By causing damage to the growing whorl and inflorescence, the fall armyworm, Spodoptera frugiperda (J.E. Smith) (Lepidoptera: Noctuidae) is a major pest of maize (Zea mays L.), sorghum [Sorghum bicolor (L.) Moench], and turfgrass in the southern United States. Grain yield losses due to fall armyworm attack on maize and sorghum can be as high as 34% (Lima et al., 2010) and 80% (Andrews 1988), respectively. Two strains of fall armyworm are known: the corn-strain that feeds on maize and sorghum and the rice-strain that feeds on rice (Oryza sativa L.) and turfgrass (Pashley 1985). Fall armyworms can be controlled by the use of insecticides, but that can be costly, because the insecticides have difficulty in penetrating the whorl and require frequent applications. Insecticides also reduce the abundance of natural enemies. Furthermore the repeated application of insecticides leads to the development of insecticide resistance in the fall armyworms. Insecticide resistance in fall armyworm populations has been reported for several different insecticide classes including organophosphates, carbamates, pyrethroids, and benzoylureas (Yu 1991, 1992; Diez-Rodrigues & Omoto 2001; Yu et al. 2003). Fall armyworms can also be managed by the use of plant genetic modification where genes from the bacterium Bacillus thuringiensis (Bt), each of which encode an insecticidal toxin, are inserted into the genomes of crop plants. However, the cultivation of transgenic plants can be controversial and may require the prevention of pollen dispersal, particularly for sorghum, which may be able to cross pollinate with wild relatives such as Johnsongrass (Sorghum halepense L. Pers.) (Morrell et al. 2005). In addition, as with conventional chemical controls, Lepidoptera can develop resistance to Bt toxins (Niu et al. 2013). An alternative to the use of synthetic insecticides or transgenic plants to control fall armyworm is the identification and use of plants that have innate genetic resistance or tolerance to fall armyworm feeding.

Screening host plants for genetic resistance to fall armyworm has been accomplished by planting under natural infestation in the field (Crubelati-Mulati et al. 2014) or by placing larvae directly on plants in the field (Diawara et al. 1992) or greenhouse (Ni et al. 2008). Natural infestations can be unpredictable and spatially variable in the field, and infestation of plants with fall armyworms in the field or greenhouse can often lead to the movement of larvae or adult moths to other host plants in close proximity or the loss of worms due to insect cannibalism or natural predators. Several different laboratory-based assays have been used to overcome these problems. Cheng et al. (2013) placed larvae in 30-mL plastic cups with agar and added fresh leaves daily from greenhouse grown plants. Plant material that is grown in the field can be lyophilized and mixed with media, and then fed to larvae in the laboratory (Williams et al. 2006), or fresh greenhouse grown tissue can be supplied daily to larvae in petri dishes with wet filter paper (Jessup et al. 2011) or in insect diet cups (Braman et al. 2000). Laboratory bioassays are best for insect confinement but as the number of accessions and the number of replicates increase, the use of insect diet cups or petri dishes becomes labor intensive because the leaf material must be replaced daily, water must be added, and containers must be cleaned. Nowierski et al. (1995) used agar plates supplemented with an antisenescence agent, benzimidizole, to assess the effects of the Russian wheat aphid (Diuraphis noxia [Kurdjumov]) on barley (Hordeum vulgare L.). The addition of benzimidizole helps to slow down the decay of excised tissue and prevents fungal growth while the agar helps retain leaf moisture. This study sought to determine if benzimidizole agar plates could be utilized to identify differences in maize and sorghum cultivars to fall armyworm feeding using minimal labor.

