Introduction

The economically important lepidopteran insect pests of sugarcane, maize and sorghum in Africa are Busseola fusca (Fuller) (Noctuidae) (Calatayud et al. 2014; Kfir et al. 2002), Chilo partellus (Swinhoe) (Crambidae) (Kfir et al. 2002), Eldana saccharina (Walker) (Pyralidae) (Assefa et al. 2008; Kfir et al. 2002), Helicoverpa armigera (Hübner) (Noctuidae) (Moran 1983), Sesamia calamistis (Hampson) (Noctuidae) (Kfir et al. 2002) and Spodoptera frugiperda (J.E. Smith) (Noctuidae) (Goergen et al. 2016). These pests often occur in mixed populations in cereal crops and can cause yield losses ranging from 10% to total crop loss (Ntiri et al. 2019; Sokame et al. 2020; Van den Berg et al. 1991). Research on these pests necessitates rearing on artificial diets to ensure reliable supplies of high quality insects. High quality insects are also needed for rearing of biocontrol agents. The monitoring of resistance evolution, as well as efficacy evaluation of specific plant-incorporated traits such as insecticidal Cry proteins produced by genetically modified crops, for example Bt maize and Bt sugarcane, also relies on availability of mass reared insects (Nair et al. 2019). Being able to successfully rear insects is also a fundamental part of developing efficient programs for integrated pest management (IPM) (Lima Pinto et al. 2019).

The lepidopteran species mentioned above are pests of crops that are increasingly being considered targets for genetic modification (Van den Berg et al. 2021) as well as development of alternative control methods (Maphumulo et al. 2023).

The mass rearing of insects has been and continues to be a crucial part of entomological research. More than 60% of papers written about studies involving insects use artificially reared insects rather than field-collected populations and approximately 45% of those insects were reared on artificial diets rather than on their natural diet (Cohen 2001). When insects are reared on artificial diets it implies that a synthetic diet is provided instead of what it would normally consume (Cohen 2001), without having an effect on its growth and development (Alfazairy et al. 2012; Nair et al. 2019). Laboratory colonies of insects reared on artificial diets reduces the labour, time, space, and in many cases, the high costs associated with rearing insects on their natural host plants (Gillespie 1993; Hervet et al. 2016). It is not always feasible to rear insects on their natural host plants because of seasonal availability, high costs and varying quality of host plants over time (Alfazairy et al. 2012; Sørensen et al. 2012; Vanderzant 1969). In many cases however, the host-plant factor can be incorporated into the artificial diet (Singh and Sarup 1987).

There are many challenges and precautions to consider when insect mass rearing is done, especially to ensure insects of high quality. High quality mass reared insects are characterised by rapid growth and maturation rates and high levels of survival to adults (Jensen et al. 2017). Life history parameters are the most widely used criteria for assessing general insect quality, with the most important characteristics being larval and pupal mass, developmental periods, and fecundity and fertility (Cohen 2001). High body mass in reproductive adults can be linked to higher reproduction capacity that ensures large numbers of offspring for the next generation (Honĕk 1993). A common method used to monitor the quality of laboratory-reared insects is to compare its performance to that of its wild relatives (Huettel 1976). Other performance traits (Chambers 1977) that are commonly used in evaluation of diet suitability are larval and moth dispersal, as well as courtship, mating and oviposition behaviour (Huettel 1976; Rahayu et al. 2018).

A challenge that mass rearing facilities with limited resources often face is, for example, access to various ingredients and equipment, which may necessitate adjustments to diet contents and protocols. As a research facility that mass rear numerous lepidopteran species, the need arose for diets with simple, easily accessible ingredients, that could accommodate the rearing of more than one species, without compromising the quality of the reared insects. Local availability of ingredients and plant material was the main motivation for the adjustments. Development of a single diet that enables mass rearing of several species is also of economic benefit in terms of time and labour input.

