An experiment originally employing ten pairs of twin heifers was carried out to assess the long-term effect of a complete hygiene regimen designed to protect the mammary gland from infection. As long as they remained in the experiment, one member of each twin pair was milked under a complete hygiene system while her sister was milked with no hygiene at all. Significantly more intramammary infections occurred in the controls (no hygiene) than in the hygiene cows. Very significantly more teat ducts became colonized by bacteria in the controls than in the hygiene cows, and significantly more of these developed into intramammary infections in the controls. Thus, hygiene effectively controlled both bacterial colonization of teat ducts and intramammary infections during lactation, but this control was to some extent offset by infections which took place during the dry period and at calving. There appeared to be a relationship between susceptibility to teat duct colonization and medical conditions other than mammary gland infections in six pairs of twins which had to be discarded before the third lactation.