis, and Vigna are hosts of M. usitatus (Chang 1987). Megalurothrips usitatus can damage cowpea at every stage from vegetative growth to flowering/fruiting. Megalurothrips usitatus infests crops by feeding on the sap of tender tissues with their piercing-and-sucking mouthparts. Both M. usitatus larvae and adults feed on flower buds. They can feed on young leaves and pods of leguminous plants when flowers are not available (Tang et al. 2015). The harmful effects of M. usitatus infestation on plants include wilting and deformity of leaves, injury of flowers, and abortion of pods (Wang et al. 2024). Megalurothrips usitatus can cause an 80% to 100% yield reduction in legume crops (Oparaeke 2006). Furthermore, M. usitatus can transmit plant viruses, such as tobacco streak virus in groundnut, cowpea, and sunflower, which cause peanut stem necrosis disease (Prasada Rao et al. 2003, Shukla et al. 2005). Frankliniella intonsa (Thysanoptera: Thripidae) is another species of thrips that occurs seriously in cowpea plants. It is indistinguishable from M. usitatus. There are quite a few similarities between the 2 species in terms of biological characteristics, behavioral habits, and occurrence patterns (Fu et al. 2022). Thrips usually lay eggs in plant tissues, which offer them a shelter to avoid natural enemies and overcome adverse environmental conditions (Lewis 1973). The postembryonic development of thrips consists of 2 nymphal instars, a prepupa stage, and a pupa stage. Pupation of thrips takes place in soil or leaf litter, which provides them an appropriate environment for emergence (Chang 1989). The small size of thrips and its hiding in the flowers make its control very difficult. Presently, chemical application is still the primary strategy for the management of thrips. However, synthetic insecticides are often not effective against thrips because of their cryptic behavior. Furthermore, excessive use of chemical insecticides has led to various detrimental effects on the environment, humans, and other living organisms (Tang et al. 2023, Cleveland et al. 2024). Particularly, it caused the poisoned cowpea incident in Hainan in 2010 (Pan et al. 2021, Camara et al. 2022, Fan et al. 2024). In recent years, there is a growing need to develop environmentally friendly pest control strategies. Nonchemical methods, including physical lure, agronomic measures, and release of natural enemy insects, have been adopted to suppress thrips (Tang et al. 2023). All of them are components of integrated pest management. The species in the genus Orius are important natural predators of M. usitatus. Orius maxidentex Ghauri (Hemiptera: Anthocoridae) (Men 1999), Orius strigicollis Poppious (Hemiptera: Anthocoridae) (Lee et al. 1991), and Orius sauteri Poppious (Hemiptera: Anthocoridae) (Liu et al. 2018) are reported to be able to effectively suppress M. usitatus population in laboratory experiments. In addition, parasitic wasps, such as Ceranisus menes, were identified as a parasitoid of M. usitatus (Loomans 2006). Green lacewings, including Chrysopa pallens Rambur, Chrysoperla sinica Tjeder, and Chrysopa formosa Stephens, etc., are broad-spectrum natural predators against a variety of soft-bodied insect pests, such as aphids, whiteflies, thrips, and mites, as well as eggs and small larvae of Lepidoptera and Coleoptera (Canard et al. 1984, McEwen et al. 2010). Recently, the control potential of C. formosa against Aleurodicus dispersus Russell (Hemiptera: Aleyrodidae), Hyphantria cunea Drury (Lepidoptera: Erebidae), Spodoptera frugiperda J.E. Smith (Lepidoptera: Noctuidae), and Spodoptera litura Fabricius (Lepidoptera: Noctuidae) was demonstrated at laboratory conditions (Lai and Liu 2020, Li et al. 2022b). However, to date, the ability of green lacewing in suppressing M. usitatus has not been studied. In this study, we evaluated the potential of the green lacewing, C. formosa against M. usitatus and F. intonsa, at laboratory or field conditions, and checked the tolerance of C. formosa to low and high temperatures. Materials and Methods Insects Megalurothrips usitatus colony used in the study was collected from a cowpea field in Yazhou District, Sanya City, Hainan Province, China (18°21′N, 109°10′E). The colony was reared in glass cylinders (20 cm in diameter, 30 cm in length) and supplied with fresh common bean pod, Phaseolus vulgaris. The cylinder was sealed with 2 layers of gauze at the 2 ends. Megalurothrips