The aim of this study was to determine the suitability of different diets for rearing of the species mentioned above. The artificial diets used for rearing Noctuidae, Crambidae and Pyralidae species in this study have been derived and adapted from diets published by Songa et al. (2004), Onyango and Ochieng’-Odero (1994), Singh and Sarup (1987), Abbasi et al. (2007), Balaji and Manvendra (2013), Hamed and Nadeem (2008), Ochieng et al. (1985), Ratnadass et al. (2001), Saranya and Samiayyan (2017) and Tefera et al. (2010). If successful, rearing of the insect species on the adapted artificial diet will enable research facilities in South Africa and other developing countries to expand and support research efforts through using readily available ingredients.

Materials and methods

Insect stock colonies and general procedures

Four lepidopteran stem borer species (B. fusca, C. partellus, S. calamistis and E. saccharina) and two maize ear and leaf feeding species (H. armigera and S. frugiperda) were used in this study. One thousand diapause larvae of B. fusca were collected from maize stalks in harvested fields during winter months. Maize stubble were uprooted and dissected to collect larvae that overwintered inside the bases of stalks. Larvae were then placed in eight 25 ℓ containers with dry maize leaves and stored in temperature controlled rooms maintained at 10–12 °C. Diapause was terminated following the methods described by Van Rensburg and Van Rensburg (1993), by placing the containers in rearing chambers maintained at 25 ± 5 °C, 60% humidity and a photoperiod of L14: D10. Pupae that formed were collected and placed in containers until moths emerged. Eighty pairs of moths were placed individually in 2 ℓ plastic bottles to mate and lay eggs. Larvae that hatched from these eggs were used in diet bioassays. Genetic diverse offspring was therefore used in these bioassays by ensuring that offspring of as many as possible females were used in assays. The experimental design for all assays was a completely randomized. All assays were conducted in an insect rearing chamber at 28 ± 2 °C, 60 ± 10% relative humidity (RH), and a photoperiod of 14:10 (L: D). All assays were conducted in the laboratories of the Entomology division at the ARC-Grain Crops experiment station (ARC-GC), Potchefstroom, South Africa.

Approximately 800 larvae of each of C. partellus and S. frugiperda were sampled from maize plants during the growing season. Larvae were placed in 100 ml containers and reared on maize plant tissue until pupation. Pupae were placed in containers until moths emerged after which 63 pairs of C. partellus and 70 pairs of S. frugiperda were placed inside 2 ℓ plastic bottles to mate and lay eggs. Eggs of S. calamistis and E. saccharina were provided by the mass-rearing facility at the South African Sugarcane Research Institute (SASRI, Mt Edgecombe, South Africa). Wild H. armigera moths (approximately 45) were collected by means of light traps in maize fields at the ARC-Grain Crops experiment station (ARC-GC), Potchefstroom, South Africa. Moths caught in the traps were paired and put into oviposition cages. Eggs from 17 different females were used to ensure that genetically diverse neonates were used in the bioassays.

Rearing procedures

The suitability of three different artificial diets for rearing of each of the species were evaluated. Lepidoptera species have previously been reared on two different artificial diets at the ARC-GC that was based on previously published diets. Adaptations were made to these diets and are referred to in this study as the UEA (maize leaf; ML)-diet and UEA (sugar cane; SC)-diet. The acronym UEA (U– Ursula du Plessis, E– Elrine Strydom and A– Annemie Erasmus) represents the contributors that made improvements towards the rearing techniques and who manages the insect mass rearing facility at the ARC-GC laboratories. The main ingredient of the UEA (SC)-diet was crushed sugarcane stalks and this diet was used to rear C. partellus, E. saccharina, H. armigera and S. frugiperda. The UEA (ML)-diet is based on ground maize leaf powder and used for rearing B. fusca and S. calamistis. An artificial diet developed at the South African Sugarcane Research Institute (SASRI) to mass rear crambid and noctuid species (Gillespie, D. 2016, pers. Comm., 8/03/2020) (South African Sugarcane Research Institute - SASRI) was also included in this study. For the remainder of this article, the UEA (ML)-diet and the UEA (SC)-diet will be referred to as the ML and SC-diets respectively.

