Introduction

The invasive mealybug, Phenacoccus solenopsis Tinsley (Hemiptera: Pseudococcidae) was first documented as a pest of cotton in Texas, United States (Fuchs et al. 1991) and now widely distributed in various ecological zones of the globe (García Morales et al. 2016). During the last decade, this mealybug species has emerged as a serious threat to cotton in India and Pakistan, causing severe economic losses (Wang et al. 2010) and is now widespread in almost all cotton growing states of India (Nagrare et al. 2009). Since it is an exotic pest, biological control must be attempted as an effective means of its management (Ram and Saini 2010).

Encrytids are the most successful natural enemies used in the biological control programs against mealybugs (Noyes and Hayat 1994). Aenasius arizonensis (Girault) is an important koinobiont nymphal endoparasitoid of P. solenopsis reported in India (Hayat 2009), Pakistan (Mahmood 2008), China (Chen et al. 2010), Iran (Abdin et al. 2012) and Australia (Khan et al. 2012). It is a key mortality factor of P. solenopsis under field conditions (Ram et al. 2009). Field parasitism by A. arizonensis to an extent of 95% has been recorded in P. solenopsis (Khuhro et al. 2011). It was first time reported by Mahmood (2008) in Pakistan and later, was described and named as A. bambawalei by Hayat (2009). However, Fallahzadeh et al. (2014) from Iran reported that this species is closer to Aenasius arizonensis (Girault) and reported on A. bambawalei as a junior synonym of A. arizonensis.

Host life stage is one of the key factors influencing oviposition preference, sex allocation and offspring fitness in solitary parasitoids (King 1989; Hagvar and Hofsvang 1991; Harvey et al. 1994; Kant et al. 2012). Furthermore, the progeny fitness is positively related to host resources; consequently, the parasitoid tends to allocate the offspring sexes according to host-size-dependent sex allocation principle (Charnov 1982; Godfray 1994; West and Sheldon 2002; West 2009). Parasitoids make a series of oviposition decisions while locating their hosts to achieve maximum fitness gains which include the number of eggs laid, the ratio of fertilized to unfertilised eggs and whether to superparasitize the host or not. Such decisions are relatively influenced by the quality of hosts encountered by the parasitoids (Kant et al. 2012). Earlier studies by many workers on different parasitoids have reported that host stage at the time of parasitism affects their reproductive fitness, resulting in altered developmental time (Sequeira and Mackauer 1993), body size (Liu 1985; Lampson et al. 1996) and progeny sex ratio (Godfray 1994; Napoleon and King 1999).

Most of the studies on host stage preference of A. arizonensis (=A. bambawalei) have been carried out by earlier workers on the basis of parasitization rate under choice and no choice tests. (Fand et al. 2011; Abdin et al. 2012; Vijaya and Ram 2013; Badshah et al. 2016; Iqbal et al. 2016; Zhang et al. 2016). Furthermore, the biology of A. arizonensis has been studied earlier on one host stage only, either on 3rd instar nymphs or adult host stage (Poorani et al. 2009; Fand et al. 2011; Abdin et al. 2013; Vijaya and Ram 2013). Information about whether and to what extent different host stages of P. solenopsis affect the detailed biological parameters of this solitary endoparasitoid is still lacking. Moreover, comprehensive biology of A. arizonensis in cotton crop seasons with varying temperature and humidity conditions is not available. The aim of this study was to investigate the effect of different crop seasons and host stages on the overall development, reproductive fitness and offspring sex allocation by A. arizonensis on cotton. This information could be helpful in developing an optimal mass rearing program for this parasitoid and devising an effective biological control strategy for P. solenopsis in cotton ecosystem.

