Previous Page Table of Contents Next Page

SESSION II (contd.)


E. R. Orskov

Rowett Research Institute, Bucksburn, AB2 9SB,
Aberdeen, Scotland, UK


The nylon bag technique is a simple, robust and powerful tool essential in most applied nutrition studies in ruminants. Among its uses are the following:


The use of nylon bags to study rumen fermentation of feeds is not new. The technique has been used for a long time but mainly for definite incubation periods as an estimate of in vivo digestibility. In 1975 it was felt that measuring degradability of proteins by post ruminal collection was not appropriate and attempts were made to adapt the nylon bag method to describe the rate and the extent of degradation of protein supplements (Mehrez and Orskov 1977). We found the method very useful and adapted it immediately in the same way to describe rate and extent of fibre digestion. In other words, the nylon bag was used not only to give information as to the possible extent of digestion or degradation but also the events that led up to it, i.e. the rate at which degradation occurred. We have found that this measurement of rate is particularly important because it provides information on factors affecting intake of roughages.

Voluntary intake of roughages in ruminants is not only affected by the digestibility, but also by the rate at which digestion occurs because both factors influence turnover rate. In many respects, therefore, a description of feeds using the nylon bag is much more instructive and gives more information than the description of in vivo digestibility.

The nylon bag method has the advantage of simplicity. It requires only some nylon bags, rumen cannulated sheep, goats or cows and an oven to dry samples. Equipped with these tools, a lot of information can be obtained without further sophistication. In fact, we use the technique extensively in our Institute simply determining dry matter disappearance.

The nylon bag incubations have advantages in so far as digestion can be studied in the environment to which it is applicable; that is the environment of in vivo rumen fermentation. Studies made during in vivo fermentation have the added advantage that there is no problem of temperature control, there is no problem of removal of end products and so on; problems which have to be coped with as soon as rumen fluid is taken into a laboratory for in vitro fermentation. It must be pointed out, however, that in the method recently developed by Czerkawski and Breckenridge (1979), continuous fermentation may sometimes be adapted for routine laboratory use. In the following pages I will attempt to describe the method, the interpretation of results and the type of problems which the nylon bag is uniquely suited to solve.


Description of technique

Nylon bags with a mesh size of 20–40 um are most suitable. Smaller mesh sizes can be used, in fact the mesh size is a compromise. It must allow entry of rumen microbes and escape of accumulated gas on the one hand, and on the other hand it must keep losses of solid particles to a minimum. Accumulation of gas in the nylon bag can cause the bags to float on top of solid digesta. The size of the bags we use is about 140 × 90 mm and we use 3–5 g of dry samples. It is possible to use larger bags and larger samples and in fact the size of the sample is quite often dependent upon the amount of residue required for analysis. It is essential and extremely important that the materials incubated are able to move freely within the bag, so as to avoid the formation of micro-environments in the bag with resultant poor replications. The diameter of the rumen cannula is also sometimes a function of the size of the bags. We use rumen cannulas with an internal diameter of 40–50 mm. With this type of cannula it is possible to incubate up to 10 samples in the rumen at one time and to withdraw them at intervals. For cattle, larger cannulas and more bags can be used if required.

Preparation of samples for incubation

The preparation of samples must, as far as possible, represent materials as they appear in the rumen had they been consumed naturally by the animal. This is not possible however as this would require use of animals with oesophageal cannulas. For protein supplements quite often we simply use it in the form it is delivered to us. Otherwise dry materials are processed through a hammer mill with a screen of 2.5 – 3 mm. The same screen size can also be used with forages and cereals. For green materials, succulent materials and silage a mincer is more appropriate with a 5 mm screen size.

Position of bags in the rumen

Many workers recommend that bags are anchored with a 25 cm of nylon cord at the top of the cannula in sheep and about 50 cm in the rumen of cattle. This length allows the bags to move freely within digesta both in liquid and solid phase. Some workers use a weight attached to the bags to ensure that the bags do not float on top of the rumen contents while others find that additional weights give little or no improvement in accuracy. Recently, we have used a method of anchoring the bags to a nylon tube which is illustrated in Figure 1. This nylon tube can hold about 5 bags at the same time. The system simplifies the withdrawal of the bags since individual bags, on an individual cord, can become tangled and difficult to withdraw from the rumen.

