Background

The fall armyworm Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noctuidae) is a worldwide agricultural pest native to the tropics and subtropics of the Western Hemisphere [1]. It is a highly polyphagous moth that reportedly attacks over 350 hosts across 76 plant families, including Poaceae (106 species), Asteraceae (31 species) and Fabaceae (31 species) [2]. S. frugiperda can migrate over long distances. In 2016, invasion of Africa by S. frugiperda was reported; this was the first time this species was found outside its origin [3], and in May 2018 it was found in Asia [4]. Since then, S. frugiperda has spread to 47 African nations and 18 Asian countries, and now to Australia, where it poses a serious threat to crops [5]. In addition, feeding by S. frugiperda larvae can introduce saprophytic and pathogenic fungi, leading to infestation of crops and grains and resulting in significant preharvest losses and loss of grain quality [6,7,8].

S. frugiperda adults and larvae were first detected in China on December 11, 2018, and January 11, 2019 [9, 10], and it was confirmed that they belonged to a corn strain through gene fragment sequencing analysis [11]. The corn strain S. frugiperda prefers to attack maize (Zea mays) during its process of spreading [12]. According to a survey conducted in 2019, maize accounts for 98.6% of the total area of crops damaged by S. frugiperda in China [13]. Furthermore, feeding on maize appears to be more beneficial for S. frugiperda development and population growth than feeding on other crops such as rice (Oryza sativa), potato (Saccharum officinarum), wheat (Triticum aestivum) and soybean (Glycine max) [14,15,16,17]. On June 26, 2019, S. frugiperda larvae were observed feeding on bananas in Wuming District (23° 9′ 42.69″ N, 108° 16′ 44.13″ E), Nanning City, Guangxi, P. R. China (see Additional file 1, 2, 3). As a strategic crop for food security, bananas are highly valued for their nutritional quality [18,19,20] and are grown in more than 100 countries and regions around the world [21]. Bananas are also the most traded and consumed fruit internationally and are a major source of economic growth and income in many rural areas, creating jobs and providing foreign exchange value for many countries [22].

Factors that contribute to fluctuation of pest populations in the field include population density, reproductive rate, climatic conditions, and the abundance of natural enemies; the quality, availability, distribution and preference of the pest for alternate hosts also play an important role [23,24,25,26,27]. Food limitations play a key role in controlling insect populations since herbivore life history traits are influenced by host-plant characteristics [28]. For instance, an insect’s body size can be affected by the quality of its host plant, thereby determining life history parameters such as survival, longevity and fecundity [29, 30]. The presence of immature individuals of a species on any crop does not necessarily imply that the plant was a host for the insect [31].

Life tables are widely used as research tools in insect population ecology and pest management [32]. Compared with the traditional life table, the age-stage and two-sex life table not only takes into account the variability in developmental duration among individuals of both sexes, but also integrates the changes in the developmental speed changes of all stages in the form of stage distribution; thus, it can accurately describe the instar differentiation of insects and the generational overlap of populations [33,34,35,36]. Elucidating how well pests are adapted to different hosts can provide insights into pest dynamics in the field and thereby facilitate the timely adoption of prevention and control strategies [37]. It is essential to explore whether S. frugiperda feeding on bananas can develop to maturity and acquire the ability to produce fertile offspring. Therefore, in this study the survival and development of S. frugiperda on banana and maize were compared using age-stage and two-sex life tables with the goal of clarifying the adaptability of the species to banana. The study was designed to provide information on potential threats and risks to banana production, analyse the population source, and monitor and forecast S. frugiperda infestation of banana.

Materials and methods

Host plants

The test plant cultivars used in this study were banana variety (Williams B6; Guangxi Rural Export-oriented Economic Development Co., Ltd., Guangxi, China) and maize hybrid (Mei Yu Jia Tian Nuo No. 3; Hainan Lvchuan Seedling Co., Ltd., Hainan, China). Banana and maize were grown under field conditions without the use of pesticides. Maize seedlings at the three-leaf stage and newly developed banana leaves were used in the experiments.

Insect culture

S. frugiperda larvae at the 4th–6th instar were collected from maize fields in Shuangdou Village (22° 52′ 47.27″N, 109° 14′ 14.85″ E) in Jiaoyi Township, Hengzhou City, Guangxi, P. R. China on June 6, 2020. The larvae were reared in 11.5-cm diameter plastic petri dishes and fed an artificial diet [38]. The dishes containing the larvae were kept in an artificial climate chamber at 25 ± 2 °C, 75 ± 5% RH and 14 h L:10 h D photoperiod until pupation occurred. The newly emerged moths were paired (one female and one male) and introduced in plastic cups (11.5 cm in diameter and 15.5 cm in height) and fed with a 10% honey solution supplied through a small cotton wick. Eggs were collected daily and deposited in plastic petri dishes (9 cm diameter) until the emergence of the neonate larvae. Newly hatched larvae of the fifth generation raised on artificial diets were used in the following experiments.

