Tetranychus urticae Koch and Tetranychus kanzawai Kishida are important pest mites of various crops of economic importance around the world. Prey consumption and functional responses of two species of phytoseiid mites on these two spider mites were evaluated at 25 ± 1°C, 65 ± 10% RH in the artificial climatic chamber with a photoperiod of 16 h light:8 h dark. The functional response of both Neoseiulus species was type II on three immature stages (egg, larva and protonymph) of Tetranychus species. The value of attack rate coefficients (α) of N. californicus to each stage of Tetranychus was greater than N. longispinosus, and the shortest handling time (Th) was obtained on larvae followed by nymphs and eggs. The maximum attack rate (T/Th) on eggs, larvae and nymphs of T. urticae was estimated to be 11.87, 37.23 and 26.95 for N. californicus, and 18.43, 28.98 and 20.67 for N. longispinosus; the maximum attack rate (T/Tf) on eggs, larvae and nymphs of T. kanzawai was estimated to be 11.90, 42.97 and 39.60 for N. californicus, and 24.15, 31.60 and 24.45 for N. longispinosus. When different densities of prey were offered to the predators, more prey was consumed at higher prey densities, and interaction between prey stage and prey density was significant for N. californicus, but not significant for N. longispinosus. The ability of N. californicus preying on larvae and nymphs of both Tetranychus species was significantly greater than N. longispinosus at high prey densities, but N. longispinosus consumed more eggs than N. californicus.
Introduction
The two-spotted spider mite, Tetranychus urticae Koch, is one of the most economically important pests on a wide range of crops around the world (Farazmand et al. 2012). Two forms are recognized in the two-spotted spider mite; one is the green form, which has green summer females, and other is the red form, which has reddish brown females (Goka et al. 1996). Both forms are distributed widely in China; the green form was named T. urticae, and the red form was named T. cinnabarinus (Boisduval) (Xie et al. 2006; Sun et al. 2012). The Kanzawai spider-mite, T. kanzawai Kishida is also red in color and similar to the red form of T. urticae in morphology and biology (Oku 2008; de Mendonça et al. 2011; Chen et al. 2014).
Neoseiulus californicus (McGregor) is one of the most efficient biological control agents of tetranychid mites. It is widely distributed and has been commercially used in various countries around the world (Gotoh et al. 2004; Canlas et al. 2006; McMurtry et al. 2013; Barbosa & de Moraes 2015). Neoseiulus longispinosus (Evans) is another efficient phytoseiid mite of large number of mite pests, and this species has primarily an Asian distribution (Zhang et al. 1998; Thongtab et al. 2001; Carrillo et al. 2012; Rahman et al. 2013).
During several surveys of pest mites in Guangzhou, Guangdong Province, China, two Tetranychus species (T. kanzawai and T. urticae) and two Neoseiulus species (N. californicus and N. longispinosus) were found in the papaya field, in which the control of mite pests is usually carried out by spraying abamectin. In the first survey, only T. kanzawai and two phytoseiid mites were found. After two surveys, T. kanzawai disappeared, and only T. urticae was left (Song et al. 2013; 2014). In the first taxonomy of Neoseiulus, the Neoseiulus sp. was identified as N. fallacis, but in the following research Xu et al. (2013) classified this species to N. californicus; however there were some differences between Chinese specimens of N. californicus and the specimens collected from elsewhere (Xu et al. 2013; Lv et al. 2016).
Functional response is defined as the relationship between predation rates of a single predator at different densities of a prey per unit time (Farazmand et al. 2012). The functional response of phytoseiids is affected by a number of factors, such as different feeding history (Castagnoli & Simoni 1999), environmental temperature (Gotoh et al. 2004), different prey species (Escudero & Ferragut 2005) and the physical characteristics of host plant (Ahn et al. 2010). Functional response can show the efficiencies of studies and predation rates for predators and evaluate the ability of predators to regulate prey population (Xiao & Fadamiro 2010; Fathipour & Maleknia 2016).
In this study, we separately compared the functional response of N. californicus (Guangdong group) and N. longispinosus (an indigenous species) to two different Tetranychus species in the same conditions. The aim of this study was to see if the efficacies of Neoseiulus to the similar types of preys are different. The predation capacities of N. californicus and N. longispinosus to the Tetranychus are also compared in this study.
