R.M. Martín, J.R. Cox, D.G. Alston,¹ and F. Ibarra F.

Forage and Range Research Laboratory, USDA-ARS, Utah State University, Logan, UT 84322-5305.

ABSTRACT The spittlebug, Aeneolamia albofasciata Lalleman, is the most economically important insect pest of buffelgrass pastures in Mexico. In 1984-1986 we studied the effect of climate on the life cycle of this insect in Sonora, Mexico, by monitoring spittlebug nymph and adult densities in relation to summer precipitation and maximum and minimum temperatures. The amount of summer precipitation appeared to be the most important climatic factor affecting spittlebug population density and duration of nymphal and adult life stages. The correlation coefficient between cumulative summer rainfall and density of nymphs was 0.93, and for adults it was 0.92 in the 3 yr. Mean maximum, and minimum temperatures, however, were not related to adult or nymphal densities in any year. Hatch of overwintered eggs was initiated after 45-60 mm of summer rainfall. Total and peak adults were higher in 1984 when precipitation was two times above normal (320 mm) than in 1985 and 1986 when precipitation was below and approximately average. Insect densities declined when precipitation was below normal (286 mm) in 1985. Life-cycle duration from the first-instar nymph through adult averaged 43� 3 d across the 3 yr. The spittlebug was univoltine in 1985 and 1986, however, during wet years (such as 1984), nymphal and adult stages can be present for a longer period.

Productivity of arid and semiarid rangelands around the world has been improved through the introduction of exotic grasses such as buffelgrass, Cenchrus ciliaris (L.) Link (Humphreys 1967, Gonzales and Dodd 1979, Cox et al. 1988, Walker and Weston 1990). This perennial, warm-season bunchgrass, native to Africa, was introduced into Sonora, Mexico, in 1954. Since its introduction, the species has been successfully established on >400,000 ha of rangelands in Sonora (Ibarra and Cox 1990, Martin and Ibarra 1991), and there is potential for its establishment on 4 million additional hectars (Cazares et al. 1986). Buffelgrass forage production exceeds that of native grasses by 10 times, and stocking rates are frequently 3- to 10- fold greater than on native rangelands (Hanselka 1985, Ibarra and Cox 1990).

Although the grass is an ideal agronomic selection to improve rangelands (Contreras 1964, Cazares et al. 1985, Martin and Ibarra 1991), it is susceptible to injury by the spittlebug, Aeneolamia albofasciata Lalleman. Because forage grasses have low production costs and low crop values per hectare compared with cultivated crops, greater insect injury will be tolerated before economic thresholds are reached and control measures are initiates. However, spittlebug damage dramatically increases the failure of buffelgrass establishment and decreases productivity of stands (Martin et al. 1985). Therefore, relatively low cost insect control practices are justified.

The spittlebug is a wide-ranging pest in cultivated and wild plants. Host plants in Sonora include corn, Zea mays L,; rice, Oriza sativa L.; sorghum, Sorghum vulgare L.; stargrass, cynodon plestostachyus (Kschum) Pielger; bermudagrass, Cynodon dactylon L.; rhodesgrass, Chloris gayana L.; Johnsongrass, Sorgum halepense L,; black grama, Bouteloua eriopoda Torr,; bahiagrass, Paspalum notatum Flugge,; and buffelgrass (Coronado 1978, Cazares et al. 1985). In 1958, spittlebug damage was reported on crops in eastern Mexico (Coronado 1978). Since then the insect has spread along the Pacific coast, and it now occurs in the states of Chiapas, Oaxaca, Guerrero, Colima, Nayarit, and Sinaloa (Arrieta and Coronado 1968, Flores 1968) (Fig. 1).

Spittlebug infestations on native and exotic grasses were first reported in Mexico in 1947 (Marrufo and Enkerlin 1974), and their dispersal to northwestern Mexico may be related to seeding rangelands with buffelgrass. Along the Gulf of Mexico, spittlebug populations dramatically increased by 1955. In the 1960s, 40,000-60,000 ha of pasture grasses were injured by spittlebug (Velazco et al. 1969). In Sonora, spittlebugs were first reported in buffelgrass pastures in 1981, and by 1984 grass plant populations began to decline (Lopez 1984). In 26 Sonoran counties, Cazares et al. (1985) reported spittlebug nymph densities between 45 and 212 per buffelgrass plant.

Spittlebug numphs feed on juices from lower plant parts (xylem and parenchyma) and from small roots at the soil surface (Coronado 1978). Injury reduces grass vigor and health (Fewkes 1963, Hagley and Blackman 1966). Adults feed on stems, crowns, and young succulent levaes. Injured leaves develop charotic areas around puncture sites. Adults suck sap and inject a systemic toxin known as diastatic oxidase. The compound disrupts plant nutrient transport and respiration (Enkerlin and Morales 1979, Willson and Valerio 1989).