Incubation times of bags in the rumen

The most appropriate times at which bags can be withdrawn from the rumen in order to best describe the disappearance rate, depend on the shape of the degradation curve. It is not possible to be absolutely specific in the guidance one can give on time for withdrawal of bags. For many protein supplements, withdrawal of bags at 2, 6, 12, 24 and 36 h gives an adequate description. For hay, straw and other fibrous materials, longer incubation intervals are required. In this case, bags are taken out at 12, 24, 48 and 72 h. The most appropriate rules for the description of the curve will be discussed later.

Figure 1

Figure 1
Illustration of plastic tube and attachments of nylon bag for suspension in the rumen

Replication of measurements

The most important source of variation is the between animal component. As a result it is necessary to use more than one animal to get an accurate estimate of a particular feed. Quite often we find that 3 animals is a good number to work with. We find, for instance, that between sheep, variation of dry matter disappearance was 6.2% while the variation between days was 4.9%; and the variability between bags only 3.3%. From this information it is possible to provide guidance as to the importance of replication between bags, between days and between sheep. Table 1 is taken from Mehrez and Orskov (1977).

TABLE 1 Estimated variance of the mean dry matter disappearance using the rumen bag technique for various numbers of bags (b), days (d) and sheep (s) (Mehrez and Orskov 1977)
Replicatesn*Variance of mean
BagsDaysSheep(b × d ×s)(% of mean)

*n is the number of incubations for each time

Use of sheep or cattle

Few comparisons have been made but results so far suggest that if sheep or cattle are given the same feeds then there is little or no difference in degradation rates measured with nylon bags. The use of cattle has the advantage that a larger number of bags can be used. On the other hand sheep are much cheaper to keep and use. It is more important to remember that the feeding level is very important; this sometimes dictates the particular species that has to be used.

Effect of diet on rate of degradation

The diets can have a great effect on rates of degradation, both of fibrous feeds and of protein supplements. Indeed, this is one of the aspects of digestion that is very interesting and important to study; namely, the effect of the rumen environment on degradation of dietary constituents. It is therefore important that protein degradabilities are determined in the rumen of an animal receiving a diet similar to that which is used in practice and for which the results of protein degradability will have to be applied. It has been shown for instance that protein supplements of vegetable origin are normally degraded more slowly in the rumen of animals given high concentrate diets compared with high roughage diets (Ganev et al 1979).


In order to exploit fully the information provided by the description of degradation, whether it be degradation of protein supplements or degradation of fibre, it is essential that the incubation times chosen are such that both the sensitive part of the curve and the asymptote are adequately described. It may well be that for some supplements it is necessary to obtain more incubation times or to correct the incubation times used the first time because of lack of experience of a particular type of feed. Having ascertained that the curve is adequately described it is then possible to apply some mathematics to it in order to assist the interpretation.

We have chosen an exponential description of degradation which fits most situations. The expression is of the type: p = a + b (1 - e -ct) (Orskov and McDonald 1979). The reason why this equation is very suitable is due to the fact that the constants in the exponential equation have some meaning. First of all ‘p’ is the degradation which has taken place during the time ‘t’; ‘a’ is essentially similar to the very rapidly disappearing fraction because it is the intercept and highly correlated with the water soluble fraction; ‘b’ is that portion of the feed which will be degraded in time; ‘c’ is the degradation rate of the ‘b’ fraction. It follows that ‘a’ + ‘b’ (i.e. the asymptote) give some measure of the potential digestibility or the potential degradability of the feed being tested.

Degradation of small particles

Having explained the exponential equation it is now essential that we distinguish between the degradation of small and large particles. The reason for this is simply that small particles can flow out from the rumen and enter the lower gut without further reduction in particle size. As a result we can now apply another equation - namely, the rate of outflow of small particles. It will then be logical to assume that the effective degradability of small particles will depend on solubility, i.e. on the ‘a’ value and on the rate at which the ‘b’ fraction is degraded. It will also depend on the rate of outflow. We have managed to put together the two expressions in one which we call the effective degradability which is equal to:

P = a + bc
 c + k

Now, it is possible to examine the extent to which the effective degradability will vary at different rates of outflow. Figure 2a illustrates three different protein supplements which show a different pattern of degradation. One has a high ‘a’ value, very little of a ‘b’ value and a low degradability. Such illustrations are often similar to that of fish meal. Another diet illustrated has a very high ‘b’ value, a very low rate constant and a high potential degradability; another diet is between the two extremes. It can readily be understood that the effects of outflow will be different for these three different supplements. Figure 2b illustrates the typical outflow rate curve from which ‘k’ is derived. This measurement is obtained using chromiummordanted protein supplements (Elimam and Orskov 1984, in press).

