The Central American locust, Schistocerca piceifrons ssp. piceifrons (Walker), often becomes a major pest of staple and cash crops in Mexico and Central America (Barrientos-Lozano et al. 2002). In 1993, laboratory and field studies were initiated to develop a biological control strategy for S. piceifrons. The National Centre for Biological Control (Centro Nacional de Referencia de Control Biológico-CNRCB) has in its entomopathogen collection 35 isolates of Metarhizium anisopliae var. acridum (M. a. acridum) from S. piceifrons in Mexico. Our laboratory studies have shown that the isolates MaPL32 and MaPL40 are 2 of the most virulent against S. piceifrons (Barrientos-Lozano et al. 2002).

Field trials with M. a. acridum oil-formulated spores have demonstrated effective control of S. gregaria Forskal in Africa (Langewald et al. 1997), Locusta migratoria (L.) and Chortoicetes terminifera (Walker) in Australia (Hunter et al. 1999, 2001), Rammatocerus schistocercoides (Rehn) in Brazil (Magalhaes et al. 2001) and S. piceifrons in Mexico (Hernández-Velázquez et al. 2003). One of the perceived disadvantages, evident in these trials, is the length of time (13 to 14 d) that the target insect takes to die after the application. However, in practical terms, successful control of locust is reflected by a reduction in food consumption; a locust that has ceased to feed due to infection is no longer a significant pest (Moore et al. 1992). Studies by Moore et al. (1992) and Seyoum et al. (1994) both reported significant reductions in feeding by S. gregaria following infection with M. a. acridum; similar results were also observed with the grasshopper Zonocerus variegatus (L.) (Thomas et al. 1997). All of these reported studies were carried out using a high dose of the pathogen, in some cases with few insects per treatment, without controls, and a mixture of instars. The aim of the present work was to examine the effect of a conidia low dose (6 × 10⁴) per S. piceifrons adult on both feeding rate and feces production.

First instars of S. piceifrons locusts were collected in Tizimín, Yucatán, Mexico. Locusts were kept in 1 × 1 × 1 m cages inside a controlled environment (CE) room and fed on a diet of leaves of the grass Cynodon dactylon (L.) and rolled oats. The mean daily temperature and relative humidity (RH) in the CE room was 27°C and 70 ± 5% respectively, with a photoperiod of 12:12 h (L:D).

The fungal isolate was a monospore isolate (EH-502/8) of M. a. acridum MaPL40 from S. piceifrons (Hernández-Velázquez et al. 1997). Fungal conidia were produced by a diphasic method. In the first phase, the fungus was cultured in a liquid medium of 2% sucrose and 2% yeast extract (Jenkins & Prior 1993). Fungal biomass was transferred to rice in a plastic bag (250 g/bag) with a cotton filter for further growth and sporulation. After 14 d the bags were opened in the CE room which allowed the fungus to dry to a moisture content of approximately 9%. Conidia were separated from the rice by sieving through a 300-μm mesh. The resulting conidia powder was then re-sieved through a 90-μm mesh to remove any remaining rice-dust particles.

Conidial suspensions were formulated in citroline (a mineral oil derived from petroleum for agricultural use in Mexico) by mixing 1 g of conidial powder in 100 mL with a magnetic stirrer. It was then adjusted to 3 × 10⁷ conidia/mL with use of a Neubauer haemocytometer. Germination rates were determined by randomly examining 100 conidia per plate with a compound microscope (400×). A conidium was considered to have germinated if the germ tube was at least as long as the width of the conidium. Germination of conidia was 95-100% immediately after formulation and just before locust infection.

Each treatment consisted of 20 adult locusts, 7 d post-fledging, each receiving 2 μL of the formulation applied on the pronotum via a Hamilton SGE syringe. Control groups received 2 μL of either citroline or sterile distilled water with Tween 80 (0.1%, v/v, Difco™) without conidia. Following treatment, locusts were placed in subgroups of 4 insects in 13 × 10.5 cm plastic containers, covered with muslin. Containers were maintained at 28°C and 70 ± 5% RH. The experiment was replicated 4 times. Mortality was recorded daily for 12 d; dead insects were placed individually in Petri dishes on moist filter paper at 27°C for 4 d to allow growth of mycelia on the insect surface. Each day, between 1.945 to 2.557 g of fresh C. dactylon seedling leaf was given to each subgroup of 4 locusts. At the beginning of the experiment, 4 samples of fresh C. dactylon seedling were dried for 24 h at 90°C to determine the seedling dry weight percentage so food supply could be calculated. Every 24 h the uneaten seedlings were removed from each container, and dried at 90°C for 24 h until reaching a constant weight. In addition, feeding was assessed indirectly by monitoring feces production. Feces were collected every 24 h. Individual samples were dried at 90°C for 48 h until reaching a constant weight.

