Fitness of the Papaya Mealybug, Paracoccus marginatus (Hemiptera: Pseudococcidae), after Transferring from Solanum tuberosum to Carica papaya, Ipomoea batatas, and Alternanthera philoxeroides

Simple Summary The papaya mealybug, Paracoccus marginatus, is a polyphagous invasive pest that causes severe damage in China. To improve our understanding of the expansion and prevalence of P. marginatus individuals on host plants, it is important to explore the fitness changes of insects after host plant shifting. In this study, we measured the development, fecundity, and population parameters in P. marginatus individuals over a span of three consecutive generations after being transferred from potato (Solanum tuberosum) to papaya (Carica papaya), sweet potato (Ipomoea batatas), and alligator weed (Alternanthera philoxeroides). Further, the population growth rates of insects on C. papaya, I. batatas, and S. tuberosum in the F2 generation were projected. We found that P. marginatus individuals transferred to C. papaya had higher fitness. When transferred to I. batatas, the fitness of P. marginatus initially decreased in F0 and then rebounded in F1 and F2. Paracoccus marginatus individuals could rapidly expand their populations on the above host plants. However, P. marginatus individuals were unable to complete their development on A. philoxeroides. Our findings provide new insights into the host plant fitness, prevalence, and potential pest control of P. marginatus. Abstract The papaya mealybug, Paracoccus marginatus Williams and Granara de Willink (Hemiptera: Pseudococcidae), is a polyphagous invasive pest in China. The effect that the shifting of the host plant has on the fitness of a polyphagous pest is critical to its prevalence and potential pest control. In order to assess the fitness changes of P. marginatus after transferal from potato (Solanum tuberosum (Tubiflorae: Solanaceae)) to papaya (Carica papaya (Parietales: Caricacea)), sweet potato (Ipomoea batatas (Tubiflorae: Convolvulaceae)), and alligator weed (Alternanthera philoxeroides (Centrospermae: Amaranthaceae)), the life table data of three consecutive generations were collected and analyzed using the age-stage, two-sex life table method. The results showed that when P. marginatus was transferred from S. tuberosum to papaya, a higher intrinsic rate of increase (r) and finite rate of increase (λ) were observed. Paracoccus marginatus individuals transferred to I. batatas had the significantly lower population parameters than those on C. papaya; however, the fitness recovered for those on I. batatas after two generations. Paracoccus marginatus individuals were unable to complete development on A. philoxeroides. Our results conclusively demonstrate that P. marginatus individuals can readily adapt to C. papaya and I. batatas even after host plant shifting, and are capable of causing severe damage to these hosts.


Introduction
The papaya mealybug, Paracoccus marginatus Williams and Granara de Willink (Hemiptera: Pseudococcidae), is a globally invasive pest, which attacks host plants by suck-S. tuberosum as a host. The female life cycle (which differs from that of the male) consists of the egg, three larval instars, and the adult stage, while the male life cycle includes the egg, three larval instars, a pupal stage, and the adult stage.

Life Table Study of P. marginatus
Egg masses laid within a 24 h period on potato leaves were randomly selected for the life table study. In order to accurately observe the lifespan of each insect, the eggs were placed on leaves of C. papaya, I. batatas, S. tuberosum, or A. philoxeroides in plastic dishes (3.5 cm in diameter and 2.0 cm in height) containing agar (3%). After hatching, each 1st instar was transferred into a fresh dish containing leaves of the same plant and reared individually. Following the advice of Mou et al. (2015), only hatched eggs were used in the life table studies to accurately estimate the life table parameters [14]. Newly emerged adult males and females were paired. The daily fecundity and survival were recorded until the death of all individuals. The life table data for three consecutive generations (F 0 , F 1 , and F 2 ) were recorded. Paracoccus marginatus reared on A. philoxeroides only survived for a single generation (F 0 ); therefore, only one life table could be constructed for insects on this host.

