Pearl millet (Pennisetum glaucum (L.)R. Br.; Poales: Poaceae) is grown on 5 continents including Asia, Africa, North America, South America and Australia, while its use varies from one continent to another. It is estimated that 26 million hectares of pearl millet are grown in Africa and India as food grain (Gulia et al. 2007). In the United States, about 607,000 hectares of pearl millet are cultivated annually mostly in North Dakota, Nebraska, South Dakota and in some southeastern states including Georgia and Florida where it is used as hay and a summer grazing crop (Dewey et al. 2009). Pearl millet is also grown for other purposes such as pasture, silage, food, building material, seed crop and fuel (Baker 1993). The commercial use of pearl millet in the United States began in early 1990s, although the crop was introduced in South Dakota during the early 1980s (Sedivec & Schatz 1991). The crop has nutritional benefits and is a good source of protein, crude fiber, calcium and phosphorous for animal feed. The protein content of pearl millet is higher than that of maize (Zea mays L.; Poales: Poaceae); it ranges from 27% to 32% and also has a higher energy content (Ejeta et al. 1987; Davis et al. 2003) compared to wheat (Triticum aestivum L.; Poales: Poaceae), maize and sorghum (Sorghum bicolor (L.) Moench; Poales: Poaceae) (Hill & Hanna 1990). In the southeast, early-maturing hybrids can be planted from late April to early August (Wilson et al. 2006). The plant attains physiological maturity at 75 to 85 days after seed germination (Dewey et al. 2009).

Pearl millet has been reported to give appreciable yield with low nutrient availability and higher yield with increased fertilizer application (Gascho et al. 1995). A boost in grain yield was reported by Maman et al. (1999) and Limon-Ortega et al. (1998) when nitrogen was applied at 78 kg ha^-1. According to Menezes et al. (1999), a nitrogen rate of 112 kg ha^-1 was found to be optimal for high grain yield. However, yield can be reduced irrespective of N-fertilizer application under late planting (Rurinda et al. 2014) and drought conditions (Bashir et al., 2014).

Pearl millet and sorghum, being C4 species, have high photosynthetic efficiencies, high dry matter production ability (Rai et al. 1999) and closely related developmental stages (Maman et al. 2004). It is estimated that 30% of yield loss in sorghum is due to insect pest and pathogen attack (Reddy & Zehr 2004). Nwanze & Harris (1992) reported that less than 12 insect species are of economic importance on pearl millet in West Africa, although millet can attract up to 150 insect species. Agriculture in the southeastern U.S. has been threatened by stink bugs such as Euschistus servus (Say) (Hemiptera: Pentatomidae), Leptoglossus phyllopus (L.) (Hemiptera: Coreidae) and southern green stink bug (Nezara viridula (L.); Hemiptera: Pentatomidae). They are considered primary pests on grains, fruits, seeds and vegetable crops (Schaefer & Panizzi 2000). Chinch bug and European corn borer are known for causing significant damage to pearl millet (Andrews et al. 1996), but according to Kennedy (2002) and Hudson (1995), chinch bug causes the most damage in pearl millet fields, especially when grown closer to maize or sorghum fields, since the insect can quickly move to the later-planted pearl millet.

Insect damage in pearl millet can occur on foliage, flowers as well as seeds, and has been recorded across all plant growth stages, i.e., 3rd leaf stage, 5th leaf stage, head initiation, flag leaf stage, boot stage, 50% stigma emergence, milk stage and dough stage (Maiti & Bidinger 1981). Before any insect control strategy is implemented, an assessment of insect occurrence and impact needs to be conducted to establish the economic threshold of the insect damage (Youm & Owusu 1998). Therefore, with increased interest in growing pearl millet because of its broad uses, the objectives of this study were to determine occurrence of insect pests from emergence to physiological maturity, and to identify and assess potential damage of major species on various pearl millet genotypes under different levels of nitrogen fertilization.

