Resource quality (plant nitrogen) and resource quantity (plant density) have often been argued to be among the most important factors influencing herbivore densities. A difficulty inherent in the studies that manipulate resource quality, by changing nutrient levels, is that resource quantity can be influenced simultaneously, i.e. fertilized plants grow more. In this study we disentangled the potentially confounding effects of plant quality and quantity on herbivore trophic dynamics by separately manipulating nutrients and plant density, while simultaneously reducing pressure from natural enemies (parasitoids) in a fully factorial design. Plant quality of the sea oxeye daisy, Borrichia frutescens, a common coastal species in Florida, was manipulated by adding nitrogen fertilizer to increase and sugar to decrease available nitrogen. Plant density was manipulated by pulling by hand 25 or 50% of Borrichia stems on each plot. Because our main focal herbivore was a gall making fly, Asphondylia borrichiae, which attacks only the apical meristems of plants, manipulating plant nitrogen levels was a convenient and reliable way to change plant quality without impacting quantity because fertilizer and sugar altered plant nitrogen content but not plant density. Our other focal herbivore was a sap-sucker, Pissonotus quadripustulatus, which taps the main veins of leaves. Parasitism of both herbivores was reduced via yellow sticky traps that caught hymenopteran parasitoids. Plant quality significantly affected the per stem density of both herbivores, with fertilization increasing, and sugar decreasing the densities of the two species but stem density manipulations had no significant effects. Parasitoid removal significantly increased the densities of both herbivores. Top-down manipulations resulted in a trophic cascade, as the density of Borrichia stems decreased significantly on parasitoid removal plots. This is because reduced parasitism increases gall density and galls can kill plant stems. In this system, plant quality and natural enemies impact per stem herbivore population densities but plant density does not.

Increased levels of atmospheric carbon dioxide (CO 2 ) are likely to affect the trophic relationships that exist between plants, their herbivores and the herbivores' natural enemies. This study takes advantage of an open-top CO 2 fertilization experiment in a Florida scrub oak community at Kennedy Space Center, Florida, consisting of eight chambers supplied with ambient CO 2 (360 ppm) and eight chambers supplied with elevated CO 2 (710 ppm). We examined the effects of elevated CO 2 on herbivore densities and levels of leaf consumption, rates of herbivore attack by natural enemies and effects on leaf abscission. Cumulative levels of herbivores and herbivore damage were significantly lower in elevated CO 2 than in ambient CO 2 . This may be because leaf nitrogen levels are lower in elevated CO 2 . More herbivores die of host plant-induced death in elevated CO 2 than in ambient CO 2 . Attack rates of herbivores by parasitoids are also higher in elevated CO 2 , possibly because herbivores need to feed for a longer time in order to accrue sufficient nitrogen (N), thus exposing themselves longer to natural enemies. Insect herbivores cause an increase in abscission rates of leaves throughout the year. Because of the lower insect density in elevated CO 2 , we thought, abscission rates would be lower in these chambers. However, abscission rates were significantly higher in elevated CO 2 . Thus, the direct effects of elevated CO 2 on abscission are greater than the indirect effects on abscission mediated via lower insect densities. A consequence of increased leaf abscission in elevated CO 2 is that nutrient deposition rates to the soil surface are accelerated.