Along the Gulf of Mexico, Spittlebug management options have been developed for guinea, Panicum maximum Jack; pangola, Digitaria decumbens Stnt.; merkeron, Pennisetum purpureum Ashum; para, Bracharia mutica L.; and buffelgrass. Options include chemicals such as carbamates (carbaryl), organochlorines (DDT, Gamma-BHC, Dieldrin), organophosphates (diazinon, malathion, parathion) (Canavati 1974, Purata 1974), shredding and grazing (Velazco et al. 1969), burning (Ibarra and Enkerlin 1974), and planting tolerant buffelgrass accessions (Stimmann and Taliaferro 1970, Flores and Velazco 1974, Agostini 1980, Villarreal 1980, Ramirez 1981, Morales 1985). In the laboratory, spittlebug has been controlled with a fungus (Metarrhizium anisopliae Metchn), but this has not been successful in the field (Gaugliumi 196), Flores et al. 1965. Painter 1968, Coronado and Sosa 1966, Ramos 1976, Coronado 1978).

The spittlebug life cycle has been documented on agronomic crops and grasses in eastern Mexico. Climatic conditions, agriculture, and livestock production systems in eastern Mexico differ from those in northwestern Mexico. The main climatic differences are the amount and distribution of precipitation, and relative humidity. Buffelgrass pastures support = 80% of the livestock produced in northwestern Mexico (Ibarra and Cox 1990). Spittlebug population control measures used in humid eastern Mexico may differ from best management practices for arid and semiarid northwestern areas. Development of an effective spittlebug control program for buffelgrass pastures in Sonora must be based on ecological and biological sutdies. No previous attempt has been made to determine the spittlebug life cycle in arid buffelgrass pastures. The objective of this study was to determine the relationship between precipitation and temperature and spittlebug life cycle in buffelgrass pastures in northwestern Mexico.

Study Site Description. The study site was located 82 Km north of Hermosillo in north central Sonora, northwestern Mexico (29^o 41′ N, 115^o 57′ W) at the Carbo Livestock Research Station. The study site, a typical desert shrubland of the Sonoran Desert, was mechanically cleared and seeded to buffelgrass during 1970. Elevation at the site is 470 m, with gentle slopes that vary from 1 to 3%. The soil is an Anthony fine loam (thermic typic torrifluvent) of recent alluvium (2-6 m deep), weathered from granitic rocks, moderately basic (pH = 8.5-8.9) (Hendricks 1985).

Average annual precipitation for a 24-yr period (1970-1994) is 320 mm (Centro de Investigaciones Pecuarias del Estado de Sonora 1989). Precipitation is bimodally distributed; = 60% occurs between July and September, and 40% occurs between October and april, May, June, and September are usually dry. Daytime temperatures average 34^oC in summer, but may frequently exceed 40^oC in June and July. Night time temperatures average 5^oC in winter, and may approach 0^oC in January and February (climatography of Mexico 1982).

Daily precipitation and mean maximum and minimum temperatures were recorded at the Carbo Livestock Research Station.

Experimental Design. Nine 1-ha, dense shrubfree buffelgrass stands naturally infested with spittlebugs were selected randomly and fenced in the spring of 1984 to exclude livestock. Three stands were selected randomly for daily sampling between 1 July and 15 September 1984, three others were sampled between 1 July and 15 September 1985, and the remaining three were sampled between 1 July and 15 September 1986. The experimental design was a completely randomized design. Treatments were the years of study.

Egg Sampling Procedure. To obtain an assessment of egg densities, five soil samples per hectare plot were collected randomly at 0-to 5-cm depth with a 12.0-cm soil auger from the base of buffelgrass plants and between buffelgrass plants. The stands were sampled between the second half of June and the beginning of July in each year from 1984 to 1986. Each sample was miced, washed, and passed through a 0.5-mm diameter sieve. Sieved material was soaked in a 50% solution of NaCL (Fewkes 1961). Each sample was placed in a culture dish (100 by 15 cm), and eggs in five 1-cm² grid sections were counted. Subsample counts were converted to represent total area of dish and thus a total soil volume of 0.05 m³.

Nymphal Sampling Procedure. On each sampling date, five quadrats (1 by 1 m each) were selected in each stand. five nymphal instars were identified by differences in size, appearance, and wing pad length after the first instar (Bodegas 1973). Nymphs in each developmental stage were counted on plants within each quadrat. There were typically five plants per square meter. Nymphs were removed from spittle mass for ease of counting and then returned. Because of disturbance to spittle masses, the same plants (i.e., quadrats) were not sampled again. Sampling was conducted every other day from 1 July to 15 September in 1984, 1985, and 1986.

