The rate of acquisition of Echinococcus granulosus and Taenia hydatigena was examined in dogs treated with praziquantel in northwestern Turkana District of Kenya. An attempt was also made to assess the effect of drought on the prevalence of E. granulosus. The infection pressure to dogs with the cestodes was shown to be high, and the prevalence returned to pretreatment levels within six months. The unreliability of arecoline purging in prevalence studies of dog cestodes is shown by mathematical contrast to the results of autopsy. Arecoline purge may underestimate the real prevalence by 10-fold. Dosing intervals in hydatid control programmes utilizing mass dog-purging to estimate re-infection rates should therefore take this factor into consideration. It is proposed that a mass dog-dosing schedule in Turkana, using praziquantel, should take into consideration the short prepatent period of E. granulosus, the financial constraints, and the slaughter practices in the area. During the prolonged droughts which occur approximately every 10 years, the programme will be adjusted accordingly to cater for the initial increase in infection intensity to dogs.