Life Table Data Analysis
The raw life history data of all individuals of P. marginatus, including the developmental duration, longevity, and female fecundity, were analyzed according to the age-stage, twosex life table procedure [15,16] using the program TWOSEX-MSChart [17]. The variances and standard errors of parameters were estimated using the bootstrap technique [18,19]. The differences between treatments were assessed using paired bootstrap tests [20]. The age-stage-specific survival rate (s xj ) is the probability that each hatched egg will survive to age x and stage j. The age-specific survival rate (l x ) was calculated as: where k is the number of stages. The age-specific fecundity (m x ) was calculated as: The intrinsic rate of increase (r) was estimated using the Euler-Lotka equation [21,22] with the age indexed from 0 [23]: The finite rate of increase (λ), net reproductive rate (R 0 ), and mean generation time (T) were calculated as follows: λ = e r R 0 = ∞ ∑ x=0 l x m x T = ln R 0 r The age-stage-specific life expectancy (e xj ), i.e., the length of time that an individual of age x and stage j is expected to survive, was calculated according to Chi and Su (2006) [24]: where s iy is the probability that an individual of age x and stage j can survive to age i and stage y by assuming that s iy = 1. The age-stage reproductive value (v xj ), which represents the contribution of each individual in age x and stage j makes to the future population [25,26], was calculated as: The population growth of P. marginatus was simulated according to Chi (1990) [27] by using the computer program TIMING-MSChart [28]. An initial population of 10 newly laid eggs was used for the simulation. The stage growth rate of stage j was calculated according to Huang et al. (2018) [29].
As the population approaches a stable age-stage distribution, the number of individuals of each stage (n j,t ) and the total population size (n total,t ) will increase at the finite rate of increase (λ) and the intrinsic rate of increase (r). These can be expressed as: λn total,t n total,t = log λ r j,t = log n j,t+1 + 1 n j,t + 1 = ln n j,t+1 + 1 − ln n j,t + 1

Development and Fecundity of P. marginatus after Host Plant Shifting
There were no significant differences in egg duration among the four host plants in the F 0 generation. However, the developmental times of female and male nymphs fed on C. papaya were significantly shorter than those on the three other hosts. Extremely long developmental times occurred in both female and male nymphs when fed on A. philoxeroides. The detailed development durations for each instar are contained in Supplementary Table S1. The female adults reared on C. papaya lived significantly longer than those fed on the other three plants, although there was no significant difference between those fed on C. papaya and S. tuberosum. The egg duration was, however, significantly longer in the F 1 and F 2 generations when reared on I. batatas and S. tuberosum. The durations of male nymphs on I. batatas and S. tuberosum were shortened in the F 1 and F 2 generations. The durations of the female nymphs were unchanged when reared on the three host plants. The female adult longevities were unchanged on C. papaya and I. batatas, but were shortened on S. tuberosum. The adult longevities of the males were shortened on I. batatas and S. tuberosum, but unchanged on C. papaya (Table 1). The data (means ± SE) followed by the same letters were not significantly different as assessed by paired bootstrap test (p < 0.05). The lowercase letters in the same row indicated comparisons among different host plants in the same generations, and the capital letters in the same column indicated comparisons among different generations on the same host plants. The 1st instar, 2nd instar, 3rd instar, and pupae samples were combined into nymph samples for the data statistics.
The age-stage life table is capable of describing the stage differentiation; therefore, obvious stage overlapping can be observed. When P. marginatus individuals were reared on A. philoxeroides in F 0 , the probability of an egg surviving to the 2nd instar was extremely low (i.e., 0.150, 11 individuals), and significantly lower than on other host plants. Only two eggs successfully developed into female adults ( Figure 1). In contrast, the survival rates to the 2nd instar when reared on C. papaya were as high as 0.924 and 0.930 in F 1 and F 2 , respectively; higher survival rates to female adulthood (0.489 and 0.490, respectively) were also observed in F 1 and F 2 . Similar high survival rates occurred in the male adults (0.435 in F 1 and 0.440 in F 2 ). Lower survival rates were observed when reared on I. batatas and S. tuberosum ( Figure 2). The narrow distribution of male adult survival curves (s xj ) showed that all male adults had shorter lifespans than the females.
