Efficiency of use of metabolizable energy for body weight gain in pasture-based,nonlactating dairy cows

Four cohorts of nonlactating, pregnant dairy cows (n = 50, 47, 45, and 42) were individually fed indoors to determine the amount of feed required for body weight (BW) gain from autumn pasture and commonly used supplementary feeds. These results were used to estimate the apparent efficiency with which metabolizable energy (ME) is used for BW gain (app_k_g). Control cows were offered autumn pasture to estimated maintenance requirements (~0.55 MJ of ME/kg of BW^0.75), with an additional 20 MJ of ME/d allocated for pregnancy and activity. All other cows received the same allowance of autumn pasture and an additional allowance (2.5 or 5.0 kg of dry matter/d) of autumn pasture (Past), spring pasture silage (Psil), maize silage (Msil), cracked maize grain (Mgr), or palm kernel expeller (PKE), resulting in a total of 11 treatments. Individual cow dry matter intake was determined daily; BW was recorded once per week for cohorts 1 and 2, and 3 times per week for cohorts 3 and 4. The ME contents of feeds were estimated from feed quality assays. Regression analyses were used on each feed to determine the ME requirement for 1 kg of BW gain. The app_k_g of Past and Msil was 0.34 and 0.47, respectively; these estimates are in line with published literature. The app_k_g of Psil (0.50) was consistent with the published k_g for spring pasture, from which the silage was made. Palm kernel expeller had the greatest app_k_g (0.61). The reasons for this cannot be deduced from the current study but may reflect the relatively high fat content of the feed and the high k_g of fat. The app_k_g for Mgr was low (0.38) in comparison with the other supplementary feeds and, in particular, relative to its feed ME and published k_g estimates. Although the reason for the low app_k_g cannot be deduced from the current data, the most plausible reason is the preferential use of propionate-derived glucose for conceptus metabolism rather than BW gain, a factor not accounted for in previous experimental models that did not use late-gestation cows. In summary, the app_k_g for autumn pasture was low but consistent with historical growth rate trials in other ruminant species. In comparison, Msil, Psil, and PKE were used with a greater apparent efficiency (app_k_g = 0.47 to 0.61), but Mgr resulted in a relatively low rate of gain per MJ of ME (app_k_g = 0.38). These differences have implications for accurate feed budgeting on farm.