Demographic comparison of sweetpotato weevil reared on a major host, Ipomoea batatas, and an alternative host, I. triloba

Abstract

In this study, we collected life table data for the sweetpotato weevil, Cylas formicarius, grown on Ipomoea batatas and Ipomoea triloba and analyzed them using an age-stage, two-sex life table. We also demonstrated the growth potential of C. formicarius on these two host plants by using population projection. These data will be useful to the growers to the selection or eradication of host plants in an integrated control strategy for C. formicarius for the entire area of the targeted areas. We found that C. formicarius developed faster on I. batatas than on I. triloba. The developmental times of the larval and pupal stages on I. batatas than on I. triloba were 37.01 and 8.3 days. The adult females emerged before and began to produce eggs at 42 days earlier when reared on I. batatas. The fecundity of females was 90.0 eggs on I. batatas significantly higher than the mean fecundity of 68.5 eggs on I. triloba. Although this insect has a higher intrinsic rate of increase on I. batatas, the study indicated that C. formicarius can successfully survive and reproduce on both host plants.

Introduction

A life table is a convenient and comprehensive method for summarizing the survival and reproductive potential of a population. Using a life table, one can compare the growth potential of an insect on different host plants, under different environmental conditions^1,2,3. Because the age-stage, two-sex life table can describe stage differentiation and include both sexes, it can precisely reveal the actual life history of the insect species and life table have been widely applied in the study of various ecological aspects of interest in relation to insect pests and their natural enemies⁴. Understanding the demography of an insect under variable conditions is the cornerstone for developing an environmentally friendly pest management strategy.

The sweetpotato weevil, Cylas formicarius (F.) (Coleoptera: Brentidae), is the most serious pest of sweet potato (Ipomoea batatas [L.] Lam., Convolvulaceae)⁵ both in the field and in storage⁶. In the Marianas Islands, Ipomoea triloba L. (Convolvulaceae) (littlebell, or Aiea morning glory) is widespread and serves as an alternative host for C. formicarius. This vine is reported to grow around sweet potato fields and has been recorded as an alternate host of another pest of sweet potato Euscepes postfasciatus (Fairmaire) (Coleoptera: Brentidae)⁷. In addition to I. batatas, the major host plant of C. formicarius⁵, at least 49 other members of the Convolvulaceae have been recorded as hosts for C. formicarius, which has been recorded feeding on seven genera in six tribes within this plant family^8,9. In Guam and other Micronesian Islands, I. triloba is widespread and serves as an alternative host for C. formicarius¹⁰. Because of the cryptic nature of the larvae and the nocturnal activity of the C. formicarius adults, it is difficult to control this pest using chemicals¹¹. In addition, the understanding the life history of C. formicarius will help and makes the pest easiest to control with long-residual pesticides that are now out of favor and often unavailable⁵.

To systematically understand the role of main and alternative hosts in the population explosion of C. formicarius, information on its demography, including development, survival and fecundity on each host plant, is needed. In this study, we measured these life table parameters of C. formicarius on I. batatas and I. triloba and analyzed them using an age-stage, two-sex life table. Furthermore, we estimated the potential rate of growth of C. formicarius on each of these two host plants using a population projection. If C. formicarius can survive on I. triloba, growers have to focus in controlling pest incidence on the alternative host as well. These data will be useful in the selection or eradication of host plants for an integrated control strategy for C. formicarius.

Materials and methods

The experiments were conducted from January 2013 to March 2014 in a shade house at the Western Pacific Quarantine Biocontrol Laboratory of the University of Guam (13.43 °N, 144.80 °E and 54.3 m).

Study area

The shade house was 100 m² in area; the walls and roof were constructed of shade cloth. Mean temperature in the shade house was 31 °C (range 29.5–32.5 °C), mean relative humidity was 78% (range 76–80%) and the natural photoperiod was 14–16:10-8 (L:D) h.

Plants

Sweet potato (I. batatas) and Aiea morning-glory (I. triloba) cuttings (ca. 30 cm) were planted in individual pots (20 cm diameter × 30 cm deep) filled with disease- and weed-free Miracle Grow potting soil. One hundred and fifty such pots were planted with each of the two plant species for use in the experiments. Fertilizer in the form of Nitrogen, Phosphorus, Potassium and Sulfur was applied at the actual time of planting according to published recommendations¹².

