Megamelus scutellaris Berg (Hemiptera: Delphacidae) is a South American species that feeds on waterhyacinth, Eichhornia crassipes Mart. (Solms). This species exhibits significant wing dimorphism whereby fully winged adults (macropters) are capable of flight while those with reduced wings (brachtypters) are not. The wing form is determined by a developmental switch triggered by environmental factors including crowding, host plant quality, temperature, and photoperiod. This study examined the influences of insect density and host plant quality on M. scutellaris wing dimorphism, development, and biomass as well as their effects on E. crassipes. Two experiments exposed a single generation of M. scutellaris to lower and higher densities of conspecifics on low and high quality plants. The first experiment involved transferring second instars to test plants at loads of 50, 5, or 1 g of fresh weight plant biomass per nymph, which resulted in mean densities of 2, 15, and 69 nymphs, respectively, on both low and high quality plants. A second experiment exposed test plants to 2, 4, or 20 M. scutellaris adults for 7 days and allowed their progeny to develop into adults which ultimately produced densities of 0, 56, and 352 F1 adults, respectively, per low and high quality plants. No macropterous adults were produced in any treatment combination in either experiment. Several plant variables were affected by insect densities and plant quality including the mean relative growth rate, the change in leaf number, and the percentage of dead leaves on a plant. Megamelus scutellaris appeared to have a relatively high density threshold for macroptery within the range of host quality used in these studies. This may promote more brachyptery which, in turn, may increase the chances of this insect reaching damaging densities in field populations of E. crassipes.
Many planthopper species in the Delphacidae exhibit significant wing dimorphism to the extent that fully winged adults (macropters) are capable of flight while those with reduced wings (brachypters) are not (Ossiannilsson 1978; Denno et al. 1985). The wing form is determined by a developmental switch that is triggered by environmental factors including crowding, host plant quality, temperature, and photoperiod (Johno 1963; Mochida 1973; Saxena et al. 1981; Denno et al. 1985). Macroptery is usually a density-dependent phenomenon but it often increases when host quality declines (May 1975; Cook & Perfect 1985; Denno et al. 1991). The triggering density threshold for macroptery is variable among and even within species (Denno et al. 1991). Although macropters have the ability to escape suboptimal conditions, this may come at the cost of delayed reproduction, reduced fecundity and longevity, and increased risk of mortality while migrating (Denno et al. 1986; Denno & Roderick 1990; Denno et al. 1989; May 1975; Dingle 1989).
Megamelus scutellaris Berg (Hemiptera: Delphacidae) is a South American species that feeds on waterhyacinth, Eichhornia crassipes Mart. (Solms) in Peru, Brazil, Uruguay, and Argentina (Sosa et al. 2005). The insect was recently released in the U.S. as a classical biological control agent for waterhyacinth (Tipping et al. 2011). This species is multivoltine in southern Florida and exhibits the wing dimorphism of both brachypterous and macropterous wing-forms.
The objective of this study was to examine and quantify the influences of insect density and host plant quality on M. scutellaris wing dimorphism, development, and biomass as well as their effects on E. crassipes.
MATERIALS AND METHODS
All tests were conducted in 37.8 liter aquaria placed in a temperature controlled greenhouse under ambient light and photoperiods. Megamelus scutellaris nymphs were collected directly from general quarantine colonies reared in an adjacent greenhouse room. One wk old M. scutellaris adults were obtained by exposing plants to adults from quarantine colonies and collecting the appropriately aged F1 adults that emerged. Plants were grown in a temperature controlled greenhouse maintained within a range of 25–28 °C under ambient light and photoperiod with a standard fertilizer regime of 10.97 g liter-1 of 15-9-12 osmocote and 0.66 g liter-1 of 10% iron chelate.
Experiment 1. Lower Density Feeding by Nymphs
The experimental design was a 3 × 2 factorial arranged in randomized complete block design with 3 insect densities, 2 plant nutrient levels, and 5 blocks (replications). Insect densities were normalized to account for differences in the biomass of the single test plant by using a ratio of one second instar M. scutellaris to 50, 5, or 1 g of fresh weight plant biomass, which resulted in respective mean (±SE) densities of 2.0 ± 0.0, 14.5 ± 0.3, and 64.2 ± 3.2 nymphs per low quality plant, and 2.0 ± 0.0, 15.5 ± 0.6, and 74.2 ± 2.3 nymphs per high quality plant. Second instars were used because first instars suffered high levels of mortality during handling. The high quality plant treatment was created by placing standard fertilized plants in aquaria with 1549.5 mg of 15-9-12 osmocote and 92.5 mg chelated iron in 5 liters of tap water. Low quality plants were derived from plants that were previously held continuously in unfertilized tap water and placed into aquaria with the standard fertilizer regime (309.5 mg of 15-9-12 osmocote and 20.5 mg chelated iron). These fertilizer regimes resulted in leaf tissue nitrogen of 2.6 ± 0.2% and 3.7 ± 0.05% of dry weight biomass for low and high quality plants, respectively.