Materials and methods

The biological parameters of A. arizonensis were studied on three nymphal instars (1st, 2nd and 3rd) and adult stage of mealybug, P. solenopsis in the screen house (4 m × 3 m × 2 m) fitted with 30 mesh, galvanized wire gauge, at the Entomological Research Farm, Punjab Agricultural University (PAU), Ludhiana during the pre-monsoon (May–June), monsoon (July–August) and post-monsoon (September–October) seasons during 2015. The meterological data were recorded using digital thermohygrometer (HTC Instruments, India). The mean temperature ranged between 25.2 to 28.7 °C in the post-monsoon (September–October), 29.9 to 30.3 °C in the monsoon (July–August) and 31.7 to 31.8 °C in the pre-monsoon (May–June) season. The mean relative humidity during the post-monsoon season was 68.0 to 72.9%, while it was 74.5 to 76.5% in the monsoon season and 31.5 to 49.5% in the pre-monsoon season.

Raising of host plant

The seeds of Bt cotton hybrid (NCS 855 BG II) were sown in earthen pots filled with soil and farm yard manure (1:1) and plants were raised without any plant protection measures (Anonymous 2015). These pots were placed in sunlight till germination. After germination, pots were transferred to screen house to avoid any pest infestation. Only one healthy seedling was retained in each pot for conducting experiments.

Rearing of P. solenopsis

P. solenopsis was reared on sprouted potatoes (Solanum tuberosum L.) according to the procedures outlined by Nagrare et al. (2011). Small to medium sized potato tubers were washed, air dried and held at room temperature in the dark until sprouting. When the potato sprouts were 2.5–5 cm in length, they were placed in glass jars (4–5 tubers/ jar) of size 20 × 15 cm, with moist soil at the base to maintain turgidity of the sprouted tubers. The sprouts were inoculated with gravid females with the help of a soft camel hair brush. The jars were then covered with muslin for the establishment of mealybug colonies. The culture of P. solenopsis was maintained for multiple generations, which was used for rearing of A. arizonensis and for the experiments.

Rearing of A. arizonensis

Parasitized mealybugs (mummies) were collected from mealybug infested cotton, Hibiscus sp., and weed hosts (Abutilon sp., Sida sp. and Parthenium sp.) and brought to the laboratory. Naive/virgin A. arizonensis parasitoid adults obtained from mummies of mealybug were sexed (Hayat 2009) and were released in glass jars containing healthy colonies of mealybug maintained on sprouted potatoes. After parasitism, the mummies were collected and kept in glass vials plugged with cotton wool. These glass vials were provided with the honey streak on a paper strip as a source of food for the emerging adult parasitoids. The newly emerged adult parasitoids were used to study various biological parameters.

Biological parameters of A. arizonensis on different stages of mealybug, P. solenopsis during pre-monsoon, monsoon and post-monsoon seasons

Twenty individuals (5 replicates; n = 100), from each of the three nymphal instars (1st, 2nd and 3rd) and adult stage of P. solenopsis were transferred from the maintained insect culture on different Bt cotton plants (NCS 855 BG II) with a soft camel hair brush. They were kept in separate wooden screen cages (15 cm × 15 cm × 20 cm). A pair of freshly emerged adults of the parasitoid was released in each cage for 48 h and provided with cotton swab dipped in 10% honey solution as a source of food for A. arizonensis adults. After parasitism, mummies were collected and kept in separate plastic vials (45 mm diameter × 60 mm height) provided with the honey streak on a paper strip as a source of food for the emerging adult wasps. Observations were recorded on the total immature period (oviposition to adult emergence), proportion of adults emerged from mummies of P. solenopsis in progeny (adult emergence), sex ratio (proportion female), adult longevity of the parasitoid. The adult emergence (proportion) was worked out by dividing the number of adults emerged with number of mummified mealybugs.

$$ \mathrm{Adult}\ \mathrm{emergence}\ \left(\mathrm{proportion}\right)=\frac{\mathrm{Number}\ \mathrm{of}\ \mathrm{adult}\ \mathrm{parasitoids}\ \mathrm{emerged}}{\mathrm{Total}\ \mathrm{number}\ \mathrm{of}\ \mathrm{mummies}} $$

