Most sham feeding studies show that about three sham feeding tests are required for intake to reach maximum. One study, however, using a dilute solution, reported maximum sham intake in the first sham feeding test, suggesting that the progressive rise in sham intake may be concentration dependent. We tested this hypothesis with six groups of rats given five sham feeding tests each with one of six concentrations of sweetened condensed milk (0.5:1, 1:1, 2:1, 4:1, 8:1, 16:1, water-to-milk dilutions). It took three sham tests for intake to reach maximum with the three most concentrated solutions, but only one with the three weakest. Thus the intake of concentrated solutions of milk is limited by two negative feedback signals, one derived from the accumulation of fluid in the gastrointestinal tract, the other from a labile signal that loses its effectiveness with sham feeding experience. In contrast, the intake of weak concentrations is limited only by the nonlabile negative feedback signal because the labile signal is missing.