Eichhornia crassipes test plants were stripped of dead matter and total plant initial fresh weight biomass and leaf counts were taken before placement into aquaria. Insects were added and net tops were placed on the aquaria and secured to prevent insect escape. As nymphs developed into adults they were removed and tallied daily by gender and wing type, placed in a freezer for 7 days, air dried for 24 h, and weighed to determine mean biomass. Plant fresh weight biomass, the number of ramets, and the number of dead and live leaves were measured for each test plant after the last adults had developed and been removed from the aquaria. The third youngest leaf of each plant was removed for carbon and nitrogen analysis. All plant material was dried at 35 °C to a constant weight to determine dry weight biomass. Mean relative growth rate MRGR (mg day-1) was calculated for whole plant biomass using the formula:
where W1 and W2 are the estimated dry weight biomass at the beginning (t1) and known dry weight biomass at the end (t2) of the sampling period, and In is the natural logarithm. Dry weight biomass at t1 was calculated using an estimate of 96% moisture in all test plants.
Experiment 2. Adult and Higher Density Feeding by Nymphs
The experimental design was as above except that 7 d old adults were added to similarly sized plants at densities of 2, 4, and 40 adults (50% female) for 7 days and then removed. The fresh weight biomass of the test plants were not different from one another and insect and plant measurements were taken in the same way as Experiment 1.
Fig. 1.
Effects of increasing nymphal Megamelus scutellaris density with two levels of plant quality on: A. adult female biomass of M. scutellaris, B. adult male biomass, C. the percent change in leaf number, and D. the MRGR of Eichhornia crassipes. *,** Variable means were different within plant quality and insect density at P = 0.05 and P = 0.01, respectively. 1FW = Fresh Weight.
Statistical Analysis
Data were tested for normality and homogeneity of variance and transformed as needed prior to analysis. Mean values are presented as arithmetic means with standard errors. The influence of host quality and insect density and their interactions on insect and plant parameters were examined using two way ANOVA and regression. Variable means were separated post-hoc using t-tests. All statistical analyses were conducted using SAS v9.1 (SAS 2004).
TABLE 1.
EFFECTS OF FEEDING BY LOWER DENSITIES OF MEGAMELUS SCUTELLARIS NYMPHS ON INSECT AND PLANT VARIABLES WITH PLANT QUALITY AND INSECT DENSITY AS MAIN FACTORS.
RESULTS
Experiment 1. Lower Density Feeding by Nymphs
No macropterous adults were produced in any treatment combination. Mortality of nymphs was unaffected by plant quality but negatively affected by insect density with no interaction present (Table 1). Although the number of adults and the sex ratio were unaffected by plant quality, female biomass was greater on higher quality plants and there was a plant quality × insect density interaction (Table 1). This interaction resulted primarily from a difference in magnitude at the low density treatment, a result which should be interpreted cautiously because so few females developed in this treatment combination (Fig. 1A). Despite this, there was a trend of increasing biomass as insect density increased on the low quality plants (Y = 0.81 + 0.05X, r2 = 0.62, P = 0.02) (Fig. 1A). Across all treatments, females weighed more than males (1.8 ±0.1 mg vs. 0.6 ±0.04 mg, respectively) (t38 = 28.7, P < 0.0001). Male biomass was affected only by insect density but, like females, also increased with increasing density on both low (Y = 0.47 + 0.007X, r2 = 0.55, P = 0.03) and high (Y = 0.39 + 0.04X, r2 = 49, P = 0.05) quality plants (Table 1) (Fig. 1B). Mean development time to the adult stage was unaffected by plant quality or insect density regardless of gender (Table 1). In general, male nymphs developed to adulthood about a day sooner than did females ([13.8 ± 0.2 d vs. 14.9 ± 0.5 d] t38 = 3.6, P = 0.0009).
TABLE 2.
EFFECTS OF FEEDING BY ADULTS AND HIGHER DENSITIES OF NYMPHS OF MEGAMELUS SCUTELLARIS ON INSECT AND PLANT VARIABLES WITH PLANT QUALITY AND INSECT DENSITY AS MAIN FACTORS.