The oviposition period was recorded by releasing a couple of 1 day-old A. arizonensis adults in separate aluminium screen cages (10 cm × 10 cm × 10 cm) having 30 individuals (5 replicates; n = 150), from each of the three nymphal instars (1st, 2nd and 3rd) and adult stage of P. solenopsis. All the experiments were replicated five times. After every 24 h, these adults were removed and transferred to new jars for determining daily fecundity. The total number of mummies and un-mummified mealybugs were recorded daily. Un-mummified mealybugs were subjected to xylene treatment (Souissi and Le Ru 1997). Total fecundity per female was calculated by adding the number of mummies and the number of un-mummified mealybugs showing signs of encapsulation. A cotton swab dipped in 10% honey solution was hung in each cage as a source of food for A. arizonensis adults.

Statistical analysis

Data are presented as mean ± standard error of the mean (SE). The data on the proportion of adults emerged and sex-ratio (expressed as proportion of females) in progeny were analysed by means of a logistic ANOVA model: Generalized Linear Models (using a logit link function) assuming quasi binomially distributed error variances with significance assessed through model simplification (Crawley 1993; Wilson and Hardy 2002). Total fecundity of A. arizonensis was analyzed using a Generalized Linear Model with a poisson error distribution (using a log link function) (Yamane et al. 2015). The data on total development period of immature A. arizonensis, female longevity and ovipositional period of A. arizonensis adults were analyzed using Generalized Linear Models with normal error distributions (using an identity link function). Generalized Linear Models results are presented in the form of Wald Chi-square χ2 test and p values. Pair-wise tests were performed by choosing appropriate contrasts, and an overall significance level of 5% was maintained using a modified Bonferroni procedure (Hochberg 1988). All statistical tests were carried out using IBM SPSS 22.0 for Windows (IBM Corporation, Armonk, New York, USA).

Results

The mean temperature in the post-monsoon (26.9 °C) differed significantly from the mean temperature in the monsoon (30.1 °C) and the pre-monsoon (31.8 °C) season (Wald χ2 = 136.55; df = 2; p = <0.0001). The mean relative humidity in the post-monsoon (70.3%) also differed significantly from the mean relative humidity in monsoon (75.4%) and pre-monsoon (43.5%) season (Wald χ2 = 361.41; df = 2; p = <0.0001).

Total immature period

A. arizonensis females parasitized all the stages of mealybug, P. solenopsis except the 1st instar nymphs. For both sexes, total immature period, i.e. from oviposition to adult emergence also showed significant differences among different seasons as well as host stages. Among different seasons, the period was significantly shorter (11.87 days) in the post-monsoon season, followed by monsoon season (13.02 days), whereas, it was longest in pre-monsoon season (14.51 days) for male parasitoids (Table 1; Wald χ2 = 45.58; df = 2; p < 0.0001). For female parasitoids, the total immature period was significantly shorter (13.56 days) in post-monsoon season; and it was not statistically different from the monsoon season (13.74 days) (Wald χ2 = 18.69; df = 2; p = 0.0004). For both sexes, developmental time decreased with increase in host age and males developed within a shorter period of time as compared to females. The total immature period of A. arizonensis males was significantly longer (15.01 days) in 2nd instar followed by 3rd instar nymphs (13.20 days), while it was least (11.19 days) in adult host stage (Wald χ2 = 37.47; df = 2; p < 0.0001). For females also, it was significantly shorter in adult host stage as compared to those recorded in 3rd instar host stage (Table 1; Wald χ2 = 23.56; df = 1; p < 0.0001). The interaction analysis for the total immature period between different seasons and host stages was found to be significant for male (Wald χ2 = 59.03; df = 4; p < 0.0001), and female (Wald χ2 = 43.58; df = 2; p < 0.0001) parasitoids.