Fall armyworm resistant maize cultivars ‘Mp708’ (Williams et al. 1990) and ‘FAW1430’ (Ni, unpublished data) were used as resistant controls, while ‘AB24E’ (Williams 1989) was used as the susceptible control for the trials. The 4 sorghum cultivars used in the trials were ‘Honey Drip’ (PI 641821), ‘Collier’ (PI 641862), ‘AN109’ (Gorz et al. 1990) and Entry 22 which has a pedigree (A ‘Wheatland’ × ‘AF28’)-6-2-2-2-B, and was developed by Dan Gorbet, University of Florida. Sorghum and maize plants were sown on Jul 24, 2014 for Trial 1 and on Sep 3, 2014 for Trial 2 with 3 seeds per pot for 5 replicates (or pots). The pots contained a mixture of 5 gallons of masonry sand (Double A Concrete, Tifton, Georgia), 5 gallons of PRO-MOSS TBK peat moss (Premier Tech Horticulture, Quakertown, Pennsylvania), 5 gallons of coarse perlite, and 1,300 g of dolomitic lime. All of the pots of plants for the trials were randomized, and maintained in a greenhouse set at 32 °C. Plants were fertilized weekly with 0.5 × Hoaglands (Harris 2007) and grown for 36 d (Trial 1) or 37 d (Trial 2) after planting before the whorl tissues of the plants were excised to start the leaf-feeding bioassay.

Agar plates were prepared following the protocol of Nowierski et al. (1995). When the plants were 36 (Trial 1) or 37 d old (Trial 2), the top 2 leaves (the whorl) of each plant were removed, cut into approximately 7.6 cm long pieces, and 0.4 g (Trial 1) or 0.5 g (Trial 2) of the youngest leaf tissue from each pot (3 plants per pot) were placed onto 3 agar plates. Leaf tissue was arranged so that a neonate larva could travel easily from piece to piece with minimal time spent on the agar. The same procedure was used to set up all 15 agar plates per entry. The excised leaves corresponded to leaf 5–6 of sorghum or leaf 8–9 of maize for Trial 1 and leaf 5–6 of sorghum or leaf 4–6 of maize for Trial 2. Neonate larvae for Trial 1 were obtained from the USDA-ARS Insectary, Crop Protection and Management Research Unit, Tifton, Georgia. For Trial 2 the USDA-ARS Insectary had been shut down and neonate fall armyworm eggs were obtained from the Insectary from the USDA-ARS Corn Host Plant Resistance Research Unit, Mississippi State, Mississippi. For both insectaries, the fall armyworm strain has not been tested but the colonies were both established from insects collected on corn plants. One fall armyworm neonate larva was placed on a leaf piece per agar plate and the plate was sealed with Parafilm (Bemis North America, Neenah, Wisconsin). Plates were randomized within each replicate and maintained in a Percival Environmental Growth Chamber (Percival, Perry, Iowa) at 25 °C with a 12:12 h L:D cycle for 7 days. Each larva was removed from the plates and weighed on an analytical balance at the end of the trial (on the 7th d). Dead larvae that died at the neonate stage (no growth) were recorded as having a weight of zero.

Data were analyzed in SAS version v. 9.2 (SAS Institute, Cary, NC) using the GLIMMIX procedure. Cultivar, trial, and cultivar × trial were included in the model as fixed effects. Replication, nested within trial, was included as the random effect. The SLICE and SLICEDIFF options were used to compare cultivars within each trial. Significant differences were determined at P ≤ 0.05, adjusted for multiple comparisons using the TUKEY multiple range Test option.

The larval weight for Trials 1 and 2 were significantly different (F = 228.93; df = 1, 28; P < 0.0001). For Trial 2, larval weight was greater than Trial 1 (3.551 mg vs. 0.703 mg) and a cultivar by trial interaction existed (F = 5.87; df = 6, 146; P < 0.0001). Thus, cultivar differences were assessed for each trial separately. Larval weights were also significantly different among the cultivars tested (F = 16.48, df = 6, 146, P < 0.0001). For Trial 1, larvae fed susceptible maize line ‘AB24E’ had a significantly greater larval weight than those fed resistant lines ‘Mp708’, ‘FAW1430’, or the 4 sorghum lines (Fig. 1A). No significant differences were observed among the 4 sorghum lines (Fig. 1A). For Trial 2, the weights of the larvae maintained on ‘AB24E’ whorl leaf tissue were significantly greater than those fed on the resistant maize lines ‘Mp708’ or ‘FAW1430’ (Fig. 1B). For Trial 2, no significant differences in larval weight were observed for the sorghum lines except ‘Collier’, which had a similar larval weight as the larvae fed the susceptible maize line ‘AB24E’ (Fig. 1B). Furthermore, the benzimidazole agar plate protocol prevented leaf desiccation and prevented insect escape (Figs. 2A and 2B).