Insects

Pheromone lures consisting of rubber septa loaded with Z3-dodecenyl-E2-butenoate, sealed in an impermeable bag for shipping and storage, were obtained from Chem Tica Internacional S.A. (San José, Costa Rica)¹³. Pherocon unitraps (Trécé Incorporated, Adair, Oklahoma, USA) baited with these lures were used to trap adult C. formicarius in sweet potato fields in Latte Heights (Guam, USA) during 2010¹⁴. The trapped adults were taken to the laboratory, placed in batches in collapsible cages (12 × 10 × 10 cm), fed leaves and pieces of the sweet potato and maintained at 22 ± 2 °C, 70–80% relative humidity and a 16:8 h L:D photoperiod¹⁵. In this colony, 5-6 generations were completed before the offspring were used for these experiments. For all experiments, 3-4 week-old adults were obtained from these laboratory colonies¹⁶.

Experiments

Each pupae were taken out of the insect colony and placed in an individual test tube. Upon emergence, fifty pairs of adults for each plant species consisting of virgin males and females were taken; each female and male was placed on the both potted plants. Enlarged stalks and woody plants of (comparable sizes of I. batatas and I. triloba) were chosen whenever possible, as females prefer to deposit eggs on these plants. The plants were covered with collapsible rearing cages (30 × 30 × 60 cm) to prevent escape. Eggs were deposited singly, in cavities or cells formed by the adult in roots and vines of the host plant⁵. The egg is creamy white, broadly oval and narrowed at the attached end. The egg becomes darker just before hatching and the dark head of the larva becomes noticeable¹⁷. Data were taken daily on the number of eggs laid, the number hatched and the duration of each developmental stage of C. formicarius on the two host plants. For the life table studies a total of 150 eggs, all laid within 24 h, were collected from the 50 pairs for both I. batatas and I. triloba. The eggs were checked daily until they hatched and then checked the larvae daily for survival. Since immature stages of the C. formicarius grow inside host plant, the stems have been taken out of the cages observed the immatures’ growth and survival. Daily counts of surviving larvae at each stage as well as number of eggs laid by each female were recorded.

Statistical analysis

Raw data of daily development and reproduction of each individual were analyzed according to the age-stage, two-sex life table theory¹⁸ as described by Chi². The population parameters calculated were the age-stage survival rate (s_xj, the probability that a newly laid egg will survive to age x and stage j), the age-specific fecundity of female (f_xj, the mean fecundity of females at age x), the age-specific survival rate (l_x, the probability of a newly laid egg survives to age x) and the age-specific fecundity (m_x, the mean fecundity of individuals at age x). In the age-stage, two-sex life table, the l_x and m_x are calculated as

where m is the number of stages.

The net reproductive rate (R₀) is calculated as

The intrinsic rate of increase (r) is estimated by using iterative bisection method from

with age x indexed from 0¹⁹. The finite rate of increase (λ) is calculated as λ = e^r. The mean generation time is defined as the length of time that a population needs to increase to R₀-fold of its size (i.e., e^rT = R₀ or λ^T = R₀) at the stable age-stage distribution and is calculated as

The life expectancy (e_xj) of individuals at age x and stage j is calculated according to Chi and Su⁴, while the reproductive value (v_xj) is calculated according to Tuan et al²⁰. The means and standard errors of the life table parameters were estimated by using the bootstrap procedure²¹ with bootstrap number m = 40,000 to ensure more precise estimates. TWOSEX-MSChart Visual BASIC (version 6, service pack 6) for Windows (available at http://140.120.197.173/Ecology/ (Chung Hsing University) and http://nhsbig.inhs.uiuc.edu.tw/www/chi.html (Illinois Natural History Survey) were used to analyse our age-stage, two-sex life table data (Chi, H. TWOSEX-MSChart: a computer program for the age-stage, two-sex life table analysis. (http://140.20.197.173/Ecology/Download/Twosex-MSChart.rar, 2013)

The paired bootstrap test^21,22 was used to compare the differences in developmental time, adult longevity, adult preoviposition period (APOP), total preoviposition period (TPOP), oviposition days and fecundity between treatments. The population parameters (r, λ, R₀ and T) between two treatments were also compared by using the paired bootstrap test based on the confidence interval of differences^21,22.

