Experimental Analysis of the Grazing Interaction Between a Mayfly and Stream Algae

The interaction between the grazing mayfly Ameletus validus and periphyton in a small, northern California stream was examined by manipulating the density of the mayfly in flow—through plexiglass channels. Containing natural cobble substrate and located in situ, the channels established an initial gradient of A. validus at 0, 0.5, 1, and 4 times the average density of the mayfly in Barnwell Creek. After 23 d, A. validus significantly depressed periphyton standing crop: ash—free dry mass (AFDM) at the 0, 0.5, 1, and 4 N grazer densities was 5.067 ± 1.389 (se), 1.829 ± 0.173, 1.741 ± 0.325, and 1.009 ± 0.199 g/m2 (ANOVA: P < .01). The mayfly also influenced two structural attributes of the periphyton, increasing the amount of chlorophyll a per unit biomass and decreasing the relative contribution of the loose, upper layer to total periphyton biomass. Principal component analysis of algal relative abundances contrasted the effect of grazing on two groups of diatoms. A group of species found primarily in the loose layer of periphyton (Nitzschia spp., Surirella spiralis, Cymatopleura elliptica, and Navicula cryptocephala) was disproportionately reduced in abundance, while an adnate group (Gomphonema clevei, Achnanthes minutissima, Synedra ulna, Rhoicosphenia curvata, and an undescribed species of Epithemia) increased its relative abundance with increasing grazing pressure. The decline in relative abundance of the loose layer diatoms did not appear to result from selective consumption by A. validus, but may have been mediated by a reduction of inorganic sediment in the periphyton by A. validus. Inorganic sediment was highly correlated with the relative abundances of the loose layer group of diatoms, a group of species that are adapted for locomotion on sediment substrates. A. validus growth in the experimental channels was strongly density dependent. Growth in length over 23 d for the 0.5, 1, and 4 N treatments was 2.24 ± 0.17, 1.80 ± 0.23, and 1.15 ± 0.25 mm (ANOVA: P < .01). The significantly greater growth of A. validus at subnormal densities in the experimental channels suggested that the A. validus population in Barnwell Creek was food—limited.