SummarySmall numbers of bovine staphylococci, usually less than 30, were infused into the teat cistern of 17 cows secreting less than 100 000 leucocytes/ml milk. Infection was assumed to have occurred if the organisms multiplied sufficiently to be recovered in the foremilk and inflammation was regarded as demonstrated if there was an increase in leucocytes. If the staphylococci were recovered at the 1st post-infusion milking they were consistently recovered thereafter and inflammation always followed but was never evident at the 1st post-infusion milking. Inflammation was usually evident by the 3rd but in one instance was delayed to the 11th post-infusion milking. There was never inflammation in the absence of infection. When there was a high probability that the infusion contained one organism or more, infection occurred in 2 of 10 quarters infused with coagulase-negative staphylococci and 23 of 25 quarters infused with coagulase-positive staphylococci.A significant positive relationship was found between the number of staphylococci found at the 1st post-infusion milking and the number of leucocytes at the 3rd postinfusion milking.