Material and methods
Prey Culture
Both species of Tetranychus (green form T. urticae and T. kanzawai) were originally collected from a papaya orchard in Nansha (22°42′N, 113°32′), Guangzhou, China in 2011, and reared on bean (Phaseolus vulgaris L.) leaf discs (rearing arena) at 25 ± 1°C, 65 ± 10% RH in an artificial climatic chamber with a photoperiod of 16 h light: 8 h dark. The leaf discs were placed upside down on a water-saturated sponge, and the leaf edge was surrounded with strips of absorbent paper, in order to minimize the escape of individual mites and providing moisture. Every rearing unit was in a square plastic box (15cm × 15cm × 6.5cm) with a lid, which was holed and covered with fine nylon mesh to allow ventilation. The plastic boxes were filled with water until the water reached just below the leaf disc, and water was added every day.
Predator Source and Rearing
The colony of N. californicus and N. longispinosus were also collected together with Tetranychus in the same location. The predators were reared on the bean leaf disc filled with enough leaf mites (two Tetranychus species mixed). To obtain females predators of the same age for experiments, predator eggs were collected daily and transferred to a new arena with adequate leaf mites (two Tetranychus species mixed).
General Experimental Conditions
The experimental unit consisted in a bean leaf disc (3.5cm diameter) placed upside down on a water-saturated sponge (3.5cm diameter). The leaf discs were surrounded with strips of absorbent paper in order to minimize the escape of individual mites and provide moisture. Individuals trapped in the wet paper surrounding the leaf discs were excluded from data analysis. Every four test units were treated as one test group, and each test group was in a square plastic box as the rearing box. The plastic boxes were filled with water until the water level was just below the leaf disc, and water was added every day. All experiments were conducted at 25 ± 1°C, 65 ± 10% RH and a photoperiod of 16 h light:8 h dark. Before each test, the female adults (emerged in 48h) were fertilized and starved for 24h.
Each mite was transferred with fine hair brush to experimental unit for both functional response experiments, and individuals injured during the transfer were excluded from the experiments.
Functional Response
Seven densities of eggs, larvae and nymphs (protonymph) (2, 4, 8, 16, 32, 64 and 128 eggs, larvae or nymphs) of T. urticae and T. kanzawai were offered separately to the adult female of N. californicus and N. longispinosus. Each treatment was utilized 16 replicates. After 24h, the number of prey eaten was determined by counting intact eggs and the carcasses of larvae or nymphs.
The different densities of prey eggs were obtained as follows: 1–25 adult females of Tetranychus from the stock culture were introduced onto bean leaf disc (experimental unit) and allowed to lay eggs for 10–24 h at the experimental conditions, and then the females were removed. The numbers of prey eggs were adjusted for each experiment by removing excess eggs with a fine brush.
For prey larvae and nymphs, enough adult females of Tetranychus from the stock culture were introduced onto bean leaf disc, and allowed to lay eggs for 10 h at the experimental conditions, and then the females were removed. The discs with eggs were put in an artificial climatic chamber for their development. As soon as the larvae or nymphs emerged; the number of prey (larvae and nymphs) was transferred to the experimental unit with a fine brush.
Data Analysis
The data on functional response were analyzed in two steps (Juliano 2001). First, the logistic regression of the proportion of prey consumed (Ne/N0) as a function of initial density (N0) was used to determine the shape of the functional response. The data were fitted to a polynomial function that describes the relationship between the proportion of preys consumed and initial density:
where (Ne/N0) is the probability a prey will be consumed, and P0, P1 P2 and P3 are the maximum likelihood estimates as being intercept, linear, quadratic and cubic coefficients respectively (Xiao & Fadamiro 2010). The values of P0, P1 P2 and P3 were estimated by using a cubic mathematical function for curve estimation (Table 1). The type of functional response was determined by fitting data to the model (1). The sign of P1 and P2 were used to distinguish the shape of the curves. When the function is negative (P1 < 0), the predator displays a Type II functional response that indicates the proportion of prey consumed declines monotonically with the initial number of prey. When a positive density-dependent result for the proportion of prey consumed (P1 > 0 and P2 < 0) is obtained, the predator displays a Type III functional response (Juliano 2001)
Second, the handling time and the attack rate coefficients of a type II response were estimated using the random predator equation:
where Ne is the number of prey killed; N0 is the initial number of prey; Th is the handling time; and T is the total time available for predator. The data was analyzed using SAS software (SAS 2007). NLIN procedure in SAS was used to estimate the attack rate and the handling time parameters.