usitatus adults were transferred to a new cylinder daily using a fine brush during the oviposition period. The bean pod was renewed daily. Chrysopa formosa was captured in the suburbs of Beijing and fed the vetch aphid Megoura crassicauda Mordvilko (Hemiptera: Aphididae). The M. usitatus and C. formosa colonies were raised in a greenhouse kept at 26 ± 1 °C, with 70% relative humidity under a 16:8 h light:dark photoperiod. Functional Response Analysis Predation functional response was analyzed in plastic cylinders (2 cm in diameter, 8 cm in length), which were sealed using a parafilm membrane at the 2 ends. Chrysopa formosa was starved for 24 h prior to the predation experiment. A density gradient was set for each C. formosa first instar larva with 10, 20, 30, 40, 50, and 60 M. usitatus second instar nymphs. A density gradient was set for each C. formosa second instar larva with 30, 60, 90, 120, 150, and 180 M. usitatus second instar nymphs, or 10, 20, 30, 40, 50, and 60 M. usitatus adults. A density gradient was set for each C. formosa third instar larva with 300, 350, 400, 450, 500, and 600 M. usitatus third instar nymphs, or 110, 140, 170, 200, 230, and 260 M. usitatus adults. A section of fresh common bean pod (2 cm in length) was placed in the cylinder to prevent the thrips from starvation. The number of thrips eaten by C. formosa was recorded by counting the remaining thrips. Predation was observed for 24 h. Fifteen replicates were analyzed for the functional response at each predator instar and each prey density. Field Experiment A field experiment on the control of cowpea thrips by C. formosa was conducted in Lingaoxin Village, Yacheng Town, Yazhou District, Sanya City, Hainan Province (18°21′N, 109°10′E). The cowpea variety tested was “good luck.” The trial began on 11 February 2023, when cowpeas were in full bloom. At the time, the cowpea plants grew well, and 2 thrip species, M. usitatus and F. intonsa, were widespread in the field. No chemical pesticides were applied within 15 days prior to the experiment. Before flowering, square cages (2 m side length, 2.5 m height) consisting of iron posts and 80-mesh net were constructed in the field, with 10 cowpea plants in each cage. The number of thrips per flower was counted on days 1, 5, 8, 11, and 16 of the field experiment, and 40 C. formosa second instar larvae were released in each cage on day 2. Chrysopa formosa was not released in control cages. Agronomic measures were the same in the treatment and control cages. Four replicates were analyzed. To investigate the number of thrips per flower, 6 cowpea flowers were randomly selected from each cage (Wu et al. 2016). The flowers were separately placed in a white plastic tub and soaked in tap water, allowing the thrips to emerge from the flowers for easy counting. Temperature Assay To analyze the effects of temperature on C. formosa predation performance, C. formosa larvae at different instars were individually placed in cylindrical plastic tubes (2 cm in diameter, 8 cm in height) and kept at 5 temperatures: 15, 20, 25, 30, and 35 °C. Fresh common bean pod (2 cm in length) was provided in the tubes for M. usitatus as a diet. The predators were starved for 24 h prior to the temperature assay. The following combinations were set up for each temperature: (i) 60 M. usitatus second instar nymphs were exposed to 1 C. formosa first instar larva; (ii) 180 M. usitatus second instar nymphs were exposed to 1 C. formosa second instar larva; (iii) 60 M. usitatus adults were exposed to 1 C. formosa second instar larva; (iv) 600 M. usitatus second instar nymphs were exposed to 1 C. formosa third instar larva; (v) 260 M. usitatus adults were exposed to 1 C. formosa third instar larva. The tubes were sealed after the introduction of M. usitatus to prevent escape. Predation was observed for 24 h. Each predator–prey combination was replicated 10 times at each temperature. The life history of C. formosa was investigated at the above temperature regime. Chrysopa formosa eggs laid on the same day were individually inoculated into plastic petri dishes for each temperature treatment. To assess hatchability, 20 C. formosa eggs were collected randomly each time. Neonates were counted and transferred 3 times a day to prevent cannibalism. Hatchability was measured 6 to 9 times at each temperature. Enough M. crassicauda were provided after hatching, and life history parameters, such as larval