Diet preparation

Diets were prepared by using the ingredients listed in Table 1. Sorbic acid was added to the boiling distilled water (Fraction A) and stirred until it was completely dissolved. Agar powder was added to distilled water (Fraction B) and added to the boiling mixture. The mixture (Fractions A & B) was allowed to boil for 10 min and was stirred throughout the process. Distilled water and dry ingredients (Fraction C) were thoroughly mixed in a separate bowl. The mixture (Fractions A & B) was allowed to cool down to 60–70 °C after which fraction C was added. This mixture was then blended in a commercial food blender. Ingredients of fraction D was mixed and added to the mixture of fractions A, B and C. Lastly, formaldehyde 40% (Fraction E) was added to the mixture and blended for 3 min. Approximately 75 ml of diet were dispensed per 100 ml container and left overnight in a flow cabinet to set.

Table 1 Ingredients of the three artificial diets evaluated for their suitability for rearing of six Lepidoptera species
Full size table

The ingredients used in these diets were similar to those in published diets of other lepidopteran species. The main changes consisted of the plant material added to the diets such as the maize leaf and sugarcane stalk powder that are the main host plants of the majority of these insect pest species. The milk powder that was added (Nespray FortiGrow. Nestle, Bryanston, South Africa) to the adapted diets served as substitutes for the sucrose, vitamin E and cholesterol that were used in some of the previously published diets.

Fitness bioassays

Two diet assays were conducted with each species. In the one experiment, larval development and mass was assessed twice a week until prepupae formed. In the other assay, larvae were inoculated onto the diet and left undisturbed until pupae formed. Larval development and pupal formation were also assessed using maize leaf whorl tissue, to enable comparison between the natural host plant and artificial diets.

Larval development and growth

Neonate larvae were used in all bioassays. For each of the stem borer species, there were five replicates per diet. Each replicate consisted of five plastic test tubes, each inoculated with five neonate larvae. For each of the stem borer species, 175 containers were inoculated with five larvae per container. The diet dispensed into the containers was enough to supply enough food for the duration of the assay. The volume of the test tubes was 100 ml (52 mm high and 30 mm in diameter) and each tube had a mesh-infused lid for aeration purposes. No cannibalism was expected with the stem borer species since previous studies conducted by Ntiri et al. (2016) and Ratnadass et al. (2001) successfully reared larger numbers of larvae (10–12) in containers of comparative sizes.

To limit the effect of disturbance on larval growth and development, only 25 test tubes of each diet were opened per evaluation day to determine larval survival and mass.

In the assays with S. frugiperda and H. armigera, there were ten replicates, with ten petri dishes per replicate for each diet. Since larvae of these species are cannibalistic, only a single larva was inoculated per petri dish (90 mm diameter). Before placing the diet into petri dishes, it was poured into 10 ml cells in 32-well rearing trays and allowed to set for 12 h. The diet was removed from the 32-well rearing trays and two 10 ml diet blocks were placed inside each petri dish. The diet was replaced with fresh diet at weekly intervals. All petri dishes of each diet were opened twice a week to determine larval survival and mass. Since the life cycles of H. armigera and S. frugiperda are shorter than those of stem borers, larval survival and mass were determined after 11 and 14 days, respectively, when pre-pupae started to form. A growth index which provided an indication of overall suitability of a diet, was determined for each species on each of the diets as follows: growth index = percentage larval survival/mean larval period (Onyango and Ochieng’-Odero 1994).

Evaluation of the species on their natural diet were done by placing larvae onto freshly cut maize leaf whorl tissue, inside plastic containers. The leaf whorl tissue was cut from 4 to 6 week old plants. Fifty containers were inoculated with five larvae per container for the stem borer species. Bioassays with S. frugiperda and H. armigera, consisted of 100 containers, inoculated with one larva each.