The rate of outflow will have very little effect on such products as fish meal since the ‘b’ value is very low and the effective degradability will be very similar to ‘a’ (Figure 3). It will have a very large effect on protein supplements with a high ‘b’ value and a low ‘c’ value. It follows that it will have an intermediary effect on other supplements which have a high ‘c’ value combined with a high ‘b’ value. It is important that this is understood because it is not possible to use a constant correction or to use a particular incubation time to represent effective degradability.

Degradation of fibrous feeds

The same expression as mentioned before can be used to describe the degradation of fibrous feeds except that the outflow rate of small particles cannot directly be applied. The (a + b) value will give some expression for potential digestibility for the fibrous feeds. The potential degradability is however seldom reached in ruminants. Sometimes the in vivo digestibility is quite similar to degradation at 48 h. But this will depend on rumen retention time and thus also on level of intake. It is not possible to make general rules even about that. It is useful though, if in vivo digestibilities differ very much from the potential degradability, to speculate as to what the reason is.

Possibly the outflow rate is too fast to allow expression of potential digestibility. It is also possible that the rumen environment is such that it could be substantially improved and the digestibility come closer to the potential degradability. I believe that a description of degradability of fibrous feeds in the rumen environment which is optimal for cellulolysis is of greater value than is a series of measurements of digestibilities and calculated metabolisabilities obtained at maintenance energy intake. If required it is possible to make the measurements more sophisticated by describing disappearance of chemical entities such as starch, acid detergent fibre, organic matter, cellulose, hemicellulose and so on. However, in many situations the degradation of dry matter and organic matter, when we are dealing with fibrous feeds, is of the greatest value.

Calculation of exponential equation from curve fitting by eye

Description of the disappearance of protein or fibre by the exponential equation is easy to do with a computer or a “scientific” calculator; this permits the estimation of the residual standard deviation of the equation. However, if no computer facilities are available it is quite possible to caluculate the equation by simple arithmetic (Orskov et al 1980; Orskov 1982).

Figure 2

Figure 2:
Examples of graphs describing degradability of different proteins and outflow rates

Figure 3

Figure 3:
The effect of outflow on effective degradability of fish meal (F), linseed meal (L) and groundnut (G)

Figure 4

Figure 4:
Illustration of calculation of rate of degradation from disappearance curve fitted by eye

Essentially, it is necessary to describe the curve as in Figure 4. The procedure is to fit the curve to the measurements obtained. It can be seen that the intercept ‘a’ is ‘6’, the asymptote in the equation here is found to be ‘92’, i.e. (a + b) is 92% which means that ‘b’ is ‘86’ (namely 92 – 6). If we now take a value on the curve where degradation is occurring most rapidly (let us in the case here use the time 8 h) we find that ‘p’ is ‘48’.

It is now possible to describe the equation as:

e -ct = (a + b b - p)

which means that:

e-c3 = (6 + 86 86 - 48) = 0.5116

By taking the natural log on both sides of the equation, it is found that:

c = 0.084

All the constants in the equation are now known and they are in agreement with those which can be obtained slightly more accurately with the computer.

Studies to be undertaken using the technique

There are many studies which can be made including simple descriptions of different diets and so on. A few examples will be referred to.

Studies on rumen environment

It is extremely important that cellulolysis occurs at the optimum rate. It is possible to ascertain this using the nylon bag method. One can make alterations in the environment of the rumen by feeding supplements; and then study the extent to which this alteration in rumen environment has increased or indeed inhibited the digestion of cellulose. Cellulosic materials can then be incubated in the rumen and the rate of disappearance determined. For these types of studies it is often of value to wash the cellulosic materials first because the soluble material may confuse the issue.

It is possible to determine whether nitrogen of protein is limiting the rate of fibre digestion by adding urea or indeed protein supplements to the feed and determining whether it has had any effect on rates of cellulose digestion.

