Milk from a Montana dairy was implicated in an outbreak of E. coli O104:H21. No specific formulation of milk was singled out in the epidemiolgic investigation. An inspection of the dairy was conducted. The dairy plant's records for internal, microbiologic, quality-control showed that the coliform counts of the milk had exceeded the state's regulations. Cultures from selected post-pasteurization piping and equipment in contact with finished milk yielded fecal coliforms, however no E. coli O104:H21 was dectected. Testing of cows providing the raw milk for the plant did not yield E. coli O104:H21. This organism was a sorbital fermenter and would not readily have been detected with then commonly available laboratory methods used to detect the presence of E. coli O157:H7. Shiga-like toxin testing was not yet routinely available to screen for organisms such as this one.