To optimize yields of long-chain fructans extracted from herbage, fructan concentrations and chromatographic profiles were compared in four cool-season grasses extracted by different methods. In a preliminary study, extraction temperatures, pH prior to extract concentration, numbers of extractions, and tissue coarseness were varied to evaluate their effects on chromatographic profiles and quantities of total long-chain fructan (degree of polymerization (DP) of ∼7 or higher). Fructans were separated by anion-exchange high-performance liquid chromatography (HPLC) with pulsed amperometric detection (PAD). The identity of the putative fructan peaks was confirmed by acid hydrolysis of fructans to fructose. Tall fescue and bluegrass fructans were below the detection limits used in this study. Pooled timothy and cocksfoot data indicated that two or three extractions at the same temperature yielded similar amounts of fructans and that adjusting pH from 6 to 7 did not affect recovery. Replicated extractions of cocksfoot by four methods chosen from the preliminary study demonstrated that the least fructan was extracted from chopped cocksfoot at ambient temperature (P = 0·019). One C₁₈ solid-phase extraction cleanup step sufficed for analysis. The extraction method recommended is two boiling water extractions of either chopped or ground material. However, ground material extracted at ambient temperature is comparable in total fructan concentration and maximum DP.