We aimed to evaluate comfort of dairy cows using behavioural and physiological indicators. In the 1st experiment, after a grazing period of 2-3 weeks and acclimatizing dairy cows to the experimental situation, access speeds of cows on a 100-m alley were measured twice to 3 grazing areas: large (900m2), medium (400m2) and small (100m2). Four parity groups of 3 cows were used: G1 (mean parity 4.7), G2 (2.7), G3 (1.7), G4 (1.3). Each area was allocated to each group in random sequence. The access speed to each area was analyzed by Kruskal-Wallis test or ANOVA. In the 2nd experiment, we established a grazing area and zero-grazing area (900m2 each) with 3 troughs, each containing fresh forage of 5kg. Four parity groups of 3 animals were used: G1 (3.0), G2 (1.7), G3 (2.0), G4 (1.0). The access speeds to both areas were measured as in the 1st experiment and analyzed by paired t-test. Additionally, plasma cortisol and oxytocin were measured in blood from cows (n=8) collected after a 10-min stay in both areas, and analyzed by Wilcoxon signed-ranks test. In the 2nd trial of experiment 1, G1, G2, and G3 did not differ in access speeds to the 3 areas. G4 moved slower to the large area than the other areas (P<0.05). In the 2nd trial of experiment 2, G1 and G2 moved faster to the grazing area (67.4±0.0 vs 15.4±1.5 (m/min);P<0.01, 71.6±3.1 vs 38.1±7.7 (m/min);P=0.07). 1 cow in G1 and G2, 2 cows in G3 and all heifers in G4 did not visit to the zero-grazing area. Plasma cortisol of animals was higher in the zero-grazing area than the grazing area (3.77±1.22 vs 2.33±0.74 (mg/dL); P<0.05), whereas plasma oxytocin showed the opposite effect (131.2±34.0 vs 155.7±34.7 (mU/ml); P<0.05). We conclude that grazing opportunity enhanced cow comfort.