The unabated increase in global atmospheric CO(2) is expected to induce physiological changes in plants, including reduced foliar nitrogen, which are likely to affect herbivore densities. This study employs a field-based CO(2 )enrichment experiment at Kennedy Space Center, Florida, to examine plant-herbivore (insect) interactions inside eight open-topped chambers with elevated CO(2) (710 ppm) and eight control chambers with ambient CO(2). In elevated CO(2) we found decreased herbivore densities per 100 leaves, especially of leaf miners, across all five plant species we examined: the oak trees Quercus myrtifolia, Q. geminata, and Q. chapmanii, the nitrogen-fixing vine Galactia elliottii and the shrub Vaccinium myrsinites. Both direct and indirect effects of lowered plant nitrogen may influence this decrease in herbivore densities. Direct effects of lowered nitrogen resulted in increased host-plant related death and an increase in compensatory feeding: per capita herbivore leaf consumption in elevated CO(2) was higher than in ambient CO(2). Indirectly, compensatory feeding may have prolonged herbivore development and increased exposure to natural enemies. For all leaf miners we examined, mortality from natural enemies increased in elevated CO(2). These increases in host-plant induced mortality and in attack rates by natural enemies decreased leaf miner survivorship, causing a reduction in leaf miner density per 100 leaves. Despite increased leaf production in elevated CO(2) from the carbon fertilization effect, absolute herbivore abundance per chamber was also reduced in elevated CO(2). Because insects cause premature leaf abscission, we also thought that leaf abscission would be decreased in elevated CO(2). However, for all plant species, leaf abscission was increased in elevated CO(2), suggesting a direct effect of CO(2) on leaf abscission that outweighs the indirect effects of reduced insect densities on leaf abscission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecology.Abstract. Few studies have attempted to determine experimentally the relative importance of direct and indirect effects of host plants on herbivorous insects in the field. This study identifies important direct and indirect effects of a coastal plant on its most common insect herbivore and assesses the relative importance of those effects. The direct effects of increased interstitial soil salinity and nitrogen on the abundance of Pissonotus quadripustulatus (Homoptera: Delphacidae), which feeds on Borrichia frutescens (Asteraceae), are reported. Also reported are the indirect effects of these treatments on parasitism of P. quadripustulatus eggs laid in plant stems by the fairyfly parasitoid Anagrus sp. nr. armatus (Hymenoptera: Mymaridae). Soil salinity was experimentally elevated by the addition of salt pellets, and plant foliar nitrogen was increased by the addition of fertilizer.Both salt and fertilizer increased the abundance of P. quadripustulatus. There was a significant salt X fertilizer interaction, suggesting that salt stress may be more important when plants are more nitrogen limited. Salt, by increasing the frequency of tough B. frutescens stems, decreased the rate at which Anagrus parasitized P. quadripustulatus eggs. Fertilizer, by increasing the frequency of B. frutescens stems that were softer and easier to penetrate, increased the rate of parasitism. Changes in parasitism, however, did not affect P. quadripustulatus density. Tests for density dependence showed that the results reported here were attributable to application of the treatments and not to spatial density dependence. This study suggests that, in a stressful salt-marsh system, the direct effects of plant quality on herbivores are more important than indirect effects of plant morphology mediated by natural enemies.Key words: abiotic factors; Borrichia frutescens; community structure; direct vs. indirect effects; egg parasitism; herbivores, affected by host-plant variation; host-plant nitrogen; Pissonotus quadripustulatus; plant morphology, affected by salinity; plant quality, effects of salinity and nitrogen; saltmarsh community; soil salinity. of its most common herbivorous insect and assess their relative importance. Using the coastal salt-marsh plant Borrichia frutescens (Asteraceae), the direct effects of plant quality on the sap-sucking insect Pissonotus quadripustulatus (Homoptera: Delphacidae) were studied. At the same time, we investigated the indirect effects of host-plant morphology on these herbivores via plant-mediated changes in parasitism rates of herbivore eggs by the parasitoid Ana...

In recent years, ecologists have begun to develop a more fine-grained and integrative approach to examining the importance of top-down and bottom-up effects on herbivore populations by investigating how changes in abiotic heterogeneity affect the relative roles of these forces. We performed two factorial field experiments to determine how increasing salinity and nutrient supply affected the relative strengths of top-down and bottom-up forces among the gall-making midge, Asphondylia borrichiae, its host plant Borrichia frutescens, and a suite of parasitoids. Salinity was increased by the addition of salt pellets, and nutrient supply was increased by the addition of fertilizer. In both experiments, parasitism pressure was decreased by trapping hymenopteran parasitoids with yellow sticky traps.In both experiments, bottom-up manipulations had significant effects on gall density. Elevated salinity levels decreased the number of galls per 200 Borrichia stems, and fertilization increased the number of galls. In the salt-stress experiment, increased salinity decreased gall parasitism, and in the fertilization experiment, fertilizer increased parasitism of galls, but the direct effects of these treatments on gall density outweighed the indirect effects mediated by parasitoids. Parasitoid-removal treatments resulted in only small (and not statistically significant) changes in gall parasitism in the salt-stress experiment, and there was no significant effect on gall density. In the fertilization experiment, parasitoidremoval treatments reduced parasitism of galls, and this resulted in an increase in gall density over time. Thus, bottom-up effects were strong and consistent in both experiments, but top-down effects appeared weaker and inconsistent. Stem morphology, which was altered by salt and fertilizer, also affected gall density and parasitism, and there was some evidence that the strength of top-down forces on gall density varied with stem type. The results of this study show that, in this system, top-down effects of parasitism are dependent upon the bottom-up effects of host plant quality and morphology resulting from abiotic heterogeneity. As such, the relative importance of top-down effects can change depending upon the relative nutrient supply or salinity of the environment.