Life history traits study

Three hundred newly hatched larvae were randomly selected and reared individually in plastic petri dishes (11.5 cm in diameter) on banana leaves or maize leaves (control) in the artificial climate chamber described above until pupation. Survival was checked daily, and larval instars were determined by checking for moulted exoskeletons. The larvae on the first day of each instar and the pupae on the second day were weighed with an electronic balance (JJ224BF; Changshu Shuangjie Testing Instrument Factory, Jiangsu, China).

The methods used to culture adults and eggs were the same as stated above. The longevity and reproduction, including the number of progeny eggs and their hatching, of each S. frugiperda adult were recorded.

Construction and analysis of the age-stage, two‑sex life table

Life tables of S. frugiperda were constructed and analysed based on the age-stage, two-sex life table theory and method [33, 34] using the TWOSEX-MSChart program [39].

The age-stage survival rate (Sxj), is the probability that a newly hatched individual will survive to age x and stage j, and age-stage specific fecundity (fxj) is the number of fertile eggs produced by the female adult at age x. These parameters accurately represent the biological characteristics of S. frugiperda [40]. The age-specific survival rate (lx) was calculated as

$${\text{l}}_{{\text{x}}} { = }\mathop \sum \limits_{{\text{j = 1}}}^{{\text{m}}} {\text{S}}_{{{\text{xj}}}} ,$$

where m is the number of stages. If all individuals of age x are included, this value expresses the age-specific fecundity (mx) of the total population:

$${\text{m}}_{{\text{x}}} { = }\frac{{\mathop \sum \nolimits_{{\text{j = 1}}}^{{\text{m}}} {\text{S}}_{{{\text{xj}}}} {\text{f}}_{{{\text{xj}}}} }}{{\mathop \sum \nolimits_{{\text{j = 1}}}^{{\text{m}}} {\text{S}}_{{{\text{xj}}}} }}.$$

The net reproductive rate (Ro) is defined as the total number of progeny that a female produces during her lifetime and is calculated as

$${\text{R}}_{{\text{o}}} { = }\mathop \sum \limits_{{\text{x = 0}}}^{\infty } {\text{l}}_{{\text{x}}} {\text{m}}_{{\text{x}}} .$$

The intrinsic rate of increase (r) is an important indicator of population characteristics. When the population is in an unrestricted environment and the age structure of the population is stable, r is the instantaneous growth rate of the population. r was calculated using the iterative bisection method with age indexed from zero, as in [35, 41]:

$$\mathop \sum \limits_{{\text{x = 0}}}^{\infty } {\text{e}}^{{ - {\text{r}}\left( {\text{x + 1}} \right)}} {\text{l}}_{{\text{x}}} {\text{m}}_{{\text{x}}} { = 1}{\text{.}}$$

The finite rate of increase (λ) is the theoretical value that represents the growth of the population per unit time; it is measured as er [42]:

$${\uplambda }\; = \;{\text{e}}^{{\text{r}}} .$$

The mean generation time (T) is defined as the amount of time a population requires to increase its size R0-fold as time approaches infinity and the population achieves a stable age-stage distribution. The mean generation time is calculated as:

$${\text{T = }}\frac{{{\text{lnR}}_{{\text{o}}} }}{{\text{r}}}.$$

Statistical analysis

The Mann–Whitney U test (U test) was used to identify differences between groups of S. frugiperda with respect to duration of development, body weight, female and male longevity, preoviposition period, oviposition days and eggs per female. Differences in adult sex ratios were compared using a nonparametric test (binomial test). The life table parameters were calculated using TWOSEX-MSChart software and the results were plotted using GraphPad Prism 8.0.1 (GraphPad Software Inc., San Diego, CA, USA). A probability level of P < 0.05 was accepted as statistically significant. All statistical analyses were performed using SPSS 26.0 (IBM Corp., Chicago, IL, USA).

Results

Developmental duration and body weight of S. frugiperda

S. frugiperda completed its life cycle by feeding on bananas (Table 1). Maize-fed larvae had six instars, while banana-fed larvae had eight instars. The developmental durations of the 7th and 8th instar larvae fed on bananas were 11.27 ± 0.14 days and 14.71 ± 0.21 days, respectively. The developmental duration of each larval instar, prepupal and pupal stage of S. frugiperda fed on bananas was significantly longer than that of the control (1st, Z = − 20.832, P < 0.001; 2nd, Z = − 19.652, P < 0.001; 3rd, Z = − 19.133, P < 0.001; 4th, Z = − 18.542, P < 0.001; 5th, Z = − 18.288, P < 0.001; 6th, Z = −  17.255, P < 0.001; prepupa, Z = − 9.991, P < 0.001; pupa, Z = − 2.663, P = 0.008). The larvae fed on bananas had a developmental duration of 64.94 ± 0.35 days, 3.43 times longer than that of the control larvae (18.93 ± 0.07 d). There was no significant difference in the hatching periods of the two groups of progeny eggs (Z = −  1.292, P = 0.197).