survival, ecdysis, pupation, and emergence, were investigated. To assess survival and pupation rates, 5 replicates were set, and each contained 20 larvae. Cocoons were individually weighed after pupation. For statistics of eclosion rate, 5 replicates were analyzed, and each contained 20 pupae. Statistical Analysis The type of functional response of C. formosa to M. usitatus was determined prior to fitting the predation data to a specific Hollings’ equation (Holling 1959). A logistic regression model is adopted to identify the type of functional response by considering the ratio of prey consumed as a function of prey supplied (Juliano 2001). The data were fitted to the following polynomial equation: where “P0” is the intercept; “P1” is the linear coefficient; “P2” is the quadratic coefficient; “P3” represents cubic coefficient; “N0” donates number of prey supplied; and “Na” is number of prey consumed. If P1 > 0 and P2 < 0, the ratio of prey eaten is positively dependent on prey density, thus belonging to a type III functional response. If P1 < 0, the ratio of prey eaten goes down monotonically with the initial number of prey offered, thus showing a type II functional response. The predation data, including the number of thrips consumed by C. formosa at different prey densities, were analyzed by fitting the Holling’s type III equation after the determination of the type of functional response. The parameters of the functional response were calculated using nonlinear square regression. The disc equation is as follows (Holling 1959, Wu et al. 2004). where “Ne” is the daily predation amount of C. formosa on M. usitatus; “b” is the instantaneous attack rate of C. formosa on M. usitatus; “N” is the initial density of M. usitatus; “T” is the experimental time (1 day); “Th” is the time it takes for C. formosa to prey on 1 M. usitatus; “1/Th” represents the theoretical maximum predation amount of C. formosa on M. usitatus, and “a/Th” represents the ability of C. formosa to prey on M. usitatus. The control efficiency (E) of C. formosa on cowpea thrips in field was calculated using the following formulas (Zhang et al. 2026). where “TW1” is the average number of cowpea thrips per flower in the treatment groups after release of C. formosa larvae; “W0” is the average number of cowpea thrips per flower in the control groups before release of C. formosa larvae; “W1” is the average number of cowpea thrips per flower in the control groups after release of C. formosa larvae; and “TW0” is the average number of cowpea thrips per flower in the treatment groups before release of C. formosa larvae. Homogeneity of variance was confirmed using the Shapiro–Wilk test before testing the means. If the data set met the normality requirement, the significance of difference was determined using the parametric Student’s t-test or one-way ANOVA. If not, the nonparametric Mann–Whitney or Kruskal–Wallis tests were used. Differences in the number of thrips per flower and survival rate were determined by generalized linear model (GLM). Differences in control efficiency, egg hatching (from egg to larva), and eclosion (from pupa to adult) were analyzed by one-way ANOVA (Least Significant Difference test, LSD). Differences in the number of prey consumed were analyzed by one-way ANOVA (LSD), Mann–Whitney test, and Kruskal–Wallis test. Differences in developmental duration and pupal weight were analyzed by the Kruskal–Wallis test. Results Functional Response of C. formosa to M. usitatus The number of M. usitatus nymphs and adults killed by C. formosa increased with increasing prey density and plateaued at high prey densities (Fig. 1). However, the ratio of M. usitatus consumed by C. formosa monotonically went down with the increase of prey density. The graphical analysis of the ratio of M. usitatus eaten versus M. usitatus supplied indicated a type III functional response for all C. formosa instars. The linear coefficient (P1) of logistic regression consistently showed positive values, whereas the quadratic coefficient (P2) consistently showed negative values for all C. formosa instars (Supplementary Table S1). Predatory functional response of C. formosa to M. usitatus. (A) Functional response of C. formosa first instar larvae to M. usitatus second instar nymphs; (B) functional response of C. formosa second instar larvae to M. usitatus second instar nymphs; (C) functional response of C. formosa third instar larvae to M. usitatus