E R Ørskov
Rowett Research Institute, Bucksburn, Aberdeen AB2 9SB
Abstract
References
The present methods of estimating nutritive value were discussed including chemical analysis such as crude fibre, acid and neutral detergent fibre, lignin and soluble carbohydrates and biological measurements such as in vitro and in vivo determination of digestibility and determination of rate and extent of digestion using the nylon bag method. It was shown that the chemical analyses were very poor predictors of differences in nutritive value between different types of fibrous residues, between different varieties of straws and also in estimating improvements occasioned by chemical treatment. Only biological measurements estimated utilization with any degree of accuracy, and of the biological measurements the nylon bag method of describing rate and extent of degradation was by far the best predictor and better than in vivo digestibility determined in restricted fed animals.
Development of laboratory methods to estimate nutritive value of feeds has been the goal of many research workers in the past. Numerous equations have been developed relating chemical analysis to nutritive value. For concentrate feeds developments of chemical methods to predict nutritive value have on the whole been very successful. This in part is due to the fact that they usually contain very little fibre and the starch sugar or fat they contain are virtually completely digestible. Estimating fibre digestibility for ruminants has been fraught with many more problems. The method used for many years was the Weende system developed in Germany to measure crude fibre, crude protein, ether extractive and ash. The apparent credibility of this system was due to the fact that the feeds used to develop the equation varied greatly in nutritive value, due mainly to different stages of maturity of the plants investigated.
The use of biological measurements to estimate nutritive value of forages was introduced particularly by development of the in vitro system by Tilley and Terry (1963) in which substrate disappearance was measured after laboratory incubation in rumen fluid. This was a spin-off from development of easy methods of fistulating ruminant animals and thus obtaining rumen fluid. The problem, however, with the in-vitro system is that it assumes that the conditions in the incubators remain similar to rumen content for 48 hours in spite of the accumulation of end products. It also depends on maintenance of constant temperature in the laboratory incubation. Power failures will render the results void and power failures are not unusual in developing countries.
Development of equations to estimate in vivo digestibility from any parameters, chemical or biological, assumes that the in-vivo digestibility gives the true value of the feed. It is now increasingly recognized that even the in vivo value is by no means constant. It varies mainly due to differences in rumen retention time, but can also vary according to the rumen environment. This infers that the value to be predicted is not static, but is itself dynamic.
This is best illustrated by describing the course of digestion, for example of straw, as in Fig. 1. This relationship can be described by the formula
p = a + b (1 - e-ct)
where a, b and c are constants, p is degradation at time (t) and e is the natural logarithm (Ørskov and McDonald, 1979). This type of equation is chosen because it gives some biological meaning to the constants insofar that a is the intercept or the immediately soluble fraction, b is the insoluble but potentially degradable material and c is the rate constant of b. It follows that a + b is the maximum potential digestibility of the feed. If an accurate measure could be obtained of these parameters it would assist greatly in predicting not only digestibility but intake as well.
In Fig. 1 it can be seen that the maximum potential, or the asymptote, is reached only after more than 90 hours of fermentation. The rumen retention time, however, is seldom as high as that. Mean retention time with straw is likely to vary from 36 to about 60 hours. This variation however implies that one measurement in vivo is only a point on the graph, and may not represent the in vivo situation in which it is to apply. In other words the use of sheep fed at maintenance level of feeding may give values quite different from cattle given a high level of feeding. The difference is not likely to be constant either, but will depend on the shape of the degradation curve, i.e. its a, b and c values. The in vivo estimates are likely to differ most from the maximum potential in feed with a high b value and a low rate constant c.
What is then the best description of nutritive value of fibrous residues? A description of degradation as in Fig. 1, using nylon bags and a rumen environment in which cellulolysis is optimal, is no doubt more informative than one in vivo value of digestibility. In fact determination of in vivo digestibility can be used subsequently as a means of learning whether the diets are being utilized adequately, which in turn can lead to further studies to investigate reasons why if this is not the case.
Although the use of nylon bags in rumen cannulated sheep is a robust tool and does not fail due to failure of power supply etc., it would still be of great value if chemical analysis could identify some differences in nutritive value of fibrous feeds.
In a recent experiment (Ramanzin et al. 1986) studied whether chemical analysis could identify differences in quality of straw and differences in quality conferred by ammonia treatment. It is clear from the results presented in Table 1 that there was no relationship between ADF and nutritive value within leaves and within stems although there were clear differences in degradability due to variety, confirmed also in voluntary intake and in vivo digestibility studies. However ADF was clearly different between stems and leaves.
