An in vitro feeding method using chicken-skin membranes and human blood was compared with an established in vivo method using anesthetized hamsters for blood-feeding mass-reared phlebotomine sand flies. Parameters measured were percentage of sand flies taking blood meals, number of eggs laid per female, and percentage of eggs that hatched. Females from a long established (>20 yr) colony of Phlebotomus papatasi (Scopoli) from Israel landed on and started feeding sooner on the hamster than on the membrane. However, when sand flies were allowed access to the membrane feeder for the same length of time as the anesthetized hamster, the feeding percentages were not significantly different and were usually better on the membrane feeder if flies were allowed access for a longer time. Egg production and percent hatch between the two feeding methods were not statistically different. Based on these results, we conclude that the chicken-skin membrane feeding method is a viable alternative to the use of live animals for feeding large numbers of P. papatasi.