second instar nymphs; (D) functional response of C. formosa second instar larvae to M. usitatus adults; and (E) functional response of C. formosa third instar larvae to M. usitatus adults. Chrysopa formosa third instar larvae consumed the most M. usitatus second instar nymphs at the highest prey density, followed by C. formosa second instar larvae and first instar larvae. Megalurothrips usitatus adults killed by C. formosa third instar larvae were significantly more than those killed by C. formosa second instar larvae at the highest prey density. Megalurothrips usitatus second instar nymphs killed by C. formosa second and third instar larvae were significantly more than M. usitatus adults killed by the same predators (Table 1). Table 1.Number of M. usitatus consumed by a C. formosa larva per day at the highest prey density Chrysopa formosa instar Megalurothrips usitatus second instar nymphs eaten Megalurothrips usitatus adults eatenFirst instar23.20 ± 1.46c…Second instar97.20 ± 3.29b30.60 ± 2.55bThird instar448.07 ± 19.98a195.53 ± 9.58aStatistics F = 374.935, df = 2, P < 0.001 U = 0, Z = −4.668, P < 0.001Data are mean ± SE (n = 15). Different letters in the same column indicate significant difference determined by one-way ANOVA (LSD) and Mann–Whitney test (P < 0.05). Number of M. usitatus consumed by a C. formosa larva per day at the highest prey density Data are mean ± SE (n = 15). Different letters in the same column indicate significant difference determined by one-way ANOVA (LSD) and Mann–Whitney test (P < 0.05). The magnitudes of the responses to M. usitatus second instar nymphs shown by the instar larvae of C. formosa were as follows: handling time (Th), first > second > third; theoretical maximum daily predation (1/Th), third > second > first. The magnitudes of the responses to M. usitatus adults shown by the instar larvae of C. formosa were as follows: handling time (Th), second > third; theoretical maximum daily predation (1/Th), third > second (Table 2). Table 2.Parameter estimates from the Holling III functional response of C. formosa preying on M. usitatus Megalurothrips usitatus life stage Chrysopa formosa instarInstantaneous attack rate (a)Handling time (Th)Theoretical maximum predation (1/Th)Predation capability (a/Th) R 2 aSecond instarFirst instar1.25630.029134.328943.12850.8273Second instar1.07530.0052189.6574203.94490.9007Third instar1.37700.00091054.14811451.56190.9312AdultSecond instar0.88650.013872.732664.47740.7636Third instar1.14130.0018570.4507651.07430.8884a R 2 represents the coefficient of determination obtained by fitting the stochastic predator equation. Parameter estimates from the Holling III functional response of C. formosa preying on M. usitatus R 2 represents the coefficient of determination obtained by fitting the stochastic predator equation. Field Control Efficiency of C. formosa on M. usitatus The efficiency of C. formosa in managing M. usitatus in the cowpea field was investigated after the release of C. formosa larvae. The M. usitatus population in the square cages was effectively suppressed by C. formosa larvae. The number of thrips per cowpea flower was significantly affected by release, period, and interaction between them (release: df = 1, P < 0.01; period: df = 4, P < 0.01; release × period: df = 4, P < 0.01) (GLM). The thrips in field cages were significantly fewer when predators were released than when no predators were released (Fig. 2A). The control efficiency increased gradually during the experimental period and reached 71.57% on day 14 after release of C. formosa larvae (Fig. 2B). The control efficiency was significantly affected by experimental duration (F = 8.297, df = 3, P = 0.003). Field control of cowpea thrips by C. formosa larvae. Number of thrips per flower was investigated on days 1, 5, 8, 11, and 16 of the experimental period. Chrysopa formosa second instar larvae were released on day 2 of the experimental period. Thrips was significantly reduced (A) (GLM) and the control efficiency of C. formosa on thrips peaked (one-way ANOVA, LSD) on day 16 of the experimental period (B). Difference letters above bars mean significant difference (P < 0.05). Predation at Different Temperatures The predation performance of C. formosa was temperature-dependent. The number of M. usitatus second instar nymphs eaten by C. formosa first instar larvae peaked at 25 °C, followed by 35 °C. Megalurothrips usitatus second instar nymphs