Fig. 1A.

Fall armyworm weights for 7 sorghum (off white) and maize (black) cultivars for Trial 1. Among the maize lines ‘AB24E’ was known to be susceptible and ‘Mp708’ and FAW1430’ were known to be resistant to fall armyworm feeding. ‘AN109’, ‘Collier’, Entry 22 and ‘Honey Drip’ are sorghum lines. In Trial 1, fall armyworm neonate larvae were taken from the Tifton, Georgia culture. Means with the same letter are not significantly different. Error bars represent one standard error of the mean.

Fig. 1B.

Fall armyworm weights for 7 sorghum (off white) and maize (black) cultivars for Trial 2. Among the maize lines ‘AB24E’ was known to be susceptible and ‘Mp708’ and FAW1430’ were known to be resistant to fall armyworm feeding. ‘AN109’, ‘Collier’, Entry 22, ‘AN109’, and ‘Honey Drip’ are sorghum lines. In Trial 2, fall armyworm neonate larvae were taken from the Mississippi State, Mississippi culture. Means with the same letter are not significantly different. Error bars represent one standard error of the mean.

For both trials, the fall armyworm larvae fed susceptible line ‘AB24E’ had significantly greater weight than those fed resistant lines ‘Mp708’ or ‘FAW1430’. Thus, the benzimidazole agar plate protocol accurately discriminated between known fall armyworm resistant and susceptible lines. Differences in larval weights between Trials 1 and 2 could be the result of many experimental factors. It is possible that the fall armyworms obtained from the Crop Protection and Management Research Unit for Trial 1 were not as vigorous as those obtained from Mississippi for Trial 2 as many of the larvae fed the resistant maize lines in Trial 1 died, whereas in Trial 2, larvae fed the resistant maize lines were alive on day 7 but smaller than the susceptible maize control ‘AB24E’. Differences in fall armyworm feeding between the two populations may be due to genetic differences between the two populations. Nagoshi et al. (2012) showed that fall armyworms in southern Georgia are progeny of migrants from Florida whereas worms in Mississippi are from migrants that overwinter in Texas.

For both trials, no significant differences in larval weight were observed for larvae fed ‘AN109’ and Entry 22. The larval weights for larvae fed the sorghum lines ‘Collier’ and ‘Honey Drip’ were significantly different only in Trial 2. To determine if there is a significant difference in larval weight between these 2 lines, further experiments should be conducted using the larvae from Mississippi.

Fig. 2A.

Seven day old fall armyworm larva (arrow) fed fall armyworm-resistant maize line ‘Mp708’ using the benzimidazole agar plate method.

Fig. 2B.

A 7 day old fall armyworm larva (arrow) maintained on fall armywormsusceptible maize line ‘AB24E’ (susceptible control) using the benzimidazole agar plate method.

Using the agar plates with the addition of benzimidazole, the excised maize and sorghum leaf tissues showed little leaf desiccation and low fungal contamination during the 7-d experimental period (Figs. 2A, 2B). Furthermore, this agar plate bioassay also eliminated the onerous daily replacement of fresh leaf tissues. In the previously reported methods, fresh leaf tissues or water had to be added daily (Braman et al. 2000; Jessup et al. 2011) or every other day (Cheng et al. 2013), which is laborious and larvae may be stressed or harmed during the process. Once the experiment was set up, there was no maintenance needed during the benzimidazole agar plate experiment until the termination of the trial at 7 days after its initiation. Thus, this protocol has greatly optimized the technique in conducting fall armyworm excised leaf-feeding bioassays, which could be a valuable alternative to the feeding bioassays using insect diet cups or Petri dish plates. Furthermore, this new method has the potential to be used with tissue from a myriad of plant species and insects.

The authors thank Darika Bunphan and Penny Tapp for their technical help, and Wilfred Vermerris (University of Florida) for providing seed for sorghum Entry 22