The importance of herd management in loose housing systems to the social behaviour of dairy cows.

In five loose housed dairy herds three different kinds of herd management w ere tested in two variants with respect to frequencies of agonistic social behaviour. Treatments were (1) a short (0.5 h) and a long (3 h) fixation ti me in the feeding rack, (2) single and group (3 animals) introduction of ne w heifers into the herd and (3) an open and a closed outdoor yard during ni ghttime. The investigated agonistic behaviour patterns were: pushing and ch asing. The herds were observed in the evening after milking during one four -hour period starting one hour after opening the feeding rack when testing treatment (3) and immediately after opening the feeding rack when testing t reatment (1) and (2). Statistical analysis were carried out with the Wilcox on signed-ranks test for matched samples. Effects of fixation time differed inconsistently and were not significant. In 3 out of 5 farms the frequency of agonistic behaviour was lower, when cows were restrained 3 h in the fee ding rack compared to 0.5 h. In two of these 3 herds, the animals had the p ossibility to drink water from bowls directly at the feeding rack, in the t hird herd the food was silage which has a higher water content. Therefore i n these herds, the agonistic interactions at the drinking facility in the s table after opening the feeding rack was low. Contrary, in the two other he rds, with dry feed (hay) and no drinking bowls at the feeding rack, frequen cy of agonistic behaviour was higher after the long restraint which might b e due to higher competition at the drinking facilities. Agonistic interacti ons per cow as well as per new introduced heifer were lower (p less than or equal to 0.05) when only a single heifer was introduced to the herd compar ed to the introduction of a group of 3 heifers. The frequency of agonistic social behaviour of homed dairy cows that had access to a yard at night was significantly lower compared to the situation when the yard was closed ove r night (p less than or equal to 0.05). This may be due to the higher space allowances per cow but also to the additional structuring of the locomotio n area.