and adults eaten by C. formosa second instar larvae were significantly more at 35 °C than at lower temperatures, demonstrating a promotion of high temperature to the predation of C. formosa second instar larvae. The number of M. usitatus second instar nymphs eaten by C. formosa third instar larvae was the maximum at 35 °C, followed by 30 °C. Chrysopa formosa third instar larvae consumed significantly more M. usitatus second instar nymphs at 35 °C and 30 °C than at other temperatures. The number of M. usitatus adults eaten by C. formosa third instar larvae also peaked at 35 °C and was significantly higher than that at lower temperatures (Table 3). Table 3.Number of M. usitatus eaten by a C. formosa larva at different temperatures Chrysopa formosa instar Megalurothrips usitatus life stagePrey densityNumber of prey consumed15 °C20 °C25 °C30 °C35 °CStatisticsFirst instarSecond instar6011.90 ± 0.97d14.40 ± 0.92 cd20.00 ± 1.20a16.00 ± 1.13bc18.90 ± 1.49ab n = 10, F = 8.071, df = 4, P < 0.001Second instarSecond instar18060.40 ± 4.38d65.40 ± 5.55 cd77.30 ± 3.59bc80.70 ± 2.64b95.10 ± 4.54a n = 10, H = 23.279, df = 4, P < 0.001Adult6020.00 ± 1.41d26.40 ± 1.10c30.50 ± 1.16b29.90 ± 1.17b34.70 ± 1.07a n = 10, H = 31.844, df = 4, P < 0.001Third instarSecond instar600343.80 ± 12.70d415.70 ± 10.11c492.40 ± 9.57b531.10 ± 18.33a552.00 ± 13.34a n = 10, H = 38.505, df = 4, P < 0.001Adult260140.80 ± 7.02c153.50 ± 9.56c184.70 ± 8.28b204.30 ± 7.78b228.6 ± 6.34a n = 10, F = 20.912, df = 4, P < 0.001Different lowercase letters within the same row indicate significant differences (one-way ANOVA followed by LSD multiple test and Kruskal–Wallis test, P < 0.05). Number of M. usitatus eaten by a C. formosa larva at different temperatures Different lowercase letters within the same row indicate significant differences (one-way ANOVA followed by LSD multiple test and Kruskal–Wallis test, P < 0.05). Developmental Duration at Different Temperatures Developmental duration of C. formosa was also dependent on temperature. For all developmental stages, including egg stage, first instar, second instar, third instar, whole larval stage, prepupal stage, and pupal stage, the order of duration is 15 > 20 > 25 > 30 °C. The duration of all developmental stages tested did not differ significantly between 30 and 35 °C. The results suggested that C. formosa development was promoted by high temperatures (30 and 35 °C) and inhibited by low temperatures (15 and 20 °C) (Table 4). Table 4.Developmental duration of C. formosa at different temperaturesTemperaturesDevelopmental duration (d)EggFirst instarSecond instarThird instarLarval stagePrepupal stagePupal stage15 °C14.76 ± 0.11a13.62 ± 0.65a10.06 ± 0.42a9.75 ± 0.70a32.33 ± 0.70a15.64 ± 0.01a31.60 ± 0.56a20 °C6.44 ± 0.05b5.85 ± 0.12b4.45 ± 0.86b4.96 ± 0.73b15.28 ± 0.19b8.75 ± 0.09b13.43 ± 0.10b25 °C3.88 ± 0.03c3.24 ± 0.08c2.90 ± 0.70c3.92 ± 0.61c10.02 ± 0.14c5.14 ± 0.07c8.24 ± 0.12c30 °C2.98 ± 0.01d2.56 ± 0.08d2.09 ± 0.74d2.68 ± 0.07d7.31 ± 0.13d4.10 ± 0.06d6.00 ± 0.00d35 °C2.84 ± 0.04d2.32 ± 0.09d2.02 ± 0.39d2.68 ± 0.09d6.70 ± 0.15d4.00 ± 0.09d6.00 ± 0.00dStatistics n = 50 to 100, H = 489.954, df = 4, P < 0.001 n = 50 to 100, H = 257.369, df = 4, P < 0.001 n = 50 to 100, H = 236.267, df = 4, P < 0.001 n = 50 to 100, H = 223.242, df = 4, P < 0.001 n = 50 to 100, H = 261.881, df = 4, P < 0.001 n = 50 to 100, H = 245.171, df = 4, P < 0.001 n = 50 to 100, H = 130.455, df = 4, P < 0.001Different lowercase letters within the same column indicate statistically significant differences (Kruskal–Wallis test, P < 0.05). Developmental duration of C. formosa at different temperatures Different lowercase letters within the same column indicate statistically significant differences (Kruskal–Wallis test, P < 0.05). Egg Hatching, Larval Survival, Pupal Weight, and Eclosion at Different Temperatures The hatchability of eggs was significantly higher at 20, 25, and 30 °C than at 15 and 35 °C (F = 28.40, df = 4, P < 0.001). Particularly, the hatching rate at 20 °C was about 60% higher than that at 35 °C. No significant difference in hatching rate was observed between 20 and 30 °C (Fig. 3A). The survival rate of C. formosa larvae was the highest at 25 °C, followed by 30 and 35 °C. The survival rates of the first, second, and third instar larvae kept at 30 °C were 80.38% ± 4.38%, 77.62% ± 4.15%, and 76.62% ± 3.81%, respectively. The larval survival was significantly affected by instar and temperature, but