Comparison of the in vitro gas production technique and the nylon bag degradability technique to measure short- and long-term processes of degradation of maize silage in dairy cows

Short-term (2-8 h) and long-term (8-48 h) processes of degradation of sampl es of 10 different maize silages were compared using two different techniqu es: (i) Hohenheimer in vitro gas production technique (GP) and (ii) the nyl on bag degradability technique. Measurements of shortterm degradability wer e complemented with determination of non-degraded dry matter loss from nylo n bags, The objectives of the work were to (i) study the influence of sampl e processing on degradability and (ii) compare the nylon bag and the gas pr oduction techniques, for determining the nutritive value of maize silage. T he maize silage samples for in situ degradability were in the form of eithe r fresh approximately 5 mm chop length (F), F dried (FD), FD ground to pass through a 3 mm screen (FD3mm) and for in vitro gas production, FD ground t o pass either a 1 or a 3 mm screens (FD3mm and FD1mm). Incubations were car ried out for 2, 3, 8, 24 and 38 h (with a supplementary measurement at 16 h in situ). Non-degraded dry matter loss from nylon bags was higher for FD3m m than for FD or F (240.9 > 179.1, 156.3 g/kg respectively, P < 0.05). The effective DM degradability (DMDEf) was 17% higher for FD3mm than for FD or F (524.6 > 448.1, 449.6 g/kg respectively; P < 0.05). Grinding of maize sil age increased the soluble fraction. However, it did not decrease the undegr adable fraction of maize silage. In vitro gas production after 38 h was low er for FD3mm compared with FD1mm (38.23 versus 39.68 ml; P < 0.01). In a Pe arson matrix correlation which included all incubation periods, all maize s amples, DM disappearance, non-degraded loss of DM, and rate of gas producti on, (39 parameters in total yielding 416 correlation coefficients), only 63 (15%) of the correlations were significant (P < 0.05; r > 0.63). Correlati ons of DM disappearance of FD samples during the first 8 h of the incubatio n with gas production of FD3mm samples for the same incubation intervals ga ve the best correlations (r = 0.70 for values at 2 h, 0.75 for values at 4 h, 0.67 for values at 8 h). Processes of degradation were more correlated w ith parameters of the short term than the long term (P < 0.05). The nylon b ag technique and Hohenheimer technique did not describe the degradation of maize silage in the same manner. This may be explained by loss of non degra ded DR I which occurred only with the nylon bag and which was not implicate d in the GP technique. Thus, dry matter degradability and GP did not give t he same predictions of nutritive value. (C) 1999 Elsevier Science B.V. All rights reserved.