DOES LIGHT-INTENSITY MODIFY THE EFFECT MAYFLY GRAZERS HAVE ON PERIPHYTON

1. A factorial experiment was conducted in artificial outdoor streams to quantify the effects of irradiance (two levels) and two mayfly graz ers (four densities of each) on periphytic community structure. The ma yflies were Ecdyonurus venosus (Heptageniidae), a grazer using brushin g mouthparts, and Baetis spp. (Baetidae) a grazer which uses mandibles and maxilla to scrape and gather periphyton. The experiment ran for 1 6 days. 2. Grazer densities in channels approximated those existing in a shoreline habitat in the River Sihl, Switzerland. Light treatments were natural (daily mean = 810 mu mol m(-2) s(-1)) and shaded (daily m ean = 286 mu mol m(-2) s(-1)). 3. Higher irradiance increased total al gal abundance by a factor of 4. Algae most affected were prostrate/mot ile and erect diatoms, filamentous chlorophytes and Hydrurus foetidus. 4. Both species of mayfly reduced periphytic and algal biomass. Mayfl y-mayfly interactions, however, were associated with statistical incre ases in algal biovolume and chlorophyll-a content, indicating that the two grazers may have interfered with one another as their densities i ncreased. The mayfly-mayfly interaction did not influence periphytic a sh-free dry mass (AFDM). Light modified the influence of Ecdyonurus su ch that this mayfly produced greater reductions in algal biovolume und er high irradiance. 5. Despite efforts to exclude other grazers, chiro nomids colonized experimental channels. Chironomid biomass was approxi mately eight times less than mayflies across treatments and was positi vely correlated with all measures of periphytic abundance, suggesting that these grazers were responding to periphyton rather than controlli ng it. Chironomids were also associated with an increase in the abunda nce of diatoms having a prostrate/motile physiognomy. The only physiog nomy to show a negative relationship with chironomid biomass was the t hallus type, a form which comprised less than 1% of the algal biovolum e across channels. 6. Ecdyonurus and Baetis had distinct influences on algal physiognomy. Ecdyonurus, for example, reduced adnate, stalked a nd Achnanthes-type physiognomies, but was associated with a significan t increase in the abundance of filamentous chlorophytes (primarily Ulo thrix sp.). Baetis reduced erect, Achnanthes-type and thallus physiogn omies. Neither mayfly influenced the abundance of prostrate/motile dia toms; a physiognomy that comprised 21% of the algae in channels. 7. Li ght and mayfly interactions affected algal community structure. The in teraction of Ecdyonurus with light had a negative effect on erect diat oms, filamentous chlorophytes and the thallus physiognomy, but a posit ive effect on stalked and Achnanthes-type physiognomies. Baetis intera cting with light had a positive effect on adnate diatoms. 8. Although both mayfly taxa influenced periphytic community structure, physiognom y was not a good predictor of algal susceptibility to grazing. The typ e of substratum to which an alga is attached (detritus or algal filame nts vs hard surfaces) and location within the periphytic matrix may be better indicators of vulnerability to grazing than physiognomy.