not by the interaction between them (instar: df = 2, P < 0.001; temperature: df = 4, P < 0.001; instar × temperature: df = 4, P = 0.121) (GLM) (Fig. 3B). The weight of C. formosa pupae was significantly affected by temperature and peaked at 20 °C (n = 20 to 86) (H = 120.494, df = 4, P < 0.001) (Fig. 3C). The eclosion rate of pupae was significantly higher at 20 and 25 °C than at 15, 30, and 35 °C (F = 192.774, df = 4, P < 0.001), indicating a suppression of lower and higher temperatures on eclosion of C. formosa (Fig. 3D). Effects of temperature on egg hatching, survival, pupal weight, and emergence of C. formosa. (A) Hatching rate of C. formosa eggs maintained at different temperatures (one-way ANOVA followed by LSD); (B) C. formosa larval survival at different temperatures (GLM); (C) pupal weight of C. formosa kept at different temperatures (Kruskal–Wallis test); and (D) eclosion rate of C. formosa pupa kept at different temperatures (one-way ANOVA, LSD). Different letters above bars represent significant difference (P < 0.05). Discussion In the present study, we investigated the potential of the green lacewing C. formosa against M. usitatus and F. intonsa both in captivity and in the field and revealed its tolerance to low and high temperatures, demonstrating that C. formosa is an ideal biological control agent against the 2 cowpea thrips. The maximum predation amount of C. formosa on M. usitatus emerged at the third instar. A C. formosa third instar larva consumed about 448 M. usitatus second instar nymphs at the highest prey density (Table 1). Chrysopa formosa consumed far fewer M. usitatus than another green lacewing species, Chrysopa pallens, which had a maximum predation rate of 1250 M. usitatus per day (Tang et al. 2017). However, this does not necessarily mean that C. pallens is more efficient than C. formosa in attacking M. usitatus in the field. This is because these predation experiments are carried out in petri dishes, rather than in the field. In natural conditions, thrips, including M. usitatus, always hide in relatively closed spaces, such as buds, flowers, and axils, to avoid natural enemies (Wu et al. 2018). Body size, one of the most fundamental biological traits of an animal, is closely associated with physical activities, such as locomotion, dispersal, and space use (Brose et al. 2006). Theoretically, a small-sized predator can easily go through the narrow gap and reach the small spaces to hunt the prey. In contrast, big predators have more difficulties crawling into the small tissues and have only limited control potential (Bartholomew et al. 2000). By this logic, small-sized C. formosa may be more efficient in controlling M. usitatus than relatively big-sized C. pallens in the field. Chrysopa formosa consumed much more M. usitatus larvae than M. usitatus adults (Tables 1 and 2). This is due to that thrips are usually more mobile at the adult stage than at the larval stage (Sutherland 2006). As a result, adult thrips are more likely to escape the attack of predators. This speculation is based on the higher attacking rates of C. formosa against M. usitatus larvae than against M. usitatus adults and the longer handling time taken for C. formosa preying on M. usitatus adults than preying on M. usitatus larvae (Table 2). Similar results have been reported in the predation of Orius spp. on thrips (Kohno and Kashio 1998, Gitonga et al. 2002). The predation rate of C. formosa showed a linear increase at the initial predation stage, then gradually increased and finally leveled off, leading to a sigmoid shape of the functional response curves. Furthermore, the P1 value was positive, and the P2 value was negative in the logistic regression analysis (Supplementary Table S1). All these results suggested that, on the basis of the previous model (Juliano 2001), the function response of C. formosa to M. usitatus is type III. However, the functional response of Orius spp. to thrips usually belongs to type II. Type III response may reflect an increase in predation efficiency of natural predators by learning and is the only type of functional response that can explain the population regulation by itself (Montserrat et al. 2000, Gitonga et al. 2002). Generally, predators with type III functional response are thought to be more efficient than those with type II functional response in long-term management of prey populations (Holling 1965). From this perspective, C. formosa