Abstract #M19

# M19
Relationships between dry matter intake, rumination time, lying behavior, and milk yield in lactating cows.
Dafu Shao*1, Fengxia Wang1, YaJing Wang1, Shengli Li1, Zhijun Cao1, 1State Key Laboratory of Animal Nutrition, College of Animal Science and Technology, China Agricultural University, Beijing, China.

The objective of this study was to investigate the relationship between dry matter intake (DMI) and rumination time (RT), lying behavior (lying time and frequency), and milk yield in dairy cows. Data were collected from 32 mature Holstein dairy cows (16 primiparous, PR; 16 multiparous, MP, parity = 1.7 ± 0.8; milk production = 10,019 ± 166 kg/year; mean ± SD). DMI was monitored using Insentec feed bins (Insentec BV, Marknesse, the Netherlands). RT was monitored electronically using HR-Tags (SCR Engineers Ltd., Netanya, Israel). Lying behavior was monitored using an activity logger (HOBO Pendant G; Onset Computer Corp., Cape Cod, MA). Data were analyzed by the R programming language (Mathsoft). According to the level of RT, cows were assigned to the low-RT group (L: RT = 277.65 ± 39.61 min/d; 8 PR and 8 MP) and the high-RT group (H: RT = 413.21 ± 46.89 min/d; 8 PR and 8 MP). One-way ANOVA was used to analyze the difference between H and L groups. There was no significant difference between H and L group in DMI (H: 20.59 ± 2.63 kg/d; L: 20.31 ± 3.34 kg/d; P = 0.79), lying behavior (lying time, H:690.44 ± 155.46; L:773.00 ± 163.41, P = 0.20; Lying frequency, H:11.25 ± 3.14, L:10.79 ± 3.39, P = 0.43), and milk yield (H: 32.04 ± 6.00 kg/d; L: 32.15 ± 4.31 kg/d; P = 0.95). Pearson correlation was used to describe associations among DMI, RT, lying behavior (lying time and frequency) and milk yield. Multiple Linear Regression analysis was performed with lm package of the R programming language. A significant positive relationship existed between DMI and milk yield (P = 0.01, r2 = 0.93). Rumination time (P = 0.34) and lying behavior (Lying time, P = 0.36; and Lying frequency, P = 0.14) were not affected by DMI. Only the milk yield (P < 0.01) was significantly affected by DMI, and the equation was DMI = 8.97934 + 0.35762 × MY, R2 = 0.3865, and which MY represents milk yield. These results indicate that milk yield can be used to estimate DMI; however, RT and lying behavior are independent variables and not the suitable indicators for DMI.

Key Words: rumination time, lying behavior, DMI