The effect of prey density on the daily consumption of Neoseiulus spp. was analyzed by oneway ANOVA followed by Tukey's test (P<O.05) (SAS 2007).
Results
The percentage of prey consumed for each prey stage declined with increasing prey density (Figure 1), which means inverse density-dependence, and the logistic regression for all prey stages had a significant linear parameter P1 < 0 and positive quadratic coefficient (P2) of the proportion of preys consumed at each density versus each initial density of prey (Table 1). This suggests that the functional response of both Neoseiulus species was type II on both stages of Tetranychus species. Therefore, the “random-predator” equation (2) for Type II was used to estimate the attack rate coefficient (a) and the handling time (Th) (Table 2).
The value of attack rate coefficients (a) of N. californicus to each stage of Tetranychus was greater than that of N. longispinosus. The shortest handling time (Th) of each Neoseiulus species was obtained when feeding on larvae of Tetranychus, followed by nymphs, the handling time (Th) when feeding the eggs was the longest. The maximum attack rate (T/Th) on eggs, larvae and nymphs of T. urticae was estimated to be 11.87, 37.23 and 26.95 for N. californicus, and 18.43, 28.98 and 20.67 for N. longispinosus; the maximum attack rate (T/Th) on eggs, larvae and nymphs of T. kanzawai was estimated to be 11.90, 42.97 and 39.60 for N. californicus, and 24.15, 31.60 and 24.45 for N. longispinosus (Table 2). In other words, when Neoseiulus spp. was with three stages of prey (eggs, larvae and nymphs), they tended to consume more larvae, followed by nymphs and eggs.
For N. californicus, when different densities of prey were offered to the predators, more prey was consumed at higher densities of prey and the interaction between prey stage and different density was significant (Table 3). In other words, with increasing density of prey, most larvae were consumed in comparison to nymphs and eggs. For N. longispinosus, when different densities of prey were offered to the predators, the interaction between prey stage and different density was not significant (Table 3); with increasing density of prey, three stages of Tetranychus were consumed equally. The data of prey consumption also showed that N. californicus attacked more larvae and nymphs per day compared to N. longispinosus, and N. longispinosus attacked more eggs per day than N. californicus.
Discussion
Consumption rate of the predators is generally inversely related to prey size (Kasap & Atlihan 2011) but in this study, consumption rate of N. californicus of Tetranychus larvae was the highest, followed by the nymphs, and the consumption rate of eggs is the lowest. This might be due to the fact that larvae and nymphs of Tetranychus fed in clusters which make them relatively easy to handle them. Li et al. (2014) also tested the consumption rate of Chinese N. californicus feeding on T. cinnabarinus at 25°C, and the results showed that the ability of N. californicus controlling three stages of T. cinnabarinus was also larvae>nymphs>eggs. For N. longispinosus, although the mean consumption of N. longispinosus at different densities of three stages of Tetranychus spp. was not significantly different, the data showed that N. longispinosus consumed more eggs than larvae and nymphs; N. longispinosus preferred eggs more. Blackwood et al. (2001) tested N. longispinosus for prey preferences between eggs and larvae, and the results showed that N. longispinosus had strong preference for eggs.
TABLE 1.
Maximum likelihood estimates from logistic regression of proportion of prey consumed as a function of initial prey densities by adults of Neoseiulus californicus and N. longispinosus.
The data from the logistic regression indicate that both N. californicus and N. longispinosus exhibited typical type II functional response on three immature stages of Tetranychus spp. Similar data were obtained for Chinese N. californicus (Qin & Li 2013; Li et al. 2014; Wang et al. 2014) and N. longispinosus (Zhang et al. 1998). Predators with higher searching efficiency (α) and lower handling time (Th) are better agents, although predators exhibiting the Type III functional response are efficient biocontrol agents. Many of the predators that have been successfully released as biological control agents have shown the Type II functional response on their prey (Fathipour & Maleknia 2016).