Table 1 Effects of banana and maize on developmental duration, body weight of S. frugiperda
Full size table

The difference in the body weights of 1st instar larvae fed maize and banana was not statistically significant (Z = −  0.551, P = 0.582) (Table 1). However, the body weights of 2nd to 6th instar banana-fed larvae were significantly lower than those of maize-fed larvae (2nd, Z =  − 10.694, P < 0.001; 3rd, Z =  − 11.526, P < 0.001; 4th, Z = − 11.008, P < 0.001; 5th, Z =  − 10.374, P < 0.001; 6th, Z = − 10.490, P < 0.001). Interestingly, the 8th instar larvae fed bananas weighed less (453.56 ± 7.03 mg) than 6th instar larvae fed maize (496.08 ± 10.23 mg). The average pupal weight of the individuals reared on maize (242.01 ± 23.87 mg) was significantly greater than that of the individuals reared on banana (132.02 ± 2.72 mg); it was 1.83 times that of the banana group (Z = −  11.012, P < 0.001).

Reproduction of S. frugiperda

There was no significant difference in the sex ratios of the two populations reared on banana and maize. The female longevity, male longevity and preoviposition period of S. frugiperda in the banana-fed populations were significantly longer than those in the controls, while oviposition days and the number of eggs laid per female in banana-fed populations were significantly lower than those in controls (Table 2).

Table 2 Effects of banana and maize on reproduction of S. frugiperda
Full size table

Life table

The age-stage survival rate (Sxj) is the probability that a newborn larva will survive to age x while in stage j (Fig. 1). Significant overlaps between stages were observed under both crops. The banana-fed populations of S. frugiperda completed the larval stage on Day 74, the pupal stage on Day 83, and eclosion on Day 68, while the corresponding times for the maize-fed populations were Day 23, 36 and 27, respectively. The Sxj that a newly hatched larva fed on maize would survive to the pupal stage was 0.90, considerably higher than that for larvae fed on banana (0.26). The Sxj values of S. frugiperda females and males from first instar larva to adult were 0.10 and 0.09, respectively, for larvae fed on banana and 0.39 and 0.40, respectively, for larvae fed on maize.

Fig. 1
figure 1

Effects of banana and maize on age-stage survival rate (Sxj) of S. frugiperda. L1, L2, L3, L4, L5, L6, L7 and L8 represent 1st, 2nd, 3rd, 4th, 5th, 6th, 7th and 8th instar larvae, respectively

Full size image

The age-specific survival rate lx is the probability that a newly hatched larva will survive to age x; because this parameter includes all individuals of the cohort and ignores stage differentiation, the lx curve is a simplified version of the Sxj curve (Fig. 2). Higher peaks of age-stage specific fecundity (fx), age-specific fecundity (mx), and lxmx were observed in S. frugiperda reared on maize than in S. frugiperda reared on banana. The maize-reared populations of S. frugiperda oviposited from Day 28 to the end of Day 39, and the banana-reared populations of S. frugiperda oviposited from Day 68 to the end of Day 88. Most of the females of in the maize-reared populations laid eggs on Days 31–33, while those in the banana-reared populations had multiple irregular oviposition peaks during the breeding period. The highest fx peak of females reared on banana occurred on Day 72, and the mean fecundity was 156.50 eggs, while the highest fx peak of females reared on maize occurred on Day 32, and the mean fecundity was 158.02 eggs.

Fig. 2
figure 2

Effects of banana and maize on age-specific survival rate (lx), age-stage specific fecundity (fx), age-specific fecundity (mx) and lxmx of S. frugiperda

Full size image

Life table parameters

The net reproductive rate (Ro) of S. frugiperda reared on maize was 253.98 progeny per female, much higher than that of S. frugiperda reared on banana (39.22 progeny per female) (Table 3). The intrinsic rate of increase (r) and the finite rate of increase (λ) for S. frugiperda reared on banana were 0.05 d−1 and 1.05 d−1, respectively, lower than those for S. frugiperda reared on maize (r = 0.17 d−1, λ = 1.18 d−1). The r and λ values for both groups were greater than 0 and greater than 1, respectively, indicating that S. frugiperda can complete generational proliferation whether feeding on banana or maize. The λ values of the banana-fed populations and maize-fed populations of S. frugiperda were 1.05 d−1 and 1.18 d−1, respectively, indicating that the two populations grew continuously and geometrically at rates of 1.05-fold and 1.18-fold per day, respectively, under these conditions. On the other hand, the mean generation time (T) of the banana-fed populations (T = 78.48 d) of S. frugiperda was 2.38 times longer than that of the maize-fed populations (T = 33.03 d).