Population projection

Survival rate and fecundity data were used to project population growth according to Chi and Liu¹⁸ and Chi²³. The computer program TIMING-MSChart (Chi, H. TWOSEX-MSChart: a computer program for age-stage, two-sex life table analysis, http://140.120.197.173/Ecology/, 2014) used in this projection is also available at the above-mentioned web sites.

Results

Out of 150 eggs (N) used at the beginning of life table study on I. batatas, 149 eggs hatched and 146 larvae successfully developed to adult stage (Table 1). We found that larvae and pupae of C. formicarius developed faster on I. batatas than on I. triloba. The developmental times of the larval and pupal stages were 37.0 and 8.3 days, respectively, when reared on I. batatas, which were significantly shorter than those observed on I. triloba (54.9 and 16.9 days, respectively) (Table 1). Adult females emerged earlier and began to produce eggs at 42 days earlier when reared on I. batatas. There was no difference between the adult preoviposition periods (APOP) reared on I. batatas and I. triloba (P = 0.5234). The mean of total preoviposition period (TPOP) on I. batatas was 50.1 d. When reared on I. triloba, the first female adult emerged much later (day 60) and the mean TPOP was 77.2 d, significantly later than on I. batatas. The mean fecundity (F) of females was 90.9 eggs on I. batatas, significantly higher than the mean fecundity of 68.57 eggs on I. triloba (Table 1). Out of 150 eggs (N) used at the beginning of life table study on I. batatas, 90 individuals emerged as female adults (N_f); meanwhile, the values of F, N_f and N on I. triloba were 68.57, 83 and 150, respectively.

Table 1 Means and standard errors of developmental time (d), adult longevity (d) and fecundity (eggs), adult preoviposition period (APOP) (d), total preoviposition period (TPOP) (d) of Cylas formicarius (F.) reared on Ipomoea batatas and Ipomoea triloba.

The faster development on I. batatas could also be observed in the age-stage survival rate (s_xj). The curves of females and males emerged at 42 and 39 d, respectively, for I. batatas (Fig. 1A); in contrast, female and male adults appeared at 60 and 64 d, respectively, on I. triloba (Fig. 1B). The female age-specific fecundity (f_xj) and age-specific fecundity (m_x) on I. batatas not only began much earlier (at 42 d) than that on I. triloba (at 60 d) (Fig. 2), but were also significantly higher than for the latter.

Figure 1

Survival rate to different developmental stages of Cylas formicarius on sweet potato Ipomoea batatas (A) and morning glory Ipomoea triloba (B).

Figure 2

Age-specific survival rate (l_x) versus age of Cylas formicarius on sweet potato Ipomoea batatas (A) and morning glory Ipomoea triloba (B).

The intrinsic rate of increase (r), the finite rate (λ) and the net reproductive rate (R₀) of the sweetpotato weevil reared on I. batatas were 0.0622 d⁻¹, 1.0663 d⁻¹ and 54.55 eggs, respectively, all significantly higher (P = 0.0001) than the values obtained on I. triloba (0.0418 d⁻¹, 1.0426 d⁻¹ and 37.94 eggs, respectively) (Table 2). On the other hand, the mean generation time (T) obtained on I. batatas (62.26 d) was shorter than that on I. triloba (87.07 d).

Table 2 Means and standard errors of the intrinsic rate of increase (r), finite rate (λ), net reproductive rate (R₀) and mean generation time (T) of Cylas formicarius (F.) reared on Ipomoea batatas and Ipomoea triloba estimated by using bootstrap technique with 40,000 resamplings.

The longevity of the C. formcarius at age zero (e₀₁) was 135.63 d on I. batatas, which did not differ significantly from the longevity of 134.68 d on I. triloba (Fig. 3). These were exactly the mean longevity of all individuals used in the life table study. At age zero, the reproductive values (v₀₁) were exactly the same as the finite rates on both host plants, i.e., 1.0663 d⁻¹ on I. batatas and 1.0426 d⁻¹ on I. triloba (Fig. 4). The value of v_xj on I. batatas jumped to 45.5 d⁻¹ at 42 d when female adults emerged; when reared on I. triloba, the v_xj value jumped to 54.93 d⁻¹ when female emerged later at 60 d.

Figure 3

Life expectancy of Cylas formicarius on sweet potato Ipomoea batatas (A) and morning glory Ipomoea triloba (B).

