. S U M .M A R Y .Although grasses infected by fungal endoptiytes often exhibit increased growth relative to uninfected plants, the results of infection may be manifested as decreased growth in plants growing under stressful conditions. A preliminary experiment compared perennial ryegrass {Lolium perenne L.) infected by Acremoiiium lolii Latch, Christensen & Samuels with uninfected plants grown at three soil moistures and three nutrient concentrations. Total biomass, tiller number, and relative tittering rate were significantty affected by nutrient concentration and moisture. Although major differences between infected and uninfected plants were not apparent, relative growth rates of infected plants were significantly greater than uninfected plants in three of nine treatment combinations.Interactions between infection and nutrient limitation were examined in .seedlings and adults of perennial rj'egrass and tall fescue {Festuca arundinacea Schreb., infected by A, coenophialiim Morgan-Jones and Gams). A significant nutrient X infection interaction for tall fescue seedlings indicated that, compared to uninfected seedlings, infected seedlings had greater biomass at the high nutrient level but a significantty tower biomass at the tow nutrient tevel, suggesting a metabolic cost to the host due to competition with the fungus for nutrients or photosynthate. Infected perennial ryegrass seedlings had significantty more biomass at high and intermediate nutrient levels, but not at low nutrient levels. Although infection did not significantly affect total biomass in perennial ryegrass adutts, in tatt fescue adutts the benefits of endophyte infection to tiost ptants (recorded as greater biomass) became increasingty greater with increasing nutrient avaitability.

Interactions among plants may be influenced by pests or parasites that differentially affect one competitor. The purpose of this study was to determine the effects of fungal parasitism and insect herbivory, alone and in concert, on plant competitive interactions. The effects of fungal endophyte (Acremonium spp.) infection and fall armyworm (Spodoptera frugiperda) herbivory on competitive interactions in one-and two-species mixtures of the grasses tall fescue (Festuca arundinacea), red fescue (F. rubra), and perennial ryegrass (Lolium perenne) were examined in greenhouse experiments. In general, herbivory reduced plant biomass whereas endophyte infection increased plant biomass. Endophyteinfected (E+) plants were less damaged by herbivory than uninfected (E-) plants of the same species. Studies on fall armyworm larval feeding and choice tests with the five grass species were generally consistent with the outcome of the competition experiments; E+ plants were less nutritious and less preferred than E-plants of the same species. There were significant interactions among factors so that the outcome of competition depended on the species identities and the presence or absence of endophytes and herbivores. In competition with Kentucky bluegrass (Poa pratensis), E+ and E-perennial ryegrass produced similar biomass in the absence of herbivory, but E+ perennial ryegrass had nearly twice the biomass ofE-plants when herbivores were present. E+ and E-tall fescue were poor competitors with orchard grass (Dactylis glomerata) when herbivores were absent, but E+ tall fescue was a better competitor thanE-plants and orchard grass when herbivores were present. This study indicates that competitive hierarchies among grasses are altered by interactions with insect herbivores and fungal endophytes, which have typically been ignored in past studies.

No abstract

SUMMARYEnhanced growth rates of several grass species infected by fungal endophytes are known, hut the underlying changes in plant physiology are not. Carbon exchange rates (CER) and leaf conductances {g) of 13 genotypes of tall fescue {Festuca arundinacea Schreb. var. KY 31) infected by the fungal endophyte Acremonimn coenophialum Morgan-Jones and Gams were measured at ambient conditions. Endophyte-free ramets of the same genotypes were also measured. Correlations were calculated hetween environmental conditions at the time of measurement, and physiological responses. The only differential response of infected and uninfected ramets was to temperature. At low leaf temperatures no difference was seen between infected and uninfected plants. However, at leaf temperatures above 35 °C infected tall fescue plants photosynthesized at a significantly greater rate (20-25 %) than uninfected plants. This resulted from a decrease in the CER of uninfected plants, not an increase in the rate of infected plants, at high temperature. There were also significant infection x genotype interactions, indicating that the response to infection was specific to a given genotype. These results indicate that physiological responses of host plants to fungal endophyte infection depend both on the physical environment and the genotype of the plants.

Increasing atmospheric carbon dioxide (CO) concentration is expected to increase plant productivity and alter plant/plant interactions, but little is known about its effects on symbiotic interactions with microorganisms. Interactions between perennial ryegrass, Lolium perenne (a C3 plant), and purpletop grass, Tridens flavus (a C4 plant), and their clavicipitaceous fungal endophytes (Acremonium lolii and Balansia epichloe, respectively) were investigated by growing the grasses under 350 and 650 μl l 1 CO at two nutrient levels. Infected and uninfected perennial ryegrass responded with increased growth to both CO enrichment and nutrient addition. Biomass and leaf area of infected and uninfected plants responded similarly to CO enrichment. When growth analysis parameters were calculated, there were significant increases in relative growth rate and net assimilation rate of infected plants compared to uninfected plants, although the differences remained constant across CO and nutrient treatments. Growth of purpletop grass did not increase with CO enrichment or nutrient addition and there were no significant differences between infected and uninfected plants. CO enrichment did not alter the interactions between these two host grasses and their endophytic-fungal symbionts.