Table 1. Effect of variety and component of straw on chemical analysis of acid detergent fibre (ADF), neutral detergent fibre (NDF), crude protein (CP) and degradation of dry matter during 48 furs.
|
Variety Component Treatment |
ADF (%) |
NDF (%) |
CP (%) |
Degradation in nylon bag for |
|
|
48 hr. |
72 hr. |
||||
|
Corgi leaves untreated |
45.2 |
83.2 |
4.4 |
76.0 |
85.9 |
|
Corgi leaves NH3 treated |
47.4 |
77.7 |
- |
83.5 |
90.1 |
|
Gerbil leaves untreated |
49.3 |
84.5 |
4.4 |
70.8 |
80.3 |
|
Gerbil leaves NH3 treated |
47.5 |
79.7 |
- |
74.4 |
86.2 |
|
Corgi internodes untreated |
61.4 |
93.0 |
2.1 |
53.3 |
58.8 |
|
Corgi internodes NH3 treated |
62.0 |
90.7 |
- |
59.1 |
66.3 |
|
Gerbil internodes untreated |
62.1 |
93.7 |
3.0 |
37.5 |
44.1 |
|
Gerbil internodes NH3 treated |
62.7 |
94.4 |
- |
51.5 |
57.2 |
Table 2. Effect of variety and type of straw on chemical analysis of acid detergent fibre (ADF), neutral detergent fibre (NDF), acid detergent lignin (ADL), cellulose digestibility (NCD) and degradation measured by in vitro or by nylon bag incubation.
|
Variety |
Treatment |
ADF |
NDF |
ADL |
NCD |
OMD |
Degradation in nylon bags for: |
|
|
In vitro |
48 hr |
72 hr. |
||||||
|
Gerbil
|
Untreated |
57.9 |
87.5 |
9.0 |
24.6 |
27.6 |
32.2 |
36.1 |
|
NH3 treated |
55.5 |
84.0 |
8.3 |
35.2 |
37.8 |
47.4 |
53.4 |
|
|
Igri
|
Untreated |
55.4 |
86.4 |
7.7 |
29.3 |
29.5 |
36.7 |
41.4 |
|
NH3 treated |
55.4 |
84.1 |
8.0 |
34.0 |
37.5 |
44.3 |
49.8 |
|
|
Corgi
|
Untreated |
51.2 |
84.0 |
6.3 |
34.4 |
39.0 |
46.8 |
50.9 |
|
NH3 treated |
50.0 |
81.0 |
6.1 |
47.3 |
54.1 |
61.6 |
63.6 |
|
|
Golden Promise
|
Untreated |
55.5 |
85.0 |
7.3 |
33.5 |
36.4 |
43.5 |
50.5 |
|
NH3 treated |
52.3 |
79.9 |
6.9 |
44.5 |
45.6 |
53.1 |
57.8 |
|
|
Wheat
|
Untreated |
52.0 |
80.7 |
8.0 |
31.5 |
31.7 |
42.1 |
44.5 |
|
NH3 treated |
48.8 |
75.5 |
7.0 |
42.9 |
44.8 |
51.3 |
57.4 |
|
Preliminary and unpublished results from Rowett Institute and North of Scotland College of Agriculture.
Figure 1. Description of degradability of fibrous residues.
In later unpublished work (G.W. Reid, M. Kay and E.R. Ørskov) we have made some comparisons of chemical and biological methods (Table 2). It can be seen again that ADF and lignin content did not distinguish between nutritive values of type and varieties of straw and could not predict increases in nutritive value due to ammonia treatment. On the other hand it can be seen that both incubation with cellulase enzymes and in vivo in rumen fluid distinguished between variety, types and ammonia treatment.
It can be asked quite legitimately, if the chemical analyses do not predict nutritive value, why use resources to measure them? It is clear that differences between nutritive values of straws could not be predicted from chemical analysis developed so far.
The available evidence suggests that while gross chemical composition may be interesting, only biological measurements can presently provide sufficient information about the nutritive value of fibrous feeds or residues. We need to search for laboratory methods which can identify the factors controlling accessibility of bacteria to the substrate and the rate at which it can occur. These factors are more likely to be found in surface chemistry of cell walls rather than in gross chemical composition.
Ørskov E.R. and McDonald I. 1979. Journal of Agricultural Science, Cambridge 92, 499.
Ramanzin M., Ørskov E.R. and Tuah A.K. 1986. Animal Production (In press.)
Tilley J.M.A. and Terry R.A. 1963. Journal of the British Grass land Society 18: 104.