Whorl tissue was provided twice a week and larval survival and mass recorded for the duration of the larval period. No insecticides were used to spray onto plants prior to their use.

Pupal development

Similar procedures were used for the evaluation of pupa development for each of the four stem borer species. For each of the stem borer species, 80 test tubes were inoculated with five neonate larvae each and left undisturbed until the first pupae were observed. Tubes were checked for the presence of pupae depending on the estimated larval development period of each species. Pupae were then removed and pupal mass and the number of days to pupation recorded. The percentage pupation was determined, and duration of pupal period calculated. The assays for S. frugiperda and H. armigera consisted of 150 petri dishes per diet. Each petri dish was inoculated with one neonate larvae. Larvae were left undisturbed until the first pupae were observed, after which petri dishes were checked daily to collect the pupae. Larval development and pupal formation on maize leaf tissue was also determined as described above to enable comparison between the natural host plant and artificial diets.

The mass of male and female pupae was recorded separately. The sex of each pupa was determined based on the external structures on the ventral surface of the last abdominal segment (Butt and Cantu 1962) and the sex ratio calculated for all species on artificial and natural diet.

Data analysis

For larval survival and mass, a one-way analysis of variance was conducted for each species separately, using the GLM (General Linear Models) Procedure of SAS software (Version 9.4; SAS Institute Inc, Cary, USA). The Shapiro-Wilk test was performed to test the residuals for deviation from normality (Shapiro and Wilk 1965). The homogeneity of the diet variances were tested using Bartlett’s test (Bartlett 1937). The larval mass was natural log transformed to normalise the residuals and stabilise diet variances. Tukey’s Studentised Range (HSD) Test were used at the 5% level to compare means for significant effects. A probability level of 5% was considered significant for all significance tests. Duration of the larval development period as well as the pupal periods and mass of males and females, were compared between artificial diets for each species by means of t-tests. Chi-square analysis was used to determine if there were differences in pupation as well as the sex ratios of pupae that developed on the different artificial diets. Data on mass were natural log transformed to normalise residuals and stabilise diet variances. Chi-square analyses were used to determine if there were differences in pupation and sex ratio on the different artificial diets. In this study the χ2 tests were done using XLSTAT to test whether pupation and sex ratio are independent of diet for the different species respectively. A significant χ2 test indicates evidence against independence. In other words, for a significant χ2 test, the proportion larvae that pupated or the sex ratio was dependent on the diet they were reared on.

Results

Larval survival, development and growth

Larval survival and mass of all six species are provided in Table 2. Larval survival of B. fusca on day 25 was the highest on the ML-diet (53%) when compared to the natural diet (45%) and very low survival was observed on the SASRI (6%) and SC-diets (0.8%) (Table 2). Larvae recovered from the ML-diet had a significantly higher mean mass compared to those on the other two diets and did not differ significantly from that on maize leaf tissue. The mean larval duration period on the ML-diet was 49.1 days, 7 days longer than on maize leaf tissue (Table 3). No larvae survived until pupation on the SC- and SASRI diets. Busseola fusca had a growth index of 1.1 on the ML-diet, which was similar to what was observed on maize leaf tissue.

Survival of C. partellus larvae on day 25 was significantly higher on the SC-diet (61.6%) compared to the ML- (57.6%), SASRI (13.64%) and natural (55.2%) diets (Table 2). A significantly higher mean larval mass was recorded on the SC-diet compared to those on the other diets. Chilo partellus had a significantly shorter larval duration on the SC-diet (23.9 days) when compared to the natural (25.8 days) and ML-diets (31.6 days). Therefore, the highest growth index was observed on the SC-diet (2.6) (Table 3).