The preadult survival rate of P. marginatus reared on A. philoxeroides in F 0 was extremely low (s a = 0.110), while no significant differences occurred among C. papaya, I. batatas, and S. tuberosum. The preadult survival rate of P. marginatus reared on I. batatas increased to 0.933 in F 2 . Higher proportions of female adults of P. marginatus were observed in F 0 on C. papaya (N f/ N = 0.495) and S. tuberosum (N f/ N = 0.450). The N f/ N value on I. batatas was 0.170. An extremely low N f/ N value (0.023) was observed on A. philoxeroides. In the F 2 generation, the N f/ N values remained constant on C. papaya and S. tuberosum, but increased to 0.367 on I. batatas. In F 0, a significantly high proportion of male adults (N m/ N) of P. marginatus was observed on I. batatas (0.650). The N m/ N values on C. papaya and S. tuberosum were 0.411 and 0.440, respectively. An extremely low N m/ N ratio (0.090) was observed on A. philoxeroides. The N m/ N ratio did not change from F 1 to F 2 ( Table 2).
In the F 0 generation, the highest fecundity (F) of P. marginatus occurred on C. papaya (202.70 hatched eggs/female), which was significantly higher than in the other three plants. Paracoccus marginatus produced, on average, 6.50 eggs/female when reared on A. philoxeroides. None of the eggs produced on this host were viable, so the mean fecundity was zero (Table 2). Lower fecundity rates were observed on I. batatas and S. tuberosum, with 98.83 hatched eggs/female and 127.46 hatched eggs/female, respectively. On I. batatas, the fecundity increased in F 1 (229.50 hatched eggs/female) and F 2 (203.76 hatched eggs/female) ( Table 2).
(202.70 hatched eggs/female), which was significantly higher than in the other three plants. Paracoccus marginatus produced, on average, 6.50 eggs/female when reared on A. philoxeroides. None of the eggs produced on this host were viable, so the mean fecundity was zero (Table 2). Lower fecundity rates were observed on I. batatas and S. tuberosum, with 98.83 hatched eggs/female and 127.46 hatched eggs/female, respectively. On I. batatas, the fecundity increased in F1 (229.50 hatched eggs/female) and F2 (203.76 hatched eggs/female) ( Table 2).    The data (means ± SE) followed by the same letters were not significantly different as assessed by paired bootstrap test (p < 0.05). The lowercase letters in the same row indicated comparisons among different host plants in the same generations, and the capital letters in the same column indicated comparisons among different generations on the same host plants. The age-specific survival rate (lx) curve is the simplified version of sxj; thus, the stage differentiation is not observable. The 50% survival rates of P. marginatus in F0 occurred at 26, 25, 26, and 13 d on C. papaya, I. batatas, S. tuberosum, and A. philoxeroides, respectively ( Figure 3). In F2, the 50% survival rates of P. marginatus on I. batatas and S. tuberosum changed at 26 and 24 d, respectively (Figure 4). Higher curves of the age-specific fecundity (mx) and net maternity (lxmx) were observed on C. papaya in F0. Although there was a relatively high peak of 27 eggs at 40 d on I. batatas, the low survival rate (lx) caused the net maternity rates (lxmx) to be very low. When reared on S. tuberosum, the high peak of mx (18.4 eggs) occurred at 26 d, and the remaining mx values were, for the most part, greater The age-specific survival rate (l x ) curve is the simplified version of s xj ; thus, the stage differentiation is not observable. The 50% survival rates of P. marginatus in F 0 occurred at 26, 25, 26, and 13 d on C. papaya, I. batatas, S. tuberosum, and A. philoxeroides, respectively ( Figure 3). In F 2 , the 50% survival rates of P. marginatus on I. batatas and S. tuberosum changed at 26 and 24 d, respectively (Figure 4). Higher curves of the age-specific fecundity (m x ) and net maternity (l x m x ) were observed on C. papaya in F 0 . Although there was a relatively high peak of 27 eggs at 40 d on I. batatas, the low survival rate (l x ) caused the net maternity rates (l x m x ) to be very low. When reared on S. tuberosum, the high peak of m x (18.4 eggs) occurred at 26 d, and the remaining m x values were, for the most part, greater than 5 eggs (Figure 3). The m x and l x m x values on C. papaya and S. tuberosum did not change significantly; they did, however, increase on I. batatas during the F 1 and F 2 generations (Figure 4).
