Sampling methods for preputial secretions and culture methods for Tritrichomonas foetus were compared in a two-by-two factorial experiment. Fourteen bulls were confirmed as infected by repeated weekly sampling and culture on Diamond's medium over the course of 10 months. In a six-week test of alternative methods of diagnosis, the bulls were randomly assigned to have their preputial contents sampled weekly either by washing or by scraping of the preputial membrane. The samples were cultured on Claussen's medium and in InPouch packets at 37 degrees C in air for seven days. The results were compared by McNemar's chi 2 test for symmetry. Washing yielded 69 positive cultures from 84 samples whereas scraping yielded 65 from 83. These differences were not statistically significant. InPouch culture detected 73 of 83 samples from positive bulls, whereas Claussen's medium detected 61 of 83. The difference in sensitivity between the two culture methods was highly significant (P = 0.0027), with the InPouch method 'missing' two positive samples that were detected by Claussen's medium and Claussen's medium missing 14 positive samples that were detected by InPouch culture.