Summary Observations on the production of animals before and after the development of laboratory evidence of mastitis were made. These data are derived from animals in the early stages of infection and animals in which the disease was latent in character. In many cases no clinical evidence was observable during most of the mastitis reacting periods. Since animals showing obvious clinical evidence of mastitis were eliminated from the herds as a matter of routine practice before the full impact of chronic mastitis on production of milk was manifested, it is evident that the full adverse effect on yield is not here measured. Rather, these data tend to show that a loss in yield may occur in the majority of cases during the incipient stage of the disease. In 240-day lactations of a group of 30 cows having a history both as mastitis free and mastitis positive based on the bromthymol blue test, the sediment test, the leucocyte count, and the shedding of organisms, there was a loss of 463 pounds of milk attributable to mastitis. In another herd of 22 cows there was a loss in yield of 425 pounds. These reductions are between 4 and 5 per cent and are not particularly significant. A loss in yield was manifested in about two thirds of the individual cases. When only one quarter was positive there was no loss in yield. Such are usually incipient stages, and possibly also compensation in yield is made by the unaffected quarters. The loss, however, increased in magnitude with each additional quarter involved amounting to about 15–20 per cent with all four quarters positive. When the results were segregated for each diagnostic test it was found that for those reacting to the bromthymol blue test the loss in yield was 837 pounds, and for those shedding S. mastiditis the loss amounted to 1100 pounds per lactation. The loss in yield of sediment positives was 596 pounds and of leucocyte positives 573 pounds. No effect on the butterfat percentage was observed.