Insects 2022, 13, x FOR PEER REVIEW 8 of than 5 eggs (Figure 3). The mx and lxmx values on C. papaya and S. tuberosum did not chan significantly; they did, however, increase on I. batatas during the F1 and F2 generatio ( Figure 4). The life expectancy rates of newly laid eggs of P. marginatus were 29.0, 24.5, 29.3, an 14.9 d in F0. The survival rate from the 1st instar to the 2nd instar on A. philoxeroides w extremely low (0.15), and individuals surviving to the 2nd instar could, for the most pa complete their development to adults; hence, the exj curve of the 2nd instar w significantly higher than in the 1st instar ( Figure 5). The detailed exj curves on C. papaya batatas, and S. tuberosum during F1 and F2 are shown in Supplementary Figure S1.
The age-stage-specific reproductive values (vxj) at age zero were exactly equal to t finite rates of increase (λ), i.e., 1.1945, 1.0970, and 1.1475. The vxj increased with age. Wh reared on I. batatas in F0, the vxj curve significantly increased when female adults emerge Similar increases in the vxj curves were observed on C. papaya and S. tuberosum; due to t high percentage of female adults, however, this increase was not obvious. The peak dat of vxj were close to the total preoviposition period (TPOP) ( Figure 6). The detailed curves on C. papaya, I. batatas, and S. tuberosum for F1 and F2 are shown in Supplementa Figure S2. The life expectancy rates of newly laid eggs of P. marginatus were 29.0, 24.5, 29.3, and 14.9 d in F 0 . The survival rate from the 1st instar to the 2nd instar on A. philoxeroides was extremely low (0.15), and individuals surviving to the 2nd instar could, for the most part, complete their development to adults; hence, the e xj curve of the 2nd instar was significantly higher than in the 1st instar ( Figure 5). The detailed e xj curves on C. papaya, I. batatas, and S. tuberosum during F 1 and F 2 are shown in Supplementary Figure S1.
The age-stage-specific reproductive values (v xj ) at age zero were exactly equal to the finite rates of increase (λ), i.e., 1.1945, 1.0970, and 1.1475. The v xj increased with age. When reared on I. batatas in F 0 , the v xj curve significantly increased when female adults emerged. Similar increases in the v xj curves were observed on C. papaya and S. tuberosum; due to the high percentage of female adults, however, this increase was not obvious. The peak dates of v xj were close to the total preoviposition period (TPOP) ( Figure 6). The detailed v xj curves on C. papaya, I. batatas, and S. tuberosum for F 1 and F 2 are shown in Supplementary Figure S2.

Population Parameters of P. marginatus after Host Plant Shifting
There were significant differences in the population parameters in the F 0 generation of P. marginatus after host plant shifting. The highest values of the net reproductive rate (R 0 ), intrinsic rate of increase (r), and finite rate of increase (λ) for P. marginatus occurred on C. papaya (i.e., 100.26 offspring, 0.1778 d −1 and 1.1945 d −1 ). Significantly lower R 0 , r, and λ values were observed when reared on I. batatas and S. tuberosum. Only inviable eggs were produced on A. philoxeroides; thus, the population parameters could not be estimated on this host. The mean generation time (T) of P. marginatus reared on C. papaya was significantly shorter than on I. batatas and S. tuberosum. Although the R 0 , r, and λ values were not significantly changed in the F 1 and F 2 individuals when reared on C. papaya and S. tuberosum, higher values did occur in the F 1 and F 2 generations when reared on I. batatas (Table 3).

Population Parameters of P. marginatus after Host Plant Shifting
There were significant differences in the population parameters in the F0 generation of P. marginatus after host plant shifting. The highest values of the net reproductive rate (R0), intrinsic rate of increase (r), and finite rate of increase (λ) for P. marginatus occurred on C. papaya (i.e., 100.26 offspring, 0.1778 d −1 and 1.1945 d −1 ). Significantly lower R0, r, and λ values were observed when reared on I. batatas and S. tuberosum. Only inviable eggs were produced on A. philoxeroides; thus, the population parameters could not be estimated on this host. The mean generation time (T) of P. marginatus reared on C. papaya was significantly shorter than on I. batatas and S. tuberosum. Although the R0, r, and λ values were not significantly changed in the F1 and F2 individuals when reared on C. papaya and S. tuberosum, higher values did occur in the F1 and F2 generations when reared on I. batatas (Table 3).

