The objective of this study was to determine the empirical efficacy of imidacloprid 10 %/flumethrin 4.5 % (Seresto®) collars in preventing long-term transmission of Ehrlichia canis by infected Rhipicephalus sanguineus ticks to dogs. The study was a parallel group design, single centre, randomised, non-blinded, controlled, long-term efficacy study. The treatment group of 8 dogs was fitted with Seresto® collars, the untreated control group of 8 dogs received no collars. Ehrlichia canis-infected ticks were released into the dogs sleeping quarters at 14-day intervals up to Day +378. Control group dogs infected with E. canis were continuously replaced to keep the control sample size constant, and a total of 39 control dogs were required. The final clinical examination and blood sampling occurred on Day +420. The primary assessment criterion was the number of dogs infected with E. canis, as confirmed by IFA and PCR, and the secondary criterion was the acaricidal efficacy based on tick counts. All scheduled blood samples taken were subject to analyses for both PCR and IFA, but only positive cases are discussed. Up to Day +378, none of the collar-treated dogs were infected with E. canis, whereas 34 of the 35 untreated dogs enrolled before Day +371 were infected. The acaricidal efficacy of the collar ranged from 90 % to 100 % for the duration of the assessment period.