In ruminants, milking induces the release of prolactin (PRL). Milkinginduced PRL release decreases as lactation advances and PRL is a survival factor for mammary epithelial cells (MEC) of cows, suggesting a galactopoietic role of this hormone. Nevertheless, suppression of PRL by bromocriptine has produced ambiguous effects on milk yield in cows. To assess the effect of inhibition of PRL release in lactating dairy cows, nine Holstein cows were assigned randomly to treatments during 3 5-d periods: 1)daily i.m. injection of 2 mg of Quinagolide (a PRL release inhibitor), 2)daily i.m. injection of 2 mg of Quinagolide and twice a day (milking time) i.v injection of PRL (2µg/kg of body weight),3) daily injection of vehicle as control. Blood and milk samples were harvested at milking. MEC were purified from milk. Daily injections of Quinagolide reduced milking-induced PRL release (P<0.05). The amount (area under curve) of PRL during milking after the PRL injections was similar to that of endogenous PRL discharges in control treatment. Quinagolide decreased milk production (P<0.05), milk protein content (P<0.05) and milk lactose content (P<0.05), without modification of milk fat composition or fat globule size. PRL injections had no effect on milk yield and milk fat composition but tended to increased milk protein content (P<0.10). Injections of Quinagolide increased the number of MEC harvested from milk (P<0.05) but PRL injections tended to decrease it (P=0.10). PRL injections also increased viability of MEC harvested from milk (P<0.05). Injections of Quinagolide decreased kappa-casein (P<0.05) and alpha-lactalbumin (P<0.05) mRNAs in milk MEC. Exogenous PRL have no affected the level of these mRNA. In conclusion, chronic administration of Quinagolide reduces milk production in dairy cows by affecting MEC activity. PRL injections at milking time were not sufficient to restore milk yield but influenced the viability of MEC purified from milk.