Population Projection of P. marginatus
Starting with an initial population of 10 newly enclosed 1st-instar nymphs, P. marginatus could develop to the third generation on C. papaya within 60 d, with a population size reaching as many as 89,552 individuals. However, only two intact generations were observed on I. batatas and S. tuberosum, where the population sizes at 60 d were 12,067 and 15,555 individuals, respectively. When the life tables of the 2.5th and 97.5th percentiles of the net reproductive rate (R 0 ) were used to project the variability of the population growth, the population sizes on C. papaya ranged from 49,033 to 144,038. However, when the life tables of the 2.5th and 97.5th percentiles of the finite rate of increase (λ) were used to project the variability of population growth, the population sizes of P. marginatus ranged from 47,452 to 131,289 (Figure 7). The growth rate curves of all stages fluctuated around the intrinsic rate of increase (r) (Figure 8).

Discussion
With the intention to improve our understanding in the fitness changes of P. marginatus after host plant shifting, we investigated the development, fecundity, and population parameters in P. marginatus within three consecutive generations after being transferred from S. tuberosum to C. papaya, I. batatas, and A. philoxeroides. In addition, the population growth rates of the insects on C. papaya, I. batatas, and S. tuberosum were projected. The study showed that P. marginatus transferred to C. papaya had a higher fitness level. When transferred to I. batatas, the fitness decreased initially and then recovered after two generations. Paracoccus marginatus individuals could rapidly expand their populations on the above host plants. Alternanthera philoxeroides was not suitable for the development of P. marginatus.
Multiple factors such as the population growth and total egg production should be adequately considered when evaluating the fitness of an insect population. The construction and comparison of life tables is the most comprehensive method for describing the population growth, development, survival, and reproduction of a species. Insects of different sexes and stages will usually demonstrate different responses when

Discussion
With the intention to improve our understanding in the fitness changes of P. marginatus after host plant shifting, we investigated the development, fecundity, and population parameters in P. marginatus within three consecutive generations after being transferred from S. tuberosum to C. papaya, I. batatas, and A. philoxeroides. In addition, the population growth rates of the insects on C. papaya, I. batatas, and S. tuberosum were projected. The study showed that P. marginatus transferred to C. papaya had a higher fitness level. When transferred to I. batatas, the fitness decreased initially and then recovered after two generations. Paracoccus marginatus individuals could rapidly expand their populations on the above host plants. Alternanthera philoxeroides was not suitable for the development of P. marginatus.
Multiple factors such as the population growth and total egg production should be adequately considered when evaluating the fitness of an insect population. The construction and comparison of life tables is the most comprehensive method for describing the population growth, development, survival, and reproduction of a species. Insects of different sexes and stages will usually demonstrate different responses when exposed to variations in their host plants, the numbers and composition of their biological enemies, extreme climate conditions, and pesticides, and consequently it is necessary to take all of these into consideration prior to formulating an effective pest management strategy [30][31][32][33]. In order to accomplish this, life tables are fundamental to achieving a comprehensive assessment of a population's fitness on a given host plant. Thus, it was important to use the age-stage, two-sex life table method to assessed the fitness changes that occurred after host plant shifting in P. marginatus.
The age-stage, two-sex life table not only includes the male component of a population, but is also capable of describing the overlapping and differentiation of each stage [29]. Although the males and females of P. marginatus have different numbers of developmental stages, the stage differentiation can still be precisely described.
The hatch rates of eggs vary with the age of the female adults; hence, using only hatched eggs will enable a more accurate estimate of the population parameters being studied [14,34]. The highest fecundity of P. marginatus was observed on C. papaya (F = 215.27 hatched eggs/female). Seni et al. (2015) reported the fecundity of P. marginatus on C. papaya as 291 total eggs/female (greater than 215.27 hatched eggs/female) [35]; however, the hatch rate was omitted in their study.
By using the age-stage, two-sex life table, He et al. (2021) reported a longer developmental duration and lower intrinsic rate of increase for Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noctuidae) when reared on soybean, while a shorter developmental duration and higher intrinsic rate occurred on sunflower [36]. Karimi-Pormehr et al. (2018) reported a shorter developmental time, higher survival rate, and greater fecundity in Sitotroga cerealella (Olivier) (Lepidoptera: Gelechiidae) on a more suitable cultivar ('19A 1 ) of barley, while noting a longer developmental time and lower fecundity when reared on a less suitable cultivar ('Fajr30 ) [37]. In this study, when P. marginatus was reared on C. papaya, the developmental durations of the 1st, 2nd, and 3rd instar (female) individuals were significantly reduced (Supplementary Table S1); however, the reverse occurred when reared on I. batatas and S. tuberosum. While the fecundity of P. marginatus was significantly higher on C. papaya overall in this study, it was significantly lower on I. batatas (F 0 ). The insects have trade-offs between development and reproduction. When the basic 'development' need of insects are met by suitable host plants, insects tend to allocate more energy to reproduction.