Cumulative mortality until d 12 of the experiment, among the treated and negative control groups was analyzed by analysis of variance (ANOVA, α = 0.05). Multiple comparisons were used to determine significant differences between means of infected and citroline negative control locusts at P < 0.05 (Tukey test). All data from food consumption and feces production between infected and citroline negative control locusts were analyses by a t-test (α = 0.05) for each day until d 7. Since there was no difference in food consumption and feces production between both negative controls (citroline and Tween 80), Tween 80 was excluded from the statistical analysis.

The cumulative percent mortality of S. piceifrons inoculated with fungal conidia reached 97.5 ± 2.5% at d 12. Significant differences (P < 0.01) in mortality were observed between the treated and control groups since d 6 and thereafter until d 12 of the experiment (data not shown). Mortality did not exceed 10% in either control group throughout the 12 d of the experiment.

The result of food consumption in milligrams of dry weight/insect between the treated and citroline control groups is shown in Table 1. A significant (P < 0.05) reduction in food consumption was observed between treated and citroline control groups from d 4 to 7, although a reduction was evident by d 3 (P = 0.07). During the first 24 h of the experiment, food intake was low in both groups; 67.9% lower than the second day. After 7 d, on a per insect basis food consumption was reduced by 82.24%.

Table 2 presents the result of feces production in milligrams of dry weight/insect between the treated and citroline control groups. A significant (P < 0.05) reduction of feces production was observed between the treated and citroline control groups at d 3. The mean feces production from the control group was 64.01 mg, and during the first 24 h of the experiment, the mean of both groups was 51.62 mg. After 7 d, on a per insect basis, feces production was reduced by 76.45%.

This study provides further evidence that infection by M. a. acridum causes a significant reduction in S. piceifrons feeding by both direct (food consumption) and indirect (feces production) evaluation. Feces production showed a significant reduction (d 3), 24 h before food consumption (d 4). During the first 24 h before insect treatment, food consumption was very low, probably due to stress induced in locusts during manipulation, as already noted by Moore et al. (1992) in S. gregaria.

The observed effect on food consumption and feces production at d 3 and 4 after infection agrees with findings by Moore et al. (1992). They noted a reduction in the desert locust feeding 24 h after application of 8 × 10⁵ conidia/insect of M. a. acridum, and at d 3 with 8 × 10⁴ conidia/insect. The effect on feeding, movement, and flying abilities of the locust is related to the development and colonization of the fungus, rather than toxin production (Freimoser et al. 2003). Seyoum et al. (1994) observed flight impairment in the desert locust, S. gregaria, after 3-4 d of M. a. acridum infection, and attributed it to a reduced availability of metabolizable fuel and physical damage to muscles and nerves caused by fungal colonization. In practical terms, the observed feeding reduction in our study would reflect a non-significant pest very early after Metarhizium infection.

This study was financed by grant G-31451-B of Consejo Nacional de Ciencia y Tecnología (CONACYT) of Mexico. We also thank Dr. Julio Bernal (Texas A&M University) for the review of the manuscript, and Mrs. Ingrid Masher for editorial assistance.

Summary

This study examined the effects of Mexican strain (MaPL40) Metarhizium anisopliae var. acridum on feeding and fecal production of adult Central American locusts, Schistocerca piceifrons ssp. piceifrons. Locusts infected with 6 × 10⁴ M. anisopliae conidia per insect showed 97.5% mortality in contrast to 13% in negative control locusts after 12 d of the infection. A significant (P < 0.01) reduction in feeding was observed, as indicated by Cynodon dactylon leaf consumption and feces production at 4 and 3 d, respectively. The data reflect a non-significant pest very early after Metarhizium infection.