is superior to Orius spp. in control of M. usitatus thrips. Green lacewings at different instars may exhibit different functional responses to the same prey species. As an example, the Chrysoperla carnea first and second instar larvae showed type II functional response to the 2-spotted spider mites, Tetranychus urticae (Acari: Tetranychidae), whereas the functional response of the C. carnea third instar larvae exhibited type III functional response to this mite (Hassanpour et al. 2009). Similarly, the C. carnea first instar larvae had type II functional response to nymphal instars of the papaya mealybug, Paracoccus marginatus (Hemiptera: Pseudococcidae), whereas the C. carnea second and third instar larvae showed type III functional response to them (Sajjad et al. 2021). The C. carnea first instar larvae showed type II functional response to sugarcane whitefly, Aleurolobus barodensis (Hemiptera: Aleyrodidae). However, both the C. carnea second and third instar larvae had type III functional response to this pest (Sultan and Farhanullah Khan 2014). To summarize, these reports suggested that the type of functional response of C. carnea larvae to their prey was closely related to their own size. In contrast, in the present study, the C. formosa larvae at different instars showed the same type of functional response, type III, to M. usitatus, suggesting that C. formosa responded consistently to the same prey, regardless of predator size. In predation assay, predator size is one of the key factors determining the type of functional response of a predator to its prey (Sajjad et al. 2021). Obviously, the type of functional response of C. carnea to its prey was affected by predator size, but that of C. formosa to its prey was not. The thrip population in the cowpea field was reduced by 71.37% on day 14 after C. formosa release. The control efficiency is comparable to that of chemical insecticides. However, we cannot compare it with that of other predators, as field control of thrips by natural predators has not been reported. It is worth mentioning that the foraging behavior of generalist predators, such as C. formosa, is affected by various factors such as prey abundance, habitat structure, and climatic conditions (Smith et al. 2022). Due to the wide prey spectrum of generalist predators, it is difficult to assess the contribution of C. formosa to the field control of cowpea thrips. In the field experiment, other insect pests, such as Maruca vitrata (Lepidoptera: Crambidae) (Anusha et al. 2016), may occur simultaneously in the cowpea field and act as alternative prey, distracting C. formosa from attacking cowpea thrips. Under this circumstances, the number of thrips consumed by C. formosa will decrease because of the consumption of the alternative prey. Therefore, if cowpea thrips are the only pest in the cowpea field, the control efficiency of C. formosa on them should be higher. So far, various natural enemy insects, such as Geocoris species, Orius spp. Wolff, minute pirate bug, Coccinellidae, and Chrysopidae, have been evaluated for their potential in biological control of thrips (Tang et al. 2017, Wu et al. 2018, Li et al. 2022a, Qiu et al. 2022, Smith et al. 2022). However, all these evaluations were carried out under laboratory conditions. Here, we first assessed the predation ability of C. formosa to cowpea thrips at controlled conditions and then tested the control efficiency of C. formosa on M. usitatus and F. intonsa in the cowpea field, an agricultural ecosystem. In this kind of ecosystem, the activity pattern and foraging efficiency of predatory insects depend on a variety of factors, including temperature, humidity, light intensity, and vegetation structure (Pandit 2023). Therefore, the predation ability of a predator in the lab cannot reflect its foraging performance in the field. By this logic, field experiment is more reliable and convincing than laboratory assay for assessing the potential of a natural predator. Our results lay a foundation for field application of this generalist predator in the management of thrips. Chrysopa formosa showed strong predation capability at high temperatures. Particularly, the number of M. usitatus eaten by C. Formosa second and third instar larva peaked at 35 °C (Table 3), demonstrating that the predation was promoted by high temperature. As insects are highly dependent on the environmental temperature to grow, develop, and regulate