Figure 4

Reproductive value of Cylas formicarius on sweet potato Ipomoea batatas (A) and morning glory Ipomoea triloba (B).

The population projection suggested that C. formicarius will grow much faster on I. batatas than on I. triloba (Fig. 5). Beginning with 150 eggs, the population would undergo four generations and the total population would exceed five million on I. batatas after 180 d, while the weevils would go through only three generation on I. troloba, for a final size of approximately 124,000.

Figure 5

Population projection of Cylas formicarius on sweet potato Ipomoea batatas (A) and morning glory Ipomoea triloba (B). Population projection started with 150 eggs of C. formicarius.

Discussion

The main host of C. formicarius and all species of sweetpotato weevil (Euscepes postfasciatus (Fairmaire), Daealus tuberosus (Zimmerman) and Cylas puncticollis (Boheman) is the sweet potato^24,10,25. Another species, Cylas brunneus (Fabricius) has found only on sweet potatoes. Carrot (Daucus carota (Hoffm.) Schübl. & G. Martens), radish (Raphanus sativus L.) and morning glory (I. triloba) are known to serve as additional hosts for C. formicarius²⁴. Since I. triloba is widespread in the Pacific Islands and other regions of the world¹⁵, it is one of the main alternative hosts for C. formicarius.

Because the age-stage, two-sex life table takes the variable developmental rate among individuals into consideration, the overlapping between stages can be observed in the curves of s_xj. However, if the same data were analyzed using the traditional female age-specific life table, such as Lewis-Leslie matrix^26,27,28,29, the stage differentiation would not be revealed. The age-specific survival rate (l_x) (Fig. 2) is the simplified version of the age-stage survival rate (Fig. 1). Although the stage differentiation and overlapping could not be observed in the l_x curve, it is nevertheless constructed by using the age-stage, two-sex life table and could correctly describe the change of the survival rate with age. Huang and Chi³⁰ demonstrated that an erroneous l_x curve would be obtained if the traditional female age-specific were used.

Chi² proved mathematically the relationship among F, N_f, N and R₀ as R₀ = F × (N_f /N). In this study, the values of F, N_f, N and R₀ on both host plants are completely consistent with the proven relationship. This relationship can be used to detect errors in life table analysis.

Although the weevil could survive longer than 160 days on both host plants (Fig. 1), the reproductive values (Fig. 4) showed that female adults did not contribute to the population growth after 92 d when reared on I. batatas, while C. formicarius on I. triloba did not have any contribution to population growth after 116 d. It seems that females from I. triloba can migrate to nearby I. batatas field after 92 d and still produce offspring.

Many plants function as oviposition sites for most herbivorous insects which deposit their eggs on all parts of the plants³¹. Leaf boundary layer effects both insect egg deposition behavior and progress of the embryo inside the egg^31,32. The effects of eggs on plants consist of egg-induced changes of photosynthetic activity and of the plant’s secondary metabolism³¹. In this study, out of 150 eggs, four eggs did not hatched. This could be possibly due to ovicidal effect of host plants of I. batatas or I. triloba^32,33.

A population projection based on an age-stage, two-sex life table can reveal the change of stage structure during population growth. Understanding stage structure is important to pest management because the dispersal and damage capability of insects varies with stage. This study demonstrates that such a life table can provide a comprehensive description of the fitness of an insect population on a given host plant. This study provides an interesting information on life table comparison of a C. formicarius on two of its host species. This shows that implications of the life-table data in IPM and/or weevil biology. The measure of the degree of dispersal from morning glory to sweet potato is clear, therefore the morning glory management around sweet potato fields is warranted. However, further studies are required on some insects occurring in alternate host plants may have to be proven to belong to genetically distinct populations or even cryptic species that do not cause much damage to crops of interest. Additional studies are also required if I. triloba is also the alternative host of another pest of sweet potato (i.e., Euscepes postfasciatus) an IPM strategy to manage I. triloba needs to consider E. postfasciatus in the picture.

This study determined that the C. formicarius can survive, on average, longer than four months on both host plants and successfully produce offspring for a month. Because I. triloba can be used successfully by C. formicarius as an alternate host, we suggest that growers should include the clearing of I. triloba from around sweet potato fields and storage facilities in their management program.