Twenty three 1- to 18-week old Ayrshire calves were exposed to infection by Mycoplasma dispar, 1 or 2 at a time, in a series of consecutive experiments. Exposure took place either at direct contact, i. e. with the possibility of physical contact between susceptible and infected calves (type I), or through separation of susceptibles from the nearest infected calf at a distance of 0.8 m (type II) or 1.3 m (type III). Frequent nasal swabbing was used for the detection of the infection and the measurement of the level of colonisation. Ten, 4, and 9 calves were subjected to type I, II, and III exposures, respectively. When the first positive nasal specimen preceding a rise in titres in the following specimens was used to signify an establishment of infection of a susceptible calf, the medians (ranges) of the times for the transmission of infection (including the latent period of early colonisation undetectable by nasal swabbing) in the 3 exposure groups were: 4.5 (1-27), 8.5 (7-9), and 17 (6-32), respectively. The difference between groups I and III was significant (p < 0.05). The last figures for type III of exposures represent the results of only 5 calves succesfully exposed; the remaining 4 of the 9 calves of this exposure type were not found to pick up the infection within the periods of 2 to 3 weeks of exposure. The results indicate a transmission mode via droplets. Four calves exposed only for 4 h by direct contact were found to be infected on days 0, 4, 4, and 7, respectively. This indicated variability in the length of the latency as regards the detection by nasal swabbing. Four of the 7 calves in the type I experiments sampled at hour 4 showed low titers for M. dispar. Only in 1 of these calves was this early finding of transfer followed by positivity and a rise in titers through subsequent specimens. The other 3 calves became positive anew, with subsequent developing colonisation, first on days 4, 4, and 12, respectively.