If materials such as sugar, molasses and starch are used in large quantities, it is possible to mix them with the feed and determine whether they have depressed cellulolysis. It is particularly important that the animals are fed at the level that is going to be applied in practice because the level of feeding will to a large extent determine the amount of easily digestible supplements that can be tolerated before cellulolysis is inhibited.

It is sometimes of value to investigate whether the rumen environment has had an effect on rates of protein degradation. It is quite often found that when cellulolysis is inhibited, so is the rate of protein degradation, particularly degradation of protein supplements of vegetable origin.

Studies on chemical treatments of roughages

The nylon bag method is very suitable for studying the extent to which a chemical treatment has caused alteration in the rate and extent of digestion of cellulosic materials. Here it is essential to ensure that the rumen environment in which the studies are made is optimal for cellulolysis. If this is not the case then the interpretation of results will be difficult and unreliable. Recently, Orskov and Kowalozyk (in press) used the technique to investigate interaction of chemical treatments (see Figure 5).

Figure 5

Effect of SO2 and NH3 treatment on disappearance of dry matter from nylon bag8

It is also possible to study the time relationship: for instance, whether with urea or ammonia treatment of straw it is necessary to keep stacks covered for one week, two weeks or three weeks before the maximum effect is obtained. Having ascertained which is the optimum chemical treatment, it is then possible to feed this material to the animals and determine whether the rumen environment is also optimal, i.e. whether further nitrogen supplementation is required; whether other supplements or trace minerals have to be added to ensure optimal cellulolysis.

Studies on protection of protein supplements

In many situations it is of interest to reduce the rate to which protein supplements are degraded in the rumen, Several chemical treatments are now available with which to accomplish this. It is possible to use the nylon bag method to determine the extent to which such methods have been successful. Here of course it must be remembered that degradation is not everything. It is also essential that the protein supplements are digestible in the intestine subsequent to the treatments with chemicals to reduce degradation.

It is also possible to determine the extent to which amino acids in the undegraded residues are similar to those of the original material. It is generally assumed that they are similar but it may not always be.

There are many other situations relating to rumen environments and to rates of digestion for which the nylon bag is an extremely powerful tool.

Use of the nylon bag technique for teaching

It is useful to conclude by pointing out that the nylon bag technique provides a very simple and effective way of teaching students the principles of rumen fermentation. It is possible for specialist students to undertake small projects using this method and in so doing gain tremendous experience in the art of feeding of ruminants.


Czerkawski, J.W. and Breckenridge, G. 1979, Br. J. Nutr. 42:217–228.

Elimam, M. E. & Orskov, E. R. 1984, Anim. Prod. (in press).

Ganev, G., Orskov, E.R. & Smart, R. 1979, J. agric. Sci. Camb. 93:651–656.

Mehrez, A.Z. & Orskov, E.R. 1977, J. agric. Sci. Camb. 88: 645–650.

Orskov, E. R. 1982, Protein nutrition in Ruminants. Academic Press:London.

Orskov, E.R., Hovel, F.D. & Mould, F.L. 1980, Trop. Anim. Prod. 5:195–213.

Orskov, E.R. & McDonald I. 1979, J. agric. Sci. Camb. 92:499–503.


M Picard1, D Bourdon2 and J Le Dividich2

I.N.R.A.1- Station de Recherches Avicoles - 37380 Nouzilly - France
I.N.R.A.2- Station de Recherches sur l'Elevage des Porcs - St Gilles - 35590 L'Hermitage - France


Digestibility tests provide useful information for the more appropriate utilization of by-products by monogastric animals, provided that they are done:
The methods described with some detail in the paper have been chosen as the simplest reliable techniques taking into consideration the goal of the seminar. These general guidelines should be improved by experience for a given type of by-product. But any selected technique must not alter the predictability of the results when the data are compared to table values for other ingredients. The aim of this type of work is to introduce the ingredient into monogastric feeding systems which are based upon linear programmed economical optimization.


“All too often in evaluating a new ingredient the general tendency is to: put it in the diet and see what it does!” (Waldroup, 1982). To improve this hasty attitude this author suggests a more comprehensive approach in four steps:

  1. Chemical analysis
  2. Acceptability by the animal: quickly testing the effects of low and high concentrations on food intake, faeces consistency and health
  3. Digestibility of energy and amino acids
  4. Comparative feeding trials.

