Selective predation by a sculpin and a stonefly on two chironomids in laboratory feeding trials Anne E. Hershey & Stanley I. Dodson Dept. of Zool., Birge Hall, University of Wisconsin, Madison, WI 53706 Keywords: chironomids, microhabitat, tube-dwelling behavior, predator selectivity, sculpins, stoneflies Abstract Sculpin and stonefly predators fed selectively on the larvae of the chironomids Paratendipes over Cricotopus in laboratory stream microcosms. In these experiments, Cricotopus were usually tube-dwelling, whereas Paratendipes were usually free-living. Paratendipes were also bright red, which may have influenced selectivity by visual feeding sculpin, but tactile feeding stoneflies were most likely influenced only by the difference in tube-dwelling behavior of the two prey types. Both chironomids were abundant in the field, but exhibited discrete microhabitat distributions. Field collected sculpin ate mostly Cricotopus, probably because Cricotopus occurred in a more accessible microhabitat. Introduction In aquatic ecosystems, invertebrate and verte- brate predators may feed selectively on benthic in- vertebrate prey. In lentic systems, some fish select larger insect taxa, thereby affecting the structure of the invertebrate community (Crowder & Cooper, 1982). But sculpin appear to feed selectively on small rather than large chironomids, possibly due to size-specific behavioral differences of prey (Hershey, 1985). In a field enclosure experiment, bluegill had greatest effect on a large, free-living predatory chironomid (Gilinsky, 1984). Bluegill fed on smaller Chironomus than predicted by a forag- ing model because larger Chironomus burrow deeper in sediments (Werner et al., 1983). In streams, as in lentic systems, predator-prey interactions may be determined by prey availability and prey behavior. Some mayflies appear to have behavioral defenses against predatory stoneflies (Peckarsky, 1980), and stoneflies may have a sim- ilar defense against trout (Otto & Sjostrom, 1983). Selectivity by stoneflies for mayflies and chirono- mids has been attributed to a greater availability of these prey types compared to other taxa (Sieg- fried & Knight, 1976; Allan, 1982). Stream chi- ronomids have been reported in the diets of most stream predators (Van Vliet, 1964; Davis & Warren, 1965; Brocksen et al., 1968; Stewart et al., 1973; Hildrew & Townsend, 1976; Devonport & Winterbourne, 1976; Siegfried & Knight, 1976; Craig & Wells, 1976; Fuller & Stewart, 1977, 1979; Pidgeon, 198 1 ; Allan, 1982; Walde & Davies, 1984). Chironomids exhibit a variety of behaviors, partic- ularly with respect to the type of tubes they con- struct (Wiley, 1978). In a stream enclosure experi- ment, density of the chironomid, Thienemanniella, was depressed by predatory stoneflies, but other chironomids were unaffected (Walde & Davies, 1984). This may have been due to Thienemanniel- las's free-living, as opposed to tube-building behav- ior (Walde & Davies, 1984). Thus, evidence from field observations and field enclosure experiments suggested to us that chi- ronomid behavior, particularly chironomid tube- dwelling habits, might be important in determining predator selectivity on chironomids. We tested this hypothesis in laboratory stream microcosms using Hydrobiologia 124,269-273 (1985). O Dr W. Junk Publishers, Dordrecht. Printed in the Netherlands.