Cattle selected for lower residual feed intake have reduced daily methane production

Seventy-six Angus steers chosen from breeding lines divergently selected for residual feed intake (RFI) were studied to quantify the relationship between RFI and the daily rate of methane production (MPR). A 70-d feeding test using a barley-based ration was conducted in which the voluntary DMI, feeding characteristics, and BW of steers were monitored. The estimated breeding value (EBV) for RFI (RFIEBV) for each steer had been calculated from 70-d RFI tests conducted on its parents. Methane production rate (gl d) was measured on each steer using SF6 as a tracer gas in a series of 10-d measurement periods. Daily DMI of steers was lower during the methane measurement period than when methane was not being measured (11.18 vs. 11.88 kg; P = 0.001). A significant relationship existed between MPR and RFI when RFI (RFI15d) was estimated over the 15 d when steers were harnessed 2 for methane collection (MPR = 13.3 x RFI15d + 179; r(2) = 0.12; P = 0.01). Animals expressing lower RFI had lower daily MPR. The relationship established between MPR and RFI15d was used to calculate a reduction in daily methane emission of 13.38 g accompanied a 1 kg/d reduction in RFIEBV in cattle consuming ad libitum a diet of 12.1 MJ of ME/kg. The magnitude of this emission reduction was between that predicted on the basis of intake reduction alone (18 g center dot d(-1)center dot kg of DMI-1) and that predicted by a model incorporating steer midtest BW and level of intake relative to maintenance (5 g center dot d(-1)center dot kg of DMI-1). Comparison of data from steers exhibiting the greatest (n = 10) and lowest (n = 10) RFI15d showed the low RFI15d group to not only have lower MPR (P = 0.017) but also reduced methane cost of growth (by 41.2 g of CH4/kg of ADG; P = 0.09). Although the opportunity to abate livestock MPR by selection against RFI seems great, RFI explained only a small proportion of the observed variation in MPR. A genotype x nutrition interaction can be anticipated, and the MPR:RFIEBV relationship will need to be defined over a range of diet types to account for this.