All plant variables were predictably influenced by plant quality but none directly by insect density. There was a plant quality × insect density interaction for the percent change in the number of leaves whereby leaf numbers decreased with increasing insect density for low quality plants but increased for high quality plants (Table 1) (Fig. 1C). Mean RGR was greater with high quality plants than low quality plants although there also was a trend of declining biomass with increasing insect density in the low quality plant treatment (Y = 0.01 - 1.001X, r2 = 0.43, P = 0.07) (Fig. 1D). No ramets were produced by low quality plants while those produced by high quality plants (mean ± SE = 1.4 ± 0.3 per plant) were unaffected by insect density (Table 1).
Fig. 2.
Effects of increasing initial adult Megamelus scutellaris density with two levels of plant quality on: A. adult male biomass of M. scutellaris, B. the MRGR of E. crassipes, C. the percent change in leaf number, and D. the percentage of dead leaves per plant. *,** Variable means were different within plant quality and insect density at P = 0.05 and P = 0.01, respectively.
Experiment 2. Adult and Higher Density Feeding by Nymphs
As in the first experiment, no macropterous adults were produced in any combination of plant quality or insect density. No F1 adults were produced in the lowest insect density treatment for unknown reason(s). The total number Of F1 adults produced was unaffected by plant quality but directly affected by insect density (Table 2). Sex ratio, mean female biomass, and mean male biomass were not influenced directly by either plant quality or insect density, although male biomass was influenced by a host quality × insect density interaction (Table 2). This interaction resulted from a change in rank whereby males were larger on low quality plants at low densities while the reverse was true on high quality plants at high densities (Fig. 2A). As in the previous experiment, females were larger than males (1.8 ± 0.1 mg vs. 0.8 ± 0.04 mg, respectively; t24 = 7.7, P < 0.0001). Several plant variables were affected not only by plant quality, but also insect densities, including MRGR (Fig. 2B), the change in leaf number (Fig. 2C), and the percentage of dead leaves on a plant (Fig. 2D) (Table 2). Although the relationships were not strong, MRGR decreased with increasing insect density with both low (Y = 0.009 - 0.0002X, r2= 0.38, P = 0.10) and high (Y = 0.01 - 0.0004X, r2 = 0.54, P = 0.15) quality plants (Fig. 2B). The percentage of dead leaves per plant increased with increasing insect density with low quality plants (Y = -20.1 + 4.4X, r2 = 0.78, P = 0.003) (Fig. 2D). Ramets were produced from low quality plants, unlike the first experiment, but more were produced from high quality plants regardless of insect density (Table 2).
DISCUSSION
Although attack by M. scutellaris did reduce E. crassipes biomass in previous lab studies (Tipping et al. 2011), this insect will probably require high densities in order to negatively suppress the plant in the field. In the lab, plants typically do not exhibit stress until insect densities exceed at least 100 per plant (P.W.T., unpublished data). It is important to understand the influence of density dependent and independent factors on macroptery because of their potential to dilute the impact of this insect on plant populations. Given the numbers of F1 adults produced in Experiment 2, it appears that M. scutellaris has a relatively high threshold for population density within the range of host quality used in these studies (Fig. 3). Assuming 70 plants per m2 in monotypic stands in the field in Florida and, based on the number of F1 adults produced on plants of lower quality in the high insect density treatment in Experiment 2, the population density of M. scutellaris could approach at least 16 K per m2 without the formation of macropters, a density that would likely prove damaging to E. crassipes.
There was evidence that insect biomass was positively dependent on density (Fig. 1A), indicating that aggregations of M. scutellaris may provide an increased level of benefit. Dixon and Wratten (1971) demonstrated that individual Aphis fabae Scopoli (Hemiptera: Aphididae) were larger when reared near or within aggregations of conspecifics than when reared alone. This phenomenon has been documented in a variety of insects with different strategies, but only on whole plants indicating that these aggregations somehow modify the plant by creating nutrient sinks or by overwhelming any existent plant defenses (Denno & Benrey 1997; Awmack & Leather 2002).
The lower threshold densities for macroptery seen in this study may promote outbreak levels of the insect which, in turn, may exert a stronger negative influence on waterhyacinth populations since outbreak populations usually consist primarily of brachypters (Zimmerman 1948). Although feeding by M. scutellaris after a single generation reduced MRGR of low quality plants in the second experiment by 24% and 37% with low and high insect densities, respectively, MRGR was still positive. However, MRGR was negative in another study when plants were exposed to two consecutive generations of feeding and produced 67% less biomass and 73% fewer leaves compared to the control (Tipping et al. 2011). Future studies will focus on quantifying the relationship between insect population demography, including wing-forms, with landscape level densities in natural field populations of the plant.
ACKNOWLEDGMENTS
We thank Eileen Pokorny for her assistance in laboratory studies. We also acknowledge the influence of the late Robert F. Denno whose creativity and encyclopedic knowledge of the population dynamics of the Delphacidae generated stimulating discussions on this topic.
