The objectives of this study were (1) to evaluate available coprological and serological tests for detection of Fasciola hepatica infection in field conditions, (2) to investigate if the season when samples were collected affects the interpretation of the test results, and (3) to evaluate if the test results are associated with the level of infection. During weekly visits to an abattoir, the whole liver, a rectal faecal sample and a blood sample were collected from 100 cows in two seasons each ("spring"=February-May 2006 and "autumn"=October-December 2006). A sedimentation-flotation technique on 4g (SF 4g) or 10g (SF 10g) of faeces, a copro-antigen ELISA and two indirect serum F. hepatica ELISAs (excretory-secretory (ES) and Pourquier ELISA) were performed and the test results were compared with the presence of infection and worm counts at liver necropsy. Over both seasons the sensitivity (Se) and specificity (Sp) were for the SF 4g 43% and 100%, for the SF 10g 64% and 93%, for the copro-antigen ELISA 94% and 93%, for the ES ELISA 87% and 90% and for the Pourquier ELISA 88% and 84%. Significant between-season differences (P<0.05) were observed in the sensitivities of the two serological ELISAs: whereas the Pourquier ELISA had a higher sensitivity in spring than in autumn, the opposite was true for the ES ELISA. There were no significant between-season differences in the specificity for any of the tests. The test results of the SF 4g, copro-antigen ELISA and ES ELISA were associated with the level of infection of the animal. Given a positive test result of the SF 4g it is at least 11 times more likely that the animal is carrying a heavy infection (>10 flukes) than that is free of infection or lightly infected (< or = 10 flukes). Weak ( approximately 0.3) and moderate ( approximately 0.6) correlations were observed within infected animals of level of infection with ES and copro-antigen ELISA results, respectively.