Table 1 Effect of different seasons and P. solenopsis stages on the total immature period (oviposition to adult emergence) of A. arizonensis
Full size table

Proportion of adults emerged in progeny and sex ratio (proportion female)

The proportion of A. arizonensis adults emerged from P. solenopsis mummies was significantly influenced by seasons (Fig. 1; Wald χ2 = 41.37; df = 2; p < 0.0001) but no significant differences were observed in proportion of A. arizonensis adults emerged from different host stages of P. solenopsis (Wald χ2 = 2.84; df = 2; p = 0.200). Moreover, the interaction between different seasons and host stages was also non-significant for proportion of adults emerged in progeny (Wald χ2 = 0.06; df = 4; p = 0.786). Significantly higher proportion of A. arizonensis adults (0.99) emerged from mummies in the post-monsoon season, followed by monsoon (0.84) and pre-monsoon seasons (0.74). Among host stages, it was highest (0.89) from mummies of 3rd nymphal instar and it was not statistically different from those which emerged from adult hosts (0.87) (Fig. 1). Only males emerged from parasitized 2nd instar P. solenopsis nymphs, whereas, it was female-biased in the later host stages. Between the two stages, the relatively high proportion of A. arizonensis female wasps emerged from 3rd instar P. solenopsis nymphs (0.59) as compared to adult host stage (0.57) (Fig. 2). However, there was no significant interaction between season and host stage on the proportion of A. arizonensis females in progeny (Wald χ2 = 0.27; df = 4; p = 0.801).

Fig. 1
figure 1

Proportion of A. arizonensis adults emerged from different stages of P. solenopsis during different crop seasons

Full size image
Fig. 2
figure 2

Proportion of A. arizonensis females emerged from different stages of P. solenopsis during different crop seasons. *only males emerged from the progeny of 2nd instar P. solenopsis nymphs

Full size image

Adult longevity

The longevity of female adults emerged from P. solenopsis mummies was significantly influenced by seasons (Table 2; Wald χ2 = 55.51; df = 2; p < 0.0001) and host stages of P. solenopsis (Wald χ2 = 34.21; df = 1; p < 0.0001). Moreover, the interaction between seasons and host stages was also significant for longevity of female adults (Table 2; Wald χ2 = 51.05; df = 2; p < 0.0001).

Table 2 Effect of different seasons and P. solenopsis stages on the longevity of A. arizonensis females
Full size table

Oviposition period

The oviposition period of A. arizonensis females was significantly higher in the post-monsoon season (13.33 days), followed by the monsoon season (6.53 days), while it was lowest (5.93 days) in pre-monsoon season (Table 3; Wald χ2 = 42.20; df = 2; p < 0.0001). Among the host stages, the oviposition period of A. arizonensis females was significantly higher (9.67 days) in adult host stage, followed by 3rd instar (8.93 days) and 2nd instar (7.20 days) host stages (Wald χ2 = 32.90; df = 2; p < 0.0001). The interaction between seasons and host stages was also significant for the oviposition period of female adults (Table 3; Wald χ2 = 63.69; df = 4; p < 0.0001).

Table 3 Effect of different seasons and P. solenopsis stages on the oviposition period of A. arizonensis females
Full size table

Total fecundity

Total fecundity of A. arizonensis females showed significant differences among different seasons (Wald χ2 = 63.15; df = 2; p = <0.0001), host stages (Wald χ2 = 46.27; df = 2; p = <0.0001) and their interaction (Fig. 3; Wald χ2 = 72.27; df = 4; p = <0.0001). It was significantly higher (82.80 eggs/female) in the post-monsoon season, followed by monsoon (31.47 eggs/female) and pre-monsoon seasons (20.53 eggs/female). Among the host stages, the fecundity was significantly higher (59 eggs/female) when females were offered 3rd instar P. solenopsis nymphs, followed by adult host stage (47.53 eggs/female); however, it was lowest in 2nd instar nymphs (28.27 eggs/female) (Fig. 3).