Significantly higher numbers of E. saccharina larvae survived (66%) on the SASRI diet (Table 2). Mean larval mass was higher on the SC-diet (184.5 mg) than on the ML-diet (153.26 mg) and SASRI diet (152.83 mg) but the differences were not significant. Larval mass was the lowest on maize leaf tissue (83.9 mg). Larval duration was the shortest on the SC-diet (24.2 days) and almost eight to 12 days longer on the SASRI and natural diets. The highest larval growth index for E. saccharina (2.0) was observed on the SASRI diet (Table 3).

Mean larval survival of H. armigera was significantly higher on the ML- (99%) and SC-diets (96%) compared to the natural (78%) and SASRI diets (46%) (Table 2). Mean larval mass on the SC-diet was significantly higher than on any of the other diets. The larval duration period of H. armigera on the SC-diet was 13 days compared to 44 days on the SASRI diet. The growth index of H. armigera was the highest on the SC-diet (7.4) (Table 3).

Sesamia calamistis larvae showed significantly higher survival on the ML-diet (80%) compared to the SC- (48%), natural (42%) and SASRI diets (36%). The highest mean larval mass was observed on maize leaf tissue but did not differ significantly from the ML-diet. Larval duration was 2 days shorter on maize leaf tissue compared to the ML-diet. The highest growth index (2.1) was recorded on the ML-diet (Table 3).

Mean larval survival of S. frugiperda on the three artificial diets was high and ranged between 84 and 90% while 76% survival was recorded on maize leaf tissue. The highest mean larval mass was recorded on the natural diet although the difference was not significant compared to the ML-diet. Duration of the larval period of S. frugiperda was significantly shorter on the ML-diet, which resulted in the highest growth index (4.7) on this diet (Table 3).

Pupal development

Percentage pupation was compared between the natural and artificial diets for each of the insect pest species (Table 4). While no B. fusca larvae survived on the SASRI or SC-diets, 42.9% pupated on the ML-diet and 14% on the natural diet. Pupation of C. partellus (67.3%) was the highest on the SC-diet and E. saccharina showed very high pupation of 89.3% on the SASRI diet. The highest incidence of pupation observed for H. armigera was on the ML-diet (96.4%) while 83.3% of S. calamistis larvae pupated on the ML-diet. The percentage pupation of S. frugiperda was high (> 77%) on all diets, with 86% pupation on the natural diet (Table 4).

The duration of pupal periods is provided in Table 5. A significant difference was observed in male pupal duration between the larvae of B. fusca feeding on the ML-(11.9 days) and natural (13.7 days) diets. There were no significant differences observed between the duration of male and female pupal periods for either C. partellus or S. calamistis on the ML-and SC-diets. Significantly, shorter male pupal periods were recorded for E. saccharina on the SC-diet, while shorter female pupal periods were observed on the SC-diet and it did not differ significantly from that on the natural diet. Pupal duration of H. armigera was significantly shorter on the natural diet for both males and females. Duration of the S. frugiperda male pupal period was significantly shorter on the SC-diet, and a significant difference was observed in pupal duration of females when artificial diets was compared to the natural diet.

Results on male and female pupal mass are provided in Table 6. No larvae of B. fusca, C. partellus and S. calamistis completed their life cycles on the SASRI diet. Male pupae from B. fusca had a higher mass on the ML-diet compared to the natural diet, while no significant difference was observed between female pupae on the two diets. Male and female pupae of C. partellus were heavier on the SC-diet. The highest pupal mass for E. saccharina was recorded on the SASRI diet. Male and female pupae of H. armigera were heavier on the SC-diet while S. calamistis pupae were heavier when larvae were reared on the ML-diet. Pupae from S. frugiperda had the highest pupal mass on the SC-diet.

Sex ratios, which were determined during the pupal stage, are compared in Table 7. The sex ratios did not differ from the expected 1:1 ratio on the different artificial diets, except for E. saccharina, for which there were significantly more males on the SC-diet.

