Technical Abstract: Field burning has traditionally been used to stimulate Kentucky bluegrass (Poa pratensis L.) seed production, but air quality issues are making this practice untenable. A Kentucky bluegrass core collection (20 entries), additional selected accessions (16 entries), and cultivar checks (9 entries) were used to determine developmental and production responses under burned, mechanically removed, and residue retained management systems. Compared to burned treatments, yield was reduced 27% when residue was mechanically removed from plots, and 63% when residue was retained. Higher yield was promoted by early heading followed by a long heading to anthesis period, a relatively short anthesis to harvest period, and an early harvest date (maturity). Although both seeds panicle-1 and fertile panicles m-2 were positively correlated with yield, lower yield with non-thermal residue management was most closely associated with panicles m-2. For six of the 15 highest yielding entries, the difference between yield in the burned and residue removed treatments was not significant, showing the dependance of genotype in the yield response under different residue management systems. Turf quality was negatively correlated with yield (r= -0.48**, n=44) and seeds panicle-1 (r= -0.55**, n=44). However, panicles m-2 were not significantly correlated with turf quality, so indirect selection for yield through genotypes with high panicles m-2 in the absence of high seeds panicle-1 would be expected to have minimal impact on turf quality. Sufficient variation for seed production appears available to encourage development of germplasm for non-thermal management systems.