Adult Sampling Procedure. Adult insects were sampled with light traps between 15 August to September 15 in each year. One light trap was installed in the center of each 1-ha buffelgrass stand. The trap consisted of a kerosene lantern hung at 1.5 m above a metallic pan (1.0 by 1.0 by 0.1 m) filled with kerosene. When the nymphal population approached fourth and fifth instars, light traps were ignited during 30 consecutive days from sundown to sunrise. Captured adult insects were counted daily and sexed. Kerosene was replenished as needed. Adult populations also were sampled within each 1-ha plot by sweeping 100 times daily walking in a zig-zag fashion.

Data Analyses. To assess differences in adult and nymphal densities among years, adult densities were averaged over all collection days and densities of each nymphal stage were averaged among the five quadrats and over all sampling dates. Adult and nymph mean densities were log-transformed (log x + 1) and a one-way analysis of variance (ANOVA) was used to determine differences among years (P<0.05). A Tukey’s test was used to separate means (Steel and Torrie 1980). Total nymph and adult spittlebug population densities per plot were correlated (CoStat 1986) with accumulative summer rainfall and daily maximum and minimum temperatures. The total duration of each developmental stage except eggs was determined by the number of days from first observation to last during each year.

Eggs. During the 3 yr, over-wintered eggs began to hatch in early summer after accumulation of 45-60 mm of rainfall. Most eggs were near buffelgrass crowns, and densities ranged from 150 to 200 eggs/m³ in the top 0-5 cm of soil.

Egg hatch began on 7 July 1984 after accumulation of 60 mm of rainfall; on 21 July 1985 after 46 mm rainfall; and 14 July 1986 after 45 mm of rainfall. We assume that the earlier hatching time in 1984 was caused by a greater amount and distribution of rainfall during the months of May and June in 1984 versus 1985 and 1986. These results differ from those in the Gulf of Mexico where egg hatch did not begin until after 80 mm of accumulative precipitation (Ibarra and Enkerlin 1974).

Along the Gulf of Mexico, egg hatch began when ambient temperatures were between 24 and 26^oC in July and August (Ibarra and Enkerlin 1974). In western Mexico, eggs began hatching when daily temperatures averaged between 28 and 37^oC during Juli (Lopez 1984). In this study, daily temperatures averaged between 28 and 30^oC during the time of egg hatch.

Observations during this 3-yr study indicate that spittlebug females preferentially select egg laying sites under the base of buffelgrass plants. During field sampling we commonly measured three to five first-instar nymphs per square meter in areas partially covered with litter between buffelgrass plants. However, densities were 5-10 times greater under dense litter accumulations at the base of plants.

Nymph Populations. We identified five instars (Fig. 2), the same number found by other researchers in eastern Mexico (Ibarra and Enkerlin 1974, Coronado 1978). In the summer of 1985, we did not find first-instar nymphs after small storms on 4, 9, 10, 11 and 12 July, but nymphs were present after heavy dew between 15 and 20 July. The first- and second-instar nymphs are small and difficult to locate in dense buffelgrass crown litter. First- and second-instar nymphs were most frequently observed feeding on fresh crown tissue and fine roots at the soil surface.

As air temperatures increased during the day, nymphs congregatied in groups (three to five nymphs per group) and secreted a bubbly (spittle) liquid. When nymph populations were high during the summer of 1984 and surface soil was moist, the spittle mass was frequently found beneath the soil surface near buffelgrass crowns. Willson and Valerio (1989) reported higher spittlebug nymph population densities under litter in Brachiaria decumbens pastures. The total time period that first and second nymphal stages were observed ranged from 4.1 to 5.2 and 4.8 to 5.6 d during the 3 yr, respectively. The mean duration of first and second nymphal instars was greater in 1984 and 1986 than 1985 (Table 1).

The third and fourth instars congregated on culms and leaves 10 cm above the soil surface and feed during morning and late afternoon hours. During the wet summer of 1984, nymphs migrated to a zone 10-30 cm above the soil surface, but in the dry summers of 1985 and 1986, nymphs migrated to the dense canopy litter at the plant base. The total time period that the third and fourth instars were observed ranged from 4.3 to 6.5 and 5.0 to 6.2 d during the 3 yr (Table 1). The mean duration of third and fourth nymphal instar was greater in 1984 than in 1985 and 1986.

The fifth-instar nymphs were observed feeding on leaves near the canopy. Mean duration of the fifth instar ranged from 5.3 to 7.4 d (Table 1) and was longer in 1984 and 1986 tnan in 1985. Final metamorphosis to adult occurred in the indivual spittlemass (Fig. 2).

Correlation (r) between mean total nymph density and accumulative precipitation (P<0.05) was 0.93 in 1984, 0.83 in 1985, and 0.92 in 1986. Nymphal densities were not (P<0.05) correlated with eitheir maximum or minimum temperatures (r = 0.060-0.35).