Methods for Lake Michigan zooplankton Zooplankton were collected with a 0.5 or 1 meter diameter, 153 or 64 µm mesh net. The net was vertically towed through the water column at a speed of 0.5 m s-1 from 1 to 2 meters above the bottom to the surface. For partial water column tows, to collect separately the epilimnion, metalimnion and hypolimnion, a messenger actuated choke-off vertical opening/closing net (1m diameter, 2m-long, 153 or 64 µm mesh) was used to obtain the zooplankton samples. The net was washed thoroughly, but gently and the contents were carefully transferred to a sample bottle, narcotized with Alka-Seltzer, and preserved with the addition of sugar formaldehyde to form a 2% solution (Haney & Hall, 1973). Preparation for counting and identifying zooplankton required measurement of the sample volume, gentle mixing of the sample, and removal of an aliquot with a Hensen-Stempel pipette. The aliquot was held in a 100 µm meshed bottom cup and rinsed with tap water to remove the preservative and finally it was poured into a circular counting dish. A minimum 550 zooplankton were identified for each sample. To count large predatory cladocerans, such as Bythotrephes, the whole sample was rinsed through a 600 µm mesh sieve, and all individuals were counted. Taxonomic groups were identified and categorized in the following manner. Copepodites were identified to genus and adult copepods were identified to species. The cladocerans, both herbivorous and predatory, were identified to species. To determine zooplankton biomasses, length measurements were made on a subsample of taxa (10 adult copepods and 25 copepodites or cladocerans) that were over 10% of the total density using Image Pro Plus, image analysis software (Media Cybernetics, Silver Spring, MD). For zooplankton taxa that comprised less than 10% of the total density, a default weight from the literature was used to determine biomass for all taxa except Bythotrephes (Hawkins and Evans 1979).