We evaluated 3 aspects of release strategies used to augment green lacewings (Chrysoperla spp.): the delivery system, the release rate and timing, and the lacewing developmental stage released. Tests were conducted in vineyards with either the common green lacewing, Chrysoperla carnea (Stephens); the Comacnhe lacewing, Chrysoperla comanche Banks; or Chrysoperla rufilabris (Burmeister). Lacewings were released to suppress 2 leaf-hopper pests, Erythroneura variabilis Beamer and the western grape leafhopper, Erythroneura elegantula Osborn. Two commercial delivery systems were compared. In the 1st delivery system, a mixture of lacewing eggs and corn grit was placed in paper cups, which were distributed to every 5th vine in every other row This delivery system resulted in poor egg hatch and larval dispersal. Egg hatch was low (approximate to 60%) in the paper cups, compared with egg hatch when lacewings were reared in individual cells (approximate to 91%). The poor egg hatch is attributed primarily to cannibalism. Larval dispersal from paper cups with corn grit was 25% lower than that from cups without corn grit, resulting in incomplete distribution of lacewings throughout the vineyard. In a 2nd delivery system, lacewing eggs, combined with corn grit, were dropped onto the vines from a moving flatbed trailer. Egg hatch was approximate to 62%. Eggs were delivered to each vine and vine row although delivery of eggs was uneven. More eggs were dropped at the beginning (approximate to 11 eggs per vine) of each release batch than at the end (approximate to 5 eggs per vine). Using a noncommercial delivery system, release rates between 6,175 and 1,235,000 eggs or larvae per hectare were tested. No correlation between release rate and prey density was found. Releases timed to approximate to 50-70% leafhopper egg hatch had a greater effect on leafhopper densities than releases timed to peak leafhopper nymph densities. In the 3rd experiment, we tested the effectiveness of egg versus larval releases. In egg release plots, there was approximate to 70% egg mortality and leafhopper densities were not significantly different from no-release plots. In larval release plots, approximate to 50% of the larvae survived until the 3rd instar and there was a significant reduction in leafhopper densities.