Impact of weighing procedures and variation in protein degradation rate on measured performance of growing lambs and cattle.

A lamb growth trial was conducted to evaluate soybean meal and blood meal and to compare two experimental designs (a completely randomized design and a switchback design). A larger amount of variation was observed with the switchback than with the randomized design. A larger number of animals, longer treatment period or increased number of weighings may be needed to overcome end point weighing errors and decrease the variation in the switchback design. In both designs, there was a significantly greater conversion of protein to gain for blood meal (BM)-supplemented lambs than for soybean meal (SBM)-supplemented lambs. The relative value of BM compared with SBM was 251% (completely randomized design) or 296% (switchback design). Weighing steers 2 or 3 d at the beginning and end of the trial reduced variation and increased the probability of detecting differences in daily gain, gain/feed and protein efficiency when compared with a standard weighing method (once at the beginning and once at the end of the trial). Regressing 11 weekly weights on an average of one, two or three initial and final weights also reduced variation and increased the probability of detecting treatment differences. Regressions of weekly weights were more useful as the number of initial and final weights was decreased. Daily gain and gain/feed increased linearly (P less than .05) with protein level when BM or SBM was fed. A nonlinear model was used to predict protein requirement, maximum gain and relative protein values. There was a range in relative values for BM of 255 to 348% depending on the method of analysis used.