various physiological processes and life activities (González-Tokman et al. 2020), these results also implied that C. formosa is physiologically competent for the field control of M. usitatus in subtropical and tropical regions. Though the egg hatching and eclosion of C. formosa was hampered by high temperatures (Fig. 3), this might not undermine the conclusion that C. formosa is an ideal predator of M. usitatus. The reason is that C. formosa was kept at high temperatures all day long in the current temperature assay. However, in nature, the temperature at night always drops, and alternating low and high temperatures increase the hatching rate (Li 2024). Therefore, the high temperature in the daytime might have little impact on egg hatching and eclosion of C. formosa. Supplementary Material Contributor Information Haihang Zhou, Key Laboratory of Natural Enemy Insects of Ministry of Agriculture and Rural Affairs, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China; State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China; Plant Protection College, Shenyang Agricultural University, Shenyang, Liaoning Province, China. Yinling Xia, Key Laboratory of Natural Enemy Insects of Ministry of Agriculture and Rural Affairs, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China; State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China. Lisheng Zhang, Key Laboratory of Natural Enemy Insects of Ministry of Agriculture and Rural Affairs, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China; State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China; Shanghai Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Shanghai, China. Mengqing Wang, Key Laboratory of Natural Enemy Insects of Ministry of Agriculture and Rural Affairs, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China; State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China. Yuyan Li, Key Laboratory of Natural Enemy Insects of Ministry of Agriculture and Rural Affairs, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China; State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China. Zhongjian Shen, Key Laboratory of Natural Enemy Insects of Ministry of Agriculture and Rural Affairs, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China; State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China. Liangzhou Xu, Agricultural Technology Promotion Center in Zhijiang City, Hubei, China. Junming Dong, Qixingtai Agricultural Service Center of Zhijiang City, Hubei, China. Yun Zhao, Qixingtai Agricultural Service Center of Zhijiang City, Hubei, China. Zhengmei Li, Pest Control and Inspection Station of the Agricultural and Rural Bureau of Jianyang District, Nanping City, Fujian, China. Hui Dong, Plant Protection College, Shenyang Agricultural University, Shenyang, Liaoning Province, China. Jianjun Mao, Key Laboratory of Natural Enemy Insects of Ministry of Agriculture and Rural Affairs, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China; State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing, China. Author Contributions Haihang Zhou (Conceptualization [equal], Formal analysis [equal], Investigation [equal], Methodology [equal]), Yinling Xia (Writing—original draft [equal]), LiSheng Zhang (Funding acquisition [equal], Project administration [equal], Supervision [equal]), Mengqing Wang (Project administration [equal], Supervision [equal]), Yuyan Li (Project administration [equal], Supervision [equal]), Zhongjian Shen (Project administration [equal], Supervision [equal]), Liangzhou Xu (Resources [equal]), Junming Dong (Resources [equal]), Yun Zhao (Resources [equal]), Zhengmei Li (Resources [equal]), Hui Dong (Conceptualization [equal], Supervision [equal]), and Jianjun Mao (Conceptualization [equal], Formal analysis [equal], Funding acquisition [equal], Methodology [equal], Project administration [equal], Validation [equal], Writing—original draft [equal], Writing—review & editing [equal]) Supplementary Material Supplementary material is available at Journal of Insect Science online. Funding This research was funded by the National Key R&D Program of China (2023YFD1400600), the Shanghai Eastern Talent Plan (T2024107), the major Special Project of Guizhou Provincial Company, China National Tobacco Corporation (2024XM07), and Major Tasks of the Innovation Project (CAAS-ZDRW202302). Conflicts of Interest None declared. References Associated Data Supplementary Materials