Fig. 3
figure 3

Total fecundity of A. arizonensis females on different stages of P. solenopsis during different crop seasons

Full size image

Discussion

The amount of nutritional resources available during larval development is a key determinant for progeny fitness in parasitoids (Jervis et al. 2008; López et al. 2009). Host size varies due to nutritional resources which are considered to be an index of host quality (Sequeira and Mackauer 1992; Godfray 1994). Furthermore, parasitoids may differ in their strategy for utilizing/exploiting the host body resources in accordance with the host size or host stage. Our studies showed that A. arizonensis parasitized all the stages of its host P. solenopsis except 1st instar nymphs. No parasitism in 1st instar nymphs may be due to inadequate resources available for the parasitoid at this stage for their survival and progeny development (Hu et al. 2003; Colinet et al. 2005; Chong and Oetting 2006). Our results are consistent with the findings of earlier workers (Fand et al. 2011; Abdin et al. 2012; Vijaya and Ram 2013; He et al. 2015; Iqbal et al. 2016; Zhang et al. 2016). Less preference for younger host stages have also been reported in other parasitoid species like Aenasius vexans Kerrich on Leptomastix dactylopii Howard and Planococcus citri (Risso) (Jong De and van Alphen 1989), A. vexans on Phenacoccus herreni Cox and Williams (Bertschy et al. 2000) and Anagyrus loecki Noyes and Menezes on Phenacoccus madeirensis Green (Chong and Oetting 2006).

Our findings showed that development period of A. arizonensis from oviposition to adult emergence was longer in younger stages and decreased with increase in host age for both sexes. This can be attributed to inadequate resources available at younger stages of the host for the immature stages of the parasitoid (Hu et al. 2003; Colinet et al. 2005; Chong and Oetting 2006) which might have delayed the duration. The present results corroborate with the findings of Fand et al. (2011), Abdin et al. (2013) and Vijaya and Ram (2013) who also found that total developmental period of A. bambawalei (= A. arizonensis) was relatively shorter in adult host stage as compared to 3rd and 2nd instar nymphal stages. Godfray (1994) also concluded that higher fitness costs were involved in the parasitism of small hosts by koinobiont parasitoids, as the developmental time is often lengthened and survival is often reduced. Decreased developmental time in larger hosts has also been reported to be a common feature in the development of different parasitoid species (Karamaouna and Copland 2000; Neveu et al. 2000).

We report here that the progeny of A. arizonensis that emerged from older host stages of P. solenopsis was female-biased which corroborate with the studies conducted by Fand et al. (2011) and Abdin et al. (2013). Similar findings have also been reported in other parasitoid species like Leptomastix epona (Walker) on Pseudococcus viburni (Signoret) (Karamaouna and Copland 2000), Aphidius ervi (Haliday) on pea aphids (Sequeira and Mackauer 1993) and A. vexans on P. herreni (Bertschy et al. 2000). Fand et al. (2011) and Abdin et al. (2013) reported that the progeny of A. bambawalei (= A. arizonensis) from 2nd instar nymphs produced primarily males, however in the present studies, progeny from 2nd instar P. solenopsis nymphs yielded only males as no females emerged from 2nd instar host stage.

Our results on sex allocation suggest an innate ability of A. arizonensis to recognise second instar hosts as less suitable for females. Males can become more abundant due to differential mortality, even when females invest equally in both sexes during oviposition (Hardy 1992; King 1993; Hardy et al. 1998; Ode and Hardy 2008). Haplodiploidy in parasitoids allows females to manipulate investment in each sex through the decision of whether or not to fertilize an egg; hence females can adjust the sex ratio of their offspring at the time of oviposition. More female progeny in older stages can be explained as per host size-dependent sex allocation model given by Charnov et al. (1981), which suggested that a higher proportion of females may be produced from larger size hosts due to greater nutritional requirement and reproductive benefits for the female progeny. Females of the hymenopteran parasitoids are known to control the sex of their offspring by a haplodiploid sex determination mechanism and they preferentially lay female (diploid) eggs in larger hosts and male (haploid) eggs in smaller hosts (King 1993).

