Knowledge of the presence or absence of lice in a flock of sheep enables wool growers to make informed decisions as to the need for insecticidal treatments. However, with inapparent infestations, traditional methods of detection are not sufficiently sensitive and, as a consequence, flocks may be left untreated. Conversely, the routine application of insecticide to sheep with no sign of infestation is an unnecessary cost. The sensitivity of 3 procedures for detecting lice was evaluated in 68 mobs of sheep from 50 farms. In 24 mobs of sheep known to be lightly infested, lice were detected in 17% (71%) [corrected] by either parting the fleece of 10 sheep or by the lamp test in which 8g samples of shorn wool from 30 randomly selected fleeces were placed under lamps for 10 min to repel the lice. Twenty of 23 mobs (87%) were found to be infested by the table locks test in which a 30 g sample of locks wool was dissolved in 10% sodium hydroxide and the filtered residue examined with x 40 magnification. A screening test, in which either fleeces on 5 sheep were examined by fleece parting or lice were repelled from 30 shorn fleeces for 5 minutes, detected about 60% of lightly infested mobs. When this was followed by the table locks test 91% of lightly infested mobs were detected. Conducting any one of the tests on more than one mob, and in large mobs testing more frequently, increases the sensitivity of detection of lice within the whole flock.