Milking routines in automatic milking systems (AMS) differ from those in co
nventional milking. While milking intervals are constant and teat cups are
attached simultaneously in conventional milking, in AMS milking intervals a
re variable and attachment of teat cups can be sequentially delayed. Experi
ments were performed to test effects of pre-milking teat cleaning, of diffe
rent milking intervals in different lactational stages, of delayed teat cup
attachment and individual teat cup removal on oxytocin release, milk eject
ion and milk removal. Teat cleaning caused oxytocin release and induced mil
k ejection. The lag time from start of teat Stimulation until milk ejection
occurred increased with decreasing milking interval and was longer in late
than in early lactation. In addition, milk fat content increased towards t
he end of lactation and was higher after short than after long intervals fr
om previous milking. Sequentially delayed attachment of teat cups every 20
s or every 60 s did not reduce oxytocin release. However, total interruptio
n of teat stimulation for 2 ruin between pre-stimulation and start of milki
ng resulted in transiently decreasing oxytocin concentration and increasing
amounts of residual milk. The quarters in which teat cups were last remove
d had non-significantly the smallest amounts of stripping and residual milk
. The stimulatory effect of single teat cups during sequentially delayed at
tachment is sufficient to maintain adequate oxytocin release and maximum mi
lk ejection. (C) 2001 Elsevier Science B.V. All rights reserved.