Table 3 Effects of banana and maize on life table parameters of S. frugiperda
Full size table

Discussion

Herbivorous insects can generally complete their entire life cycles (egg to adult) on a host plant that can be considered an alternative host. S. frugiperda has been reported to damage a variety of plants [2, 38, 43]. Some plant species may support the complete development of S. frugiperda. For example, this pest can complete its life cycle on maize, sugarcane (Saccharum officinarum), rice (Oryza sativa), potato, cotton (Gossypium spp.), and amaranth (Amaranthus viridis) [44,45,46,47,48,49,50]. However, other plant species may not support complete development of S. frugiperda but may still be used by larvae or adults for feeding and laying eggs. For example, although damage to cabbage (Brassica oleracea), maranta (Maranta arundinacea), and coix (Coix lacryma-jobi) has been reported, no evidence that S. frugiperda can complete its life cycle on these plants [51,52,53]. The results of current study showed that S. frugiperda can complete its life cycle on banana plants, suggesting that banana is an alternative host plant for this insect pest.

Since S. frugiperda larvae were first observed to damage bananas in this study, and the larvae used in the study were either hatched on bananas or transferred from weeds of the family Gramineae, such as Eleusine indica, Setaria viridis and Digitaria sanguinalis [54, 55]. In Guangxi, spring maize is planted in early February, and it enters the late growth or harvest period in June. At this time, autumn maize and fresh maize were not yet been sown [56]. It may be that the absence of an ideal host causes female moths to lay their eggs on more numerous and occasional hosts, such as the perennial herb banana, rather than on their preferred host, but their larvae are underfed on the occasional hosts. Nevertheless, the presence of even a few surviving larvae can ensure the presence of some individuals on the occasional hosts until the population increases in the next growing season of the preferred crop. In addition, the abundance of natural enemies on some host plants may also cause females to lay eggs on less nutritious hosts to better protect their offspring [57, 58]. However, whether deposition of eggs on bananas by female S. frugiperda protects their offspring from predation or parasitization by natural enemies needs further study.

Undoubtedly, differences in the type of food consumed have a great impact on the growth and development of herbivorous insect larvae and on the reproduction of adults even under the same environmental conditions, and this in turn affects the change trend of the entire insect population [30, 59]. Although our results show that banana is an alternative host plant for S. frugiperda, we found that S. frugiperda feeding on bananas have a longer ontogeny cycle and lower survival and fecundity than maize-fed S. frugiperda. This finding indicates that although banana plants supply S. frugiperda with the nutrients required to complete its entire life cycle, growth on banana is not conducive to optimal development of its population.

A number of studies have reported that S. frugiperda has six larval instars [15, 17, 37, 48, 50]. However, the present study identified up to eight larval instars, and He et al. [17, 60] also reported a similar result. To date, no other study of S. frugiperda has shown this phenomenon, but there are examples showing a similar occurrence of eight instars in other species, including Malacosoma disstria [61], S. exigua [62] and Chilo suppressalis [63].

Notably, our findings indicated that although the sex ratios of in the banana- and maize-reared populations did not differ significantly, the females and males in the banana-reared population lived significantly longer than those in the maize-reared population and that the females in the maize-reared populations had shorter preoviposition periods, more oviposition days, and higher fecundity. Previous studies have shown that the opportunity to reproduce closely related to longevity; therefore, decreased longevity in response to current reproductive efforts was used to estimate the costs of reproduction [64]. For example, the parasitic wasp Itoplectis naranyae has a shortened lifespan after parasitizing its hosts, suggesting that parasitization has reproductive costs in terms of egg production [65]. It is also possible that nutritional restriction is responsible for this difference. For instance, in the case of complete feeding, Grandison et al. [66] found that the addition of amino acids increased fecundity and shortened longevity in flies.

Conclusions

In conclusion, bananas are alternative but not ideal hosts of S. frugiperda compared to maize. Even so, in situations in which the population density is too high or the preferred host is scarce, it is still critical to prevent S. frugiperda from transferring to bananas and thereby increasing the number of sources of outbreak. On the other hand, larval instars of S. frugiperda reared on banana had longer developmental times than those reared on maize. These findings may be applied to the design of a comprehensive integrated pest management strategy and may help explain the rapid expansion of this polyphagous species across different areas in China. Our results show that banana can serve as an alternative host for S. frugiperda during the maize harvest or during offseason planting.