The current experiments were conducted to look at the impact of differing pre-milking routines on milking time parameters, milk quality and oxytocin profiles. All experiments were performed on Holstein cows milked 3X daily and producing 13-16 Kg/milking. Lag times of 0, 60, 90, 120 and 240 s along with forestripping or not forestripping were applied to 786 Holstein cows. Cows in early to mid-lactation had the highest milk yield in the first 2 min when lag time was 60 s and forestripping was included; however, lag time or forestripping had no impact on unit on-time for early to mid-lactation cows. The combination of forestripping with lag times of 90 and 120 s increased the amount of milk harvested in the first 2 min to >60% of the total and of milk and lag times beyond 60 s reduced milking unit on-time for late-lactation cows independent of forestripping. The quantity and quality of foremilk was analyzed to evaluate the relationship between foremilk and harvested milk plate loop count (PLC) and somatic cell count (SCC). The SCC and PLC of foremilk were not good predictors of the SCC and PLC in the harvested milk fraction. The foremilk represents 0.12% of the total milk harvested and the somatic cells and bacteria found in the foremilk represent <0.28% and1.44% of the total somatic cells and bacteria in milk. Manual stimulation, no stimulation and mechanical stimulation by increased cycle pulsation (300 cycles/min) with low vacuum were included as treatments to analyze their impact on oxytocin profiles and milking parameters. Increased cycle pulsation elicits a similar oxytocin profile to that of cows which were subjected to manual stimulation. Proper cleaning of the teats and detection of mastitis during mechanical stimulation needs further investigation in regard to identification of abnormal milk in real-time and by mechanical means.