Table 2 Mean larval survival and mass of lepidopteran species reared on three different artificial diets
Full size table
Table 3 Mean values of different larval parameters of lepidopteran species reared on three different artificial diets
Full size table
Table 4 Incidence of pupation (%) of lepidopteran species reared on three artificial diets
Full size table
Table 5 Duration of the pupal period of male and female pupae of six lepidopteran species reared on three different artificial diets
Full size table
Table 6 Mean mass of male and female pupae of six lepidopteran species reared on three different artificial diets
Full size table
Table 7 The sex ratios of six lepidopteran pest species reared on three different artificial diets
Full size table

Discussion

The adapted diets used in this study are suitable to sustain the Lepidopteran species B. fusca, C. partellus, E. saccharina, H. armigera, S. calamistis and S. frugiperda and were successful in producing healthy individuals that had the same fitness as wild or similarly reared relatives. The life history parameters observed in this study was compared to those reported in similar studies and are provided, together with the specific rearing conditions (e.g., temperature and R.H.) where possible.

Busseola fusca is an oligophagous pest of which the host plant range consists of only a few grass species (Moolman et al. 2014; Van den Berg et al. 2001). Busseola fusca moths are highly selective in their host plant selection process, which is usually maize and sorghum (Kfir et al. 2002). This stem borer completed its life cycle only on the ML-diet that contained incorporated maize leaf powder. Larval duration of B. fusca on the ML-diet was similar to that reported by Onyango and Ochieng’-Odero (1994)(32.3–70 days) (25–30 °C; 50–80% R.H.) and Songa et al. (2004), (40–45 days) (28 °C; 60–70% R.H.). Survival rates of this species on the natural diet were similar to that reported by Kotey et al. (2017). This species is challenging to rear and life history parameters measured in this study are similar to those reported previously (Onyango and Ochieng’-Odero 1994) on the diet that is considered the most effective for rearing of B. fusca. Pupal duration on the ML-diet is comparable to that reported by Onyango and Ochieng’-Odero (1994) (pupal duration: 14.3–14.5 days; pupal mass: 160.0–292.1 mg). Pupal duration on the ML-diet was, therefore, about two days shorter compared to the duration period reported by Onyango and Ochieng’-Odero (1994). Pupal mass observed in the current study was in a range similar to that reported by Onyango and Ochieng’-Odero (1994) and Kruger et al. (2012). Kruger et al. (2012) reared B. fusca on maize leaf tissue and reported comparatively similar pupal mass of males and females separately. Larvae that were reared on the ML-diet had the same range compared to field-collected individuals (Kruger et al. 2012; Strydom et al. 2019) which indicates that they have the same level of fitness although survival was not high on artificial diet. It is generally acknowledged that larger female pupae provide for moths that are more fecund than smaller females (Calvo and Molina 2005). The 1:1 sex ratio of B. fusca observed in this study was similar to that published by Kruger et al. (2012). The results obtained for the ML-diet justifies the adaptation of the previously published diet (Onyango and Ochieng’-Odero 1994; Ratnadass et al. 2001; Songa et al. 2004; Tefera et al. 2010) by supplementing with milk powder to achieve a simplified ingredient list. According to Onyango and Ochieng’-Odero (1994) the cannibalistic tendency that the larvae exhibit at times toward other larvae and pupae in the same container makes mass rearing a challenge. Busseola fusca is diet sensitive, for example, the age of incorporated plant material (sorghum leaf powder) have also been reported to have a direct effect on larval survival (Onyango and Ochieng’-Odero 1994). It was previously reported by Songa et al. (2004) that B. fusca larvae can be reared in multiples per container without cannibalism if there is an adequate amount of diet available. Microbial contamination of the artificial diet is a problem with this species and has a direct impact on the successful rearing of B. fusca since contamination inhibits larval growth (Onyango and Ochieng’-Odero 1994).

