Ten transmission trials with Actinobacillus pleuropneumoniae were carried out. The observed transmission was highly variable, which was surprising since the design of the trials was very similar. We investigated whether the variable transmission could be explained by variation in infectivity of A. pleuropneumoniae infected pigs. We looked for measurable characteristics, which could be indicative for infectious pigs or for the level of infectivity. The characteristic that appeared to be most indicative for a pig being infectious was an A. pleuropneumoniae positive tonsil at necropsy. The characteristic that was correlated to the level of infectivity was the number of A. pleuropneumoniae colonies isolated from the nasal swab, i.e. the probability for an infectious pig to infect a susceptible pig was tenfold higher on days where at least ten colonies were isolated. In this study it is shown that it is possible to measure the bacterial transmission of A. pleuropneumoniae under controlled circumstances if variation in infectivity is taken into account.