Materials and Methods

This study was conducted at Alabama A & M University's Winfred Thomas Agricultural Research Station (WTARS) located in Hazel Green, Alabama (N 34° 54′ 57.600″ W 86° 38′ 49.600″; 248.1 m). The experiment was arranged in a randomized complete block design with 4 replications in which genotypes and N fertilizer levels were allocated plots in factorial combinations. Each plot measured 3.05 × 1.14 m and seeds were sown at the rate of 5 kg ha^-1 at a spacing of 12.7 cm within row and 70 cm between rows. The genotypes consisted of two openpollinated lines, ‘2304’ and ‘LHB08’ and 2 hybrids, ‘606A1*2304’ and ‘707A1*4280’. The plots were planted on 7 June, 2010 and nitrogen treatments of 0, 40, 80, and 120 kg ha^-1 were applied upon crop establishment.

The plants were monitored for insect incidence from the time of emergence to harvest (14 Oct 2010). The insect sampling on weekly interval was carried out starting booting stage (pre-emergence of head) when insect density on plant became noticeable and carried through plant maturity. Insect sampling was carried out between 8:00 a.m. and 11:00 a.m. on 5 randomly selected plants along a diagonal transect positioned across each plot. The insects were collected using sweep net and by shaking the heads/panicles into transparent plastic bags. The insect samples were placed in a standard freezer for storage, subsequent insect count and identification.

Damage to foliage was determined and rated on a 0 to 5 scale where 0 = no damage, 1 = less than 25% of leaves damaged, 2 = more than 25% of leaves damaged but less than 50%, 3 = more than 50% of leaves damaged but less than 75% and 4 = more than 75% of leaves damaged but less than 100% and 5 = 100% of leaves damaged. Percentage leaf damage was the relative percentage of leaf damaged on the entire plot in relation to the canopy. At maturity, a similar scale was used to assess the head damage. Head damage is the cumulative ratio of damaged panicle to the total number of panicles within each plot.

At harvest, 5 randomly selected plants were cut at ground level, panicles removed and remainder dried to a constant weight individually for vegetative dry weight plant-1 estimation. The whole plots were harvested for estimation of grain yield. Insect count was transformed using log10 before analysis. Data were analyzed using General Linear Model (GLM) procedure in Statistical Analysis System (SAS) package ver. 9.3. Treatment means were separated using Duncan's Multiple Range Test at < 0.05.

Results

NITROGEN

The rate of nitrogen applied did not significantly affect leaf damage across the genotypes (Table 2). However, incidence of different insects varied with nitrogen and genotypes. The pattern observed for L. phyllopus indicated high frequency for all 4 nitrogen rates, on all 4 genotypes and nitrogen rates. The number of L. phyllopus increased with increasing N levels (0 kg ha^-1 < 40 kg ha^-1 < 80 kg ha^-1< 120 kg ha^-1) as shown in Fig. 1. Schistocerca americana was the second most domi nant insect across all treatments. It was more prevalent in the pearl millet genotypes treated with 0, 80 and 120 kg ha^-1 nitrogen (Fig. 1), while A. mellifera was the third most prevalent insect species but the population decreased with increase in N rate. Responses of beneficial insects to N treatment varied. Some beneficial species increased with increasing nitrogen rate e.g. Halictus spp. and D. plexippus, while low numbers of H. axyridis, T. geminatus and Halictus spp. were recorded across genotypes irrespective of nitrogen treatment. The relative abundance of beneficial insect on 4 genotypes showed the following trend: LHB08 > 606A1*2304 > 707A1*4280 > 2304. Although general insect damage was observed across test plots, leaf damage on plants that received different levels of nitrogen did not differ statistically. A trend, however, indicated an increase of damage from low to high nitrogen rates (Table 2).

Table 1.

Insect species, and their time and stage of occurrence on pearl millet grown in Northern Alabama.