The duration of the larval period recorded in this study for C. partellus was significantly shorter than the 43.5 days observed by Saranya and Samiayyan (2017) (28 °C; 60 ± 10% R.H.) on artificial diet. Nair et al. (2019) (25 ± 2 °C; 65 ± 5% R.H) recorded a larval duration period of 30.9 days. Singh and Sarup (Singh and Sarup 1987) evaluated 11 different artificial diets to improve mass rearing of C. partellus and reported the developmental period to range between 27.23 and 34.10 days (27 ± 1 °C; 80–90% R.H.). A larval period of 32 to 39 days (26 °C; 60 ± 10 R.H.) was observed by Peddakasim et al. (2018) when larvae fed on artificial diet. According to Cohen (2015) shorter developmental periods are one of the most used biological parameters to positively correlate with the suitability and quality of the diet. In a study done by Ochieng’-Odero et al. (1994) (26 ± 2 °C; 65 ± 5% R.H.) the pupal period of C. partellus was also similar for females and males. Peddakasim et al. (2018) recorded pupal periods for males between eight to nine days and females nine to ten days. Male and female pupal mass was similar to what was found in a study by Ochieng’-Odero et al. (1994). A study on suitability of six different poaceous host plants for development of C. partellus larvae (Rebe et al. 2004), reported similar ranges in pupal mass after larvae were reared on these hosts. Male and female pupal mass (59.95 mg; 170.00 mg) on maize, sorghum (62.92 mg; 98.93 mg), sweet sorghum (64.26 mg; 115.15 mg) and pearl millet (61.63 mg; 104.15 mg) was similar to those recorded on the SC-diet in this study. Rao et al. (2003) also reported that females that emerged from pupae with a higher mass were able to produce more eggs. Based on the comparatively higher larval survival and pupal mass on the SC-diet in this study, this diet is considered the most appropriate diet for rearing of C. partellus.

In a study done by Songa et al. (2004) (28 °C; 60–70% R.H.), larval duration of E. saccharina ranged between 30 and 33 days and pupal period lasted for 10 days. Ngomane et al. (2017) (26 ± 2 °C; 75 ± 5 R.H.) recorded larval duration periods of between 27 and 33 days on two different artificial diets. In this study, larval duration on the SC-diet was between three to six days shorter to compared studies. The pupal duration for both males and females were 3 days shorter when compared to that of Ngomane et al. (2017). Graham (1990) reported similar mean pupal weights of E. saccharina on artificial diet (126 mg) and field collected pupae (83 mg), indicating that pupae retrieved from all three artificial diets in this study, were equal in size and fitness, if not heavier. The range of pupal mass of E. saccharina reported by Gillespie (1993) ranged between 167 and 184 mg for females and 108–112 mg for males which is in the same range as this study. There was a difference in the E. saccharina sex ratio observed on the three different artificial diets. On the SC-diet there were far fewer females than males. It has been reported that nutritional variability in food resources could result in biased sex ratios, as differential mortality of sexes may occur, which lead to development of a skewed sex ratio under such conditions (Quezada-Garcia et al. 2014). The reason for the strong male-bias observed in this study is unclear.

Hamed and Nadeem (2008) evaluated different artificial diets for the rearing of H. armigera and reported the mean mass of final-instar larvae to range between 274.20 and 450.0 mg. Larvae from this study, feeding on the SC-diet was observed to be much heavier. The larval development period observed in this study was therefore shorter than those reported by others (14.5 to 42.8 days) (25 ± 2 °C, 75 ± 5% R.H.) (Hamed and Nadeem 2008) and (19.1 days) (25 ± 2 °C, 65 ± 5% R.H.) (Nair et al. 2019). In a study done by Abbasi et al. (2007) (24–26 °C; 70–75% R.H.) and Nair et al. (2019) (25 ± 2 °C; 65 ± 5% R.H.), pupation of H. armigera on their tapioca-based- and other artificial diets was recorded at 79% and 94.9% respectively, which indicates that the ML- and SC-diets used in this study were successful for rearing this species. The shortest pupal periods were recorded on the natural diet but males (13 days) and females (11 days) reared on artificial diets compared favourably with that reported by Hamed and Nadeem (2008) (11 to 14.8 days) (25 ± 2 °C; 75 ± 5% R.H.). The mean mass of pupae recorded on the SC-diet was higher than values between 250 and 325 mg reported on pigeon pea, sorghum and cotton (Bird and Akhurst 2007). Pupae retrieved from the ML- and SC-diets had a mass greater than those reported by Bird and Akhurst (2007), indicating good fitness and reproductive capabilities of H. armigera reared on diets used in this study.

