Avian tuberculosis (ATBC) is a significant cause of morbidity and mortality in birds in zoos and breeding establishments. The primary sources of Mycobacterium avium subsp. avium (MAA) of serotypes 1, 2 and 3 are infected animals (esp. birds), who shed MAA in different ways and thus contaminate the environment. The first aim of this work was to compare the efficiency of the diagnostic methods that are routinely used for the diagnosis of ATBC (skin test, serology and culture of faeces and eggs) in naturally infected hens with different levels of infection. The second aim was to determine the excretion rate of MAA in faeces and eggs. The tuberculin skin test gave a positive result in nine (42.9%) infected hens of which four (57.1%) and one (14.3%) were heavily and slightly infected hens, respectively. A positive serological response to MAA-b antigen (water bird isolate of serotype 1) was observed in five (23.8%) and to MAA-p antigen (pig isolate serotype 2) in seven (33.3%) hens. No correlation between serological and skin-test data was found. The results show that both techniques, serological and skin-test data are inadequate for the diagnosis of ATBC. In consecutively euthanized hens, with heavy infection and tuberculous lesions, serological positivity was significant (P < 0.05) in comparison with slightly infected hens lacking tuberculous lesions. Faecal culture detected MAA in 50 (29.8%) of 168 samples collected for eight days before euthanasia. MAA excretion in faeces was intermittent, but significantly (P < 0.01) higher in heavily infected hens. No mycobacteria were detected in any of the 43 examined eggs, which implies that the shedding of MAA and/or transmission of ATBC through eggs may not be frequent events.