The size of female wasps is an important measure of parasitoid fitness and is known to influence other fitness parameters like mating capacity, longevity, fecundity and reproductive rate (Jervis and Copland 1996; Harvey 2005). Large sized insects tend to live longer than small ones (Godfray 1994). This positive relationship between adult body size and their longevity has been reported for several Hymenopteran parasitoid species (Liu et al. 2001; Babendreier 2007; Gulec et al. 2007). For both sexes, adults of A. arizonensis that emerged from larger hosts lived longer as compared to those which came out from younger host stages. This may be attributed to inadequate/ reduced body resources available at younger stages of the host to the parasitoid progeny (Colinet et al. 2005; Chong and Oetting 2006). Godfray (1994) also showed that higher fitness costs were involved in the parasitizing of small hosts by koinobiont parasitoids, as the survival is often reduced. However, Fand et al. (2011) have reported that adult longevity was significantly higher in 3rd instar P. solenopsis nymphs followed by adult and 2nd instar host stages. The development time of immature stages was longer for females than males. Likewise, female adults live longer as compared to male adults.

Most of the earlier studies on parasitism and sex allocation were conducted under constant (Fand et al. 2011; Abdin et al. 2013) and different (Zhang et al. 2016) temperature conditions under laboratory conditions. In this study, we tested the effect of different crop seasons with varying temperature and humidity conditions on the A. arizonensis under screen house conditions. Our results demonstrate that post-monsoon season (September–October months) with mean temperatures varying from 25 to 29 °C along with 68 to 73% relative humidity was the best season for the parasitoid in terms of the shorter development period, higher adult emergence, and more fecundity. Moreover, the parasitism in this season also yielded more females in the progeny. The prevailing weather conditions particularly temperature and humidity during September–October months might have contributed to higher fitness of the parasitoid. The mean temperature ranged between 25.2 to 28.7 °C in the post-monsoon (September–October), 29.9 to 30.3 °C in the monsoon (July–August) and 31.7 to 31.8 °C in the pre-monsoon (May–June) season. The mean relative humidity during the post-monsoon season was 68.0 to 72.9%, while it was 74.5 to 76.5% in the monsoon season and 31.5 to 49.5% in the pre-monsoon season. The parasitism rate and biological attributes of A. arizonensis were severely affected during the pre-monsoon season under heat stress. Our results are consistent with the findings of He et al. (2018) who reported that emergence rate of A. arizonensis adults increased significantly with an increase in temperature from 19 to 31 °C, but dropped when the temperature exceeded 31 °C. The emergence of the parasitoid adults could be affected by high temperatures due to increasing rates of evaporation from host egg leading to desiccation and killing the parasitoid progeny inside (Mohammad et al. 2015). However, Zhang et al. (2016) studied the parasitism of A. bambawalei (= A. arizonensis) on P. solenopsis at different temperature conditions ranging from 21 °C to 39 °C under laboratory conditions and found the more beneficial effect in the temperature range of 30-36 °C.

The third instar P. solenopsis nymphs appear to be the most appropriate hosts for a mass-rearing program because the parasitoid gains greater fitness return and yields more female progeny which is highly desirable for increasing the efficiency of any biological control program. Since the parasitoid had the higher parasitism, emergence rate and female ratio in the post-monsoon season (September–October), we conclude that the temperature varying from 25 to 29 °C and relative humidity from 68 to 73% are optimum for the mass rearing of A. arizonensis. The findings of the study will help in mass rearing of the parasitoid vis-à-vis biological management of P. solenopsis.