A descriptive study of teat morphology, milking machine characteristics, and milking practices in a sample of Brazilian dairy herds

The objective of this study was to describe teat morphology, milking equipment characteristics and settings, and milking practices adopted in a sample of Brazilian dairy herds. A secondary objective was to compare the study variables among herd size categories. Forty herds were visited once. A vacuum recorder was used to analyze all pulsators prior to milking. During milking, the device was mounted to a milking cluster and vacuum levels were recorded at four points (short pulsation tube, short milk tube, and mouthpiece chamber of the front and rear liners) to estimate the following variables: short milk tube vacuum (VSMT), VSMT during peak milk flow, VSMT during overmilking, liner mouthpiece chamber vacuum (VMPC) during peak milk flow, time in overmilking, total milking time, and presence of bimodal milk flow. The dataset included 678 teats from 339 cows (average of 17 teats from 8.5 cows sampled per herd) and 592 measurements from 296 pulsators (average of 14.8 measurements from 7.4 pulsators per herd). Front teats were longer and wider (4.93 +/- 1.34 and 2.58 +/- 0.71 cm) than rear teats (4.04 +/- 1.14 and 2.43 +/- 0.57 cm), respectively. Teat length was negatively associated with herd size and a higher proportion of flat teats was observed in large herds, as compared with medium or small herds. Preparation lag time was longer in small herds (median = 3.67 min) than in large herds (2.19 min). The prevalence of bimodal milk flow was 6.49%. Cows of small herds remained a longer time in overmilking than those of large herds. Higher VMPC at peak milk flow, higher pulsation rate, longer B-phase, and shorter D-phase were observed in large herds (median = 15.20 kPa, 61.93 pulses/min[p/m], 465.43 ms, and 242.01 ms, respectively), as compared with small (10.20 kPa, 58.74 p/m, 428.52 ms, and 498.14 ms) or medium (8.50 kPa, 60.46 p/m, 443.52 ms, and 264.05 ms) herds. The VSMT during overmilking (39.60 kPa) was independent of herd size category. Desirable milking practices, such as wearing of gloves, use of automatic cluster removal, use of liners in good condition, training of staff, no use of oxytocin, and reduced time in overmilking were more frequent in large and medium herds, than in small herds. Results of this study can be useful for situational diagnosis and implementation of milk quality and animal welfare improvement programs.