Previous studies have shown that surgical castration wounds take between 10 and 61 d to heal. The objectives of this work were to describe healing, inflammation, lying behavior and serum concentration of substance P after surgical castration in beef calves and to evaluate the effect of a possible intervention, a single injection of flunixin meglumine (1.1 mg/kg IV, a NSAID), on the healing process. Calves (mean ± SE: 25 ± 2.0 d of age; 54 ± 1.4 kg BW) were surgically castrated with or without an injection of flunixin immediately before the procedure (n = 24/treatment). Healing was measured with a 5-point scale (1 = fresh wound, 5 = no visible incision or inflammation) as well as weight gain, scrotal size and scrotal surface temperature, on d 1, 2, 3, 7, 14, 21, 28, 35, 49 and 63 after castration. Serum concentration of substance P was recorded on all d, including d 0, but not d 63. Lying behavior was recorded with loggers from 2 d before to 29 d after castration. Inflammation, as measured by scrotal size, peaked on d 2 and 3 after the procedure (e.g. 51 ± 1.0 mm on d 2 versus 28 ± 1.3 mm before castration) and then declined with time (P < 0.001). The first wound to score as fully healed (i.e. 5/5) was seen on d 28; by d 63, 98 per cent of wounds were fully healed. The greatest changes in healing score occurred between d 21 and 35; this was also the peak of wound surface temperature and may correspond with revascularization. Serum concentration of substance P was highest before castration (41 ± 1.2 pg/mL), possibly because the sample was collected after the lidocaine ring block was administered, which was likely painful, and because separation from the dam and restraint. Values began to drop by d 3 (34 ± 1.2 pg/mL) and leveled out by d 21 (30 ± 1.2 pg/mL; P < 0.001). Calves given flunixin had more lying bouts than those that received saline (flunixin by time interaction, P = 0.052), but this pattern emerged on and after d 8, well after the 3 to 8 h half life of this NSAID. In conclusion, castration caused inflammation in the days that followed, and the wounds required a minimum of 4 wks to heal. Provision of an NSAID had no effect on these outcomes. The study is from the Department of Animal Science and the School of Veterinary Medicine, University of California, Davis, USA; Animal Welfare Program, University of British Columbia, Vancouver, Canada; and College of Veterinary Medicine, Iowa State University, Ames, USA. Mintline EM, Varga A, Banuelos J, et al. J Anim Sci 2014 Nov 3 [Epub ahead of print].