When compared between artificial diets, S. calamistis had the highest larval survival and mass, as well as shortest larval development period on the ML-diet. This was predictable since this stem borers’ biology is very similar to that of B. fusca. The larval duration on the ML-diet for S. calamistis was similar to that reported by Songa et al. (2004) (35–36 days) (28 °C; 60–70% R.H.). A high pupation percentage was observed on the ML-diet, with a short pupal period compared to that reported by Songa et al. (2004) (12-days). The mean mass of pupae recovered from the ML-diet was in the same range and sometimes exceeded those reported in other studies. The mass of S. calamistis pupae on the ML-diet in this study was high. Shanower et al. (1993) reported mean pupal mass that ranged between 86 and 145 mg for S. calamistis reared on maize and different grass species, while the mean pupal mass of larvae reared on artificial diet was 210 mg (25 °C). It can be concluded that the ML-diet used in this study is suitable for the mass rearing of S. calamistis and that this diet can be used as a standard diet in laboratory bioassays.

A high percentage S. frugiperda larvae survived on all three artificial diets. The highest larval survival was recorded on the SC-diet, which was similar to the 90.7% reported by Da Silva and Parra (2013) for this pest on artificial diet (24 ± 1 °C; 70% R.H.). Jin et al. (2020) recorded pre-adult survival between 50 and 80% on artificial diet (25 ± 2 °C; 70–80% R.H.). One of the most recent studies by Truzi et al. (2021) also had high larval survival of 84–90% on artificial diets with larval periods ranging between 17.9 and 18.9 days. Lynch et al. (1989) reported similar larval periods (14.2–19.9 days) on different artificial diets (26.7 ± 2 °C; 80 ± 5 R.H.). Larval developmental period on artificial diet ranged between 12.1 and 13.9 days in a study done by Jin et al. (2020). The pupation rate on the three diets ranged between 77 and 79%, and the duration of pupal periods of males and females were 8–9 days and 7–8 days respectively. The pupal period ranged between 7.9 and 8.9 days in the study conducted by Jin et al. (2020). Martinez et al. (2004) and Truzi et al. (2021) reported the pupal period to be approximately 10 days long and pupal mass of between 238.0 and 252.7 mg on different artificial diets (26 ± 2 C°; 70–80% R.H.), which correlated with observations in this study. Lynch et al. (1989) reported a pupal duration of 22–29 days which is long compared to that observed in the current study. The highest mean pupal mass recorded in this study was on the SC-diet, which is within the same range than those reported by Botha et al. (2019) (180–237 mg) for different S. frugiperda populations reared on maize (Botha et al. 2019), Lynch et al. (1989) (202.1–256.9 mg) and Lima Pinto et al. (2019) (156.7–258.5 mg) on artificial diet. Pupal mass ranged between 173.5 and 232.2 mg for males and 200.0 to 236.8 mg for females on the different artificial diets in the study by Jin et al. (2020).

Conclusions

This study compared and identified suitable diets for six lepidopteran pest species of maize. Although B. fusca remains difficult to rear, the ML-diet evaluated in this study is currently the most suitable. The most suitable diets for rearing of the respective species were the SC-diet for C. partellus and S. frugiperda, the ML- and SC-diets for H. armigera, the ML-diet for S. calamistis, and the SASRI diet for E. saccharina.

The low level of survival (< 70%) of larvae on some of the artificial diets might be an indication of improper ratios of dietary components, the absence of crucial substances or presence of undesirable substances (Ochieng’-Odero 1994). These aspects should be further investigated to improve the diets use for mass rearing of these species.