LEAF AND HEAD DAMAGE, AND YIELD

Leaf damage on pearl millet treated with the 4 nitrogen rates and the 4 genotypes over the growing season did not differ statistically. Foliar feeding damage was caused by S. americana, M. differentialis, H. zea and E. pestifera (Table 2). Head damage was prominent on some of the plants and could be attributed to pod sucking insects, i.e. L. phyllopus and E. servus. Head/panicle damage was greatest at 120 kg ha^-1 N, which was significantly different from that at 80 kg ha^-1 N. Higher levels of head damage by insects on genotype ‘707A1x4280’ and ‘2304’ were recorded, which were significantly different from damage on ‘LHBO8’ and ‘606A1*2304’ (Table 2). Although there was not any significant difference in grain yield across nitrogen regimes, grain yield varied among genotypes (Table 2). There was no significant difference in the vegetative dry weight between N treatments and the 4 genotypes (Table 2). Estimated grain yield of genotype LHBO8 (6,154 kg ha^-1) was significantly greater than that of all the other genotypes across nitrogen rates.

Table 2.

Leaf damage, head damage, vegetative dry weight and estimated yield for 4 nitrogen rates and 4 pearl millet genotypes.

Discussion

This study showed that L. phyllopus can be considered as the most prevalent insect on pearl millet grown in Northern Alabama, followed by S. americana. According to Kuhar et al. (2010), the piercing and sucking injury caused by L. phyllopus is very similar to that of the chinch bug (Blissus leucopterus Say), which has been identified as the most economically important insect of pearl millet (Hudson 1995). There was a general increase in the population of L. phyllopus and S. americana with increasing nitrogen rate. This indicates a pattern of preference by these insects for plants with higher nitrogen levels. Coley et al. (1985) have explained that such insect population increases could be due to the decrease in plant defenses against herbivores at enhanced nutrient conditions. Nabity et al. (2012) found an increase in efficiency of conversion of feed to body mass in insects at higher soil N fertility levels.

Fig. 1.

Number (log ₁₀) of insects per plot on pearl millet under 4 nitrogen rates.

Increases in the populations of L. phyllopus and S. americana at different times during the season were inverse in occurrence. Reduction in the population of S. americana occurred later in the growing season, which may be due to onset of senescence, drying and shedding of foliage and thus, decreased food availability. Sharma et al. (1997) reported that several species of grasshopper including the desert locust, S. gregaria, caused damage to grasses such as sorghum by feeding on leaves, flowers and grain. Leptoglossus phyllopus was more prevalent later in the season when grain filling was at its peak. Leptoglossus phyllopus is known to be a grain feeder (Abudulai et al. 2001), hence increasing the potential for head damage and yield loss. Cover crop management in neighboring field can be chopped and disked in September before the L. phyllopus becomes attracted to crops (Mead, 2010). Although the population of E. servus was sparse in general, it had its peak at 66 DAP when pearl millet was at milk stage (Maiti & Bidinger 1981). Mizell et al. (2008) also reported that stink bug prefer to feed primarily on the seeds of grain crop when they are in milk stage.

Table 3.

Densities of various insect species (per plot) on pearl millet from 61 to 109 days after planting.

This study gives credence to the fact that pearl millet growing in this region is a desirable host plant for many species of insect pests that can cause severe crop damage if not controlled. Our results indicate that certain insect species congregate and harm the crop at higher incidence with increase in nitrogen rate. Because of the diverse uses of pearl millet as animal feed, hay, pasture and silage in the U.S., growing pearl millet requires careful selection of high yielding genotypes in combination with proper nitrogen management to reduce pest damage.

Fig. 2.

Number (log ₁₀) of different insects per plot on 4 pearl millet genotypes.

Pearl millet has been reported to produce substantial yield on marginal land and with minimal N input (Obeng et al. 2012). This study shows that higher N rates may result in enhanced insect infestation impacting the yield. However, further studies are needed to establish appropriate nitrogen regimes that can support optimum reproductive growth and grain development with minimal insect pest infestation.