In previous reports, N. longispinosus was considered to be oligophagous, specialized spider mite predators; and N. californicus was considered to be more polyphagous, generalist predators (Blackwood et al. 2001). However, Croft et al. (2004) suggested N. californicus should be the same type as N. longispinosus. In this manuscript, we showed that N. californicus was more effective against Tetranychus's larvae and nymphs than N. longispinosus, and N. longispinosus was more effective against Tetranychus's eggs.
TABLE 2.
Estimate (±SE) of instantaneous attack rate and handling time of Neoseiulus californicus and N. longispinosus feeding on different stages of Tetranychus urticae and T. kanzawai
Comparing the mean daily consumption of Neoseiulus spp. at different densities of different stages of Tetranychus urticae and T. kanzawai, more T. kanzawai larvae and nymphs were consumed by Neoseiulus spp, although the differences were not significant after analyzed by ANO VA followed by Tukey's test. In the draft of this paper, we analyzed the data by Duncan's Multiple Range Test, and the results showed that the mean daily consumption of Neoseiulus spp. at the higher densities of T. urticae and T. kanzawai were significantly different. This showed that the Tukey's test was more appropriate and precise for multiple comparisons in this paper.
TABLE 3.
Mean (±SE) daily consumption of Neoseiulus californicus and N. longispinosus at different densities of different stages of Tetranychus urticae and T. kanzawai
FIGURE 1.
Mean proportion (±SE) of prey consumed by adult females of Neoseiulus californicus and N. longispinosus at seven densities of three stages of Tetranychus urticae and T. kanzawai.
Phytoseiids are able to perceive chemical cues produced by their spider-mite prey; these chemical compounds involved were called “kairomones” (Sabelis & Van de Baan 1983; Sabelis& Dicke 1985; Dicke & Sabelis 1988; Dicke et al. 1990). Recent research shows that these chemical volatiles were produced by plants infected by herbivores; the volatiles emitted from the plant are called “HIPVs” (herbivore-induced plant volatiles) (Takabayashi et al. 1994; Shimoda et al. 2005; Nachappa et al. 2006; Dicke & Baldwin 2010; Dicke 2015). “HIPVs” comprise a complex mixture of tens up to more than 200 compounds, the composition of which may vary with herbivore species, herbivore developmental instar, plant tissue, and abiotic conditions (Dicke 2015). In this study, at the high density of each prey stage, N. californicus consumed more Tetranychus's larvae and nymphs than eggs, because the volatiles are concentrated at the lower epidermis of the leaves, which are exploited by the spider mites. The more prey on the leaf, the more HIPVs are released (Sabelis & Van de Baan 1983); the larvae and nymphs of spider mites damaged the leaf more directly than the eggs; N. californicus would sense more volatiles from leaves infested with Tetranychus larvae and nymphs. However, N. longispinosus consumed three stages of Tetranychus spp. equally, and the predatory ability of N. longispinosus was lower than the predatory ability of N. californicus; perhaps the abilities of sensing HIPVs between these two Neoseiulus spp. were different. N. californicus and N. longispinosus are all good candidates for the control of Tetranychus mites. To confirm the relationship between the predatory abilities and chemosensory abilities, the chemical sense capacities of these two Neoseiulus spp. to differently treated bean leaves should be measured by the olfactometer, then by SEM and TEM to confirm the composition and structure of the olfactory system.
Acknowledgements
This study was supported in part by a Grant-in-aid from the Science and Technology Program of the National Public Welfare Professional Fund (201103020) from the Ministry of Agriculture of China, a Grant-in-aid from PhD Start-up Fund (S2012040007007) from the Natural Science Foundation of Guangdong Province, and a Grant-in-aid from Science and Technology Project of Guangdong Province, China (the Construction of Green Management R&D Center for Crop Pests in Guangdong Province). We also thank Prof. Xiao-Yue Hong, Prof. Xue-Nong Xu and the anonymous reviewers for reviewing and giving constructive suggestions.
