More insight into the transmission dynamics of bacteria between animals is gained with help of transmission experiments. In a transmission experiment various aspects of transmission can be studied. For example, more insight into the transmission dynamics can be gained, transmission can be quantified, or the effect of interventions on transmission can be quantified so that better-directed intervention strategies can be devised. The main goal of the research described in this thesis was the development of methods to quantify bacterial transmission in an experimental setting. We restricted the research to the transmission of one specific bacterium, i.e. Actinobacillus pleuropneumoniae serotype 9 in pigs. A. pleuropneumoniae is regarded as a primary pathogen that causes pleuropneumonia in pigs and brings considerable economic losses about world-wide. Direct transmission from pig to pig is believed to be the most important transmission route of this bacterium, therefore, prevention or reduction of transmission in direct animal to animal contact should in principle lead to eradication. By conducting several transmission experiments we got a better understanding of the transmission dynamics of the bacterium. It was concluded that an infectious state is related to an A. pleuropneumoniae positive tonsil at necropsy. Another conclusion was that the infectivity of a pig is a tenfold higher on days where more than ten A. pleuropneumoniae bacteria were isolated from the nasal swab than on the other days. Furthermore, new statistical and mathematical methods were developed to estimate or test hypothesis about the level of transmission. Statistical methods were based on the transient state (TS) algorithm. The TS algorithm is based on the stochastic susceptible-infectious-removed (SIR) model and provides a time-dependent probability distribution over the number of infected individuals during an epidemic. TS methods are difficult to calculate due to numerical limitations. Therefore, one would probably resort to the easily applicable but less appropriate final size (FS) methods. So, we investigated the error made when FS methods are used instead of TS methods. This error was generally not substantial. Furthermore, a new method to find a difference in transmission between two treatment groups ( MaxDiff test) has been developed and compared to tests based on FS and TS algorithms and a test based on a Generalised Linear Model (GLM). The GLM test was most powerful in finding a difference in transmission. Next were the TS test and the MaxDiff test, which were approximately equally powerful, but more powerful than the FS test especially when the R 0 in both treatment groups are larger than 1. At the end, we tested the effect of vaccination on the transmission of A. pleuropneumoniae in the newly developed experimental design. The effect of vaccination was quantified with a method based on a generalised linear model, which appeared to be most appropriate for the quantification of A. pleuropneumoniae transmission. The effect of vaccination on the susceptibility could not been demonstrated.