Our results showed that P. marginatus reared on C. papaya had a significantly higher proportion of female adults (N f /N), while a lower N f /N occurred on I. batatas in the F 0 generation. Lewontin (1965) demonstrated that the first age of reproduction plays an important role in the values of r and λ [38]. When P. marginatus was reared on C. papaya, reproduction in the F 0 generation started at 18 d, but advanced to 15 d in F 2 . The threeday change resulted in the value of r increasing from 0.1778 d −1 (F 0 ) to 0.1824 d −1 (F 2 ), while λ increased from 1.1945 d −1 (F 0 ) to 1.2000 d −1 (F 2 ) (Figures 4 and 5). Consequently, this change resulted in P. marginatus reared on C. papaya having higher values for their population parameters (r and λ) due to their higher survival and fecundity rates on this host. The opposite was true when reared on I. batatas and S. tuberosum.
By using the age-stage, two-sex life table, the stage structure and fluctuations in growth rate in different stages can be observed using population projection. In addition, the life tables constructed based on the 2.5th and 97.5th percentiles of R 0 and λ can be used to disclose the variabilities that occur during population growth [29].
When host plant shifting happens, the fitness of the insect population to the new host plant may recover after a few generations. Quezada et al. (2015) showed that Choristoneura fumiferana (Clemens) (Lepidoptera: Tortricidae) consecutively reared on less nutritional host plants for three generations would show an adaptive response [39]. Meihls et al. (2008) demonstrated that after three generations of being reared on Bt corn, the survival rate of Diabrotica v. virgifera (LeConte) (Coleoptera: Chrysomelidae) was comparable to beetles reared on normal corn [40]. In this study, when P. marginatus transferred from S. tuberosum to C. papaya, all population parameters were significantly higher than on other plants during three generations. This demonstrated that even though P. marginatus initially survived on S. tuberosum for multiple generations, the insects transferred to C. papaya still had a higher fitness level. However, after transferal to I. batatas, the fitness of P. marginatus initially decreased in F 0 and then rebounded in F 1 and F 2 . Paracoccus marginatus showed a higher ability to recover fitness on I. batata. Based on the observation that females were unable to produce viable eggs on A. philoxeroides, we concluded that this host was unsuitable for P. marginatus. This differences in the fitness of P. marginatus to host plants may be due to the volatiles, nutrients of host plants, and so on (unpublished data from the authors).
The age-stage, two-sex life table has been used in a number of studies involving the adaptation of insects on different host plants. Guo et al. (2021) reported that compared with being reared on potato and tobacco, Spodoptera frugiperda reared on maize exhibited a shorter developmental time in the larval period, more female individuals, and a higher reproductive rate [41]. Nemati-Kalkhoran et al. (2018) reported the life table characteristics of Rhyzopertha dominica (Coleoptera: Bostrichidae) on different barley cultivars, demonstrating that a higher net reproductive rate and intrinsic rate of increase occurred on the cultivar 'Mahoor' [42]. Jaleel et al. (2018) reported that Bactrocera dorsalis (Diptera: Tephritidae) females produced more eggs on guava than banana [43].
Cipollini and Peterson (2018) pointed out the potential effects of host shifting, including the importance of phytophagous insects being able to find and utilize their ancestral hosts, potentially leading to host range expansions [44]. The present study reports the fitness of P. marginatus after transferal from S. tuberosum to C. papaya, I. batatas, and A. philoxeroides. Ipomoea batatas and S. tuberosum are important food crops and C. papaya is an important fruit [45][46][47]. Our results demonstrate the potential damage of P. marginatus to I. batatas and S. tuberosum, and again verify the severe damage of P. marginatus to C. papaya, even if the insects transfer from suboptimal host plants. These results indicate that outbreaks of P. marginatus are possible in the future, and should they occur may result in serious economic damage. This study provides new insights into the host plant fitness, prevalence, and potential pest control of P. marginatus.