Therefore a product with a known chemical composition will be tested for digestibility with regard to a “goal-concentration” i.e. a level of acceptability. But above all, a residue or a by-product has to be defined precisely. Zombade et al (1977) for rice bran and Lipstein et al (1982) for activated sludges gave good examples of the wide variation in composition of such products. When variability between samples of the same raw material is large it may be worthwhile to split them up into categories prior to any animal test.

Bioavailable energy and amino acids account for over 90% of feed costs. We will then focus on the methods to estimate these nutrients for poultry and pigs.

To remain clear, simplicity will always be favoured against a more exhaustive but complex description of techniques, considering the general goal of the seminar.


Informative reviews on this subject have been published in 1972 by Vohra and more recently by Sibbald (1982). The major event of the decade separating these two articles appears to be the development of rapid tests.

1. Metabolizable energy is today the only practical system for poultry.

Digestible or net energy determinations may be desirable for research purposes. Too many experimental problems prevent them from being used efficiently routinely. Furthermore, Metabolizable Energy (ME) is the worldwide accepted unit and new results can be compared to a large literature background.

2. Adult cockerels must be preferred to growing chickens or laying hens

Differences in ME determination between adults and growing birds have been occasionally shown. But the methods used are also different. The only way to prove something is to measure nutrient digestibilities. Only lipids have been consistently reported to be less digestible in young animals (Figure 1).

Figure 1

Figure 1:
Age and fatty acid digestibility (Lessire et al 1982)

Extrapolation from the adult to the young should therefore be adjusted for fat sources such as animal by-products or agro-industrial residues. With this exception the adult cockerel is a good model, easier for ME measurement than any other alternative.

3. Level of inclusion in the test diet and food intake : THE KEY POINT

The inclusion of 5% of a by-product in poultry diets may be of real practical and economical values for a developing country. Unfortunately the usual methods are not sensitive enough to estimate the ME content of a raw material precisely at a low inclusion level in the test diet. A high level of inclusion may bring about a risk of abnormal value. The acceptability test is necessary to fix a “goal concentration”. What is the significance of ME measurements done on sal seed (Zombade et al, 1979; Kaduskar et al, 1981) at 15, 30 or 60% of the test diet, when tannins limit its use to less than 3 to 5%?

Figure 1 illustrates the danger of the regression technique when various levels of inclusion of brewer's grains in a test diet lead to the extrapolation of their ME values to 100%. For regular use of 10% what is the ME value : 2750 or 2270 Kcal/kg?

These discrepancies in the Apparent ME (AME) measurement can often be explained by a depressive effect of high concentration of the ingredient on food intake. Jonsson and Mc Nab (1983) suggested a correction to adjust for these food intake variations.

Figure 2

Figure 2: ME of brewers' grains for poultry (Lopez et al 1981)

As shown by many researchers the “training technique” proposed by Farrell (1980) which restricts the test meal to one hour cannot overcome a depressive effect on food intake.

Sibbald (see 1982) advocated a force feeding method using the pure raw material. This technique avoids any problems of feed spillage in the plate or decreases in food intake. However, the digestion may be affected by some pure by-products. Ravindran et al (1983) obtained consistent results with cassava leaf for a 48 h collection period, while Ngoupayou et al (1982) measured a very low ME value for jojoba meal.

When food intake is not affected, a non linear decrease in the ME value of the meat meal may be observed with increasing concentrations (Figure 3). Such an effect must be confirmed by measurement of nutrient digestibilities (i.e. fats) before drawing a conclusion.

The suggested procedure (Annex 1) is a compromise. It will work for “goal concentrations” ranging from 10 to 30 p. 100. For lower values, critical discussion of the results and repetitions of the measurement are necessary.

Figure 3

Figure 3:
ME content of meat meals according to the incorporation level in the diet of cockerels (Lessire et al 1983)

4. Precision : a “leit motif”

Feed spillage in the plate is a major critical point. It may be limited by :

Any indicator adds variability to the results due to its own dosage and transit discrepancies (Guilloteau and Toullec 1980…). Acid Insoluble Ash appears more reliable than Cr2O3 (Schang et al, 1982).

Total collection of the droppings is more accurate than indicators if the plates are carefully and frequently (twice daily) emptied with immediate cooling of the samples (less than + 4°C). This is especially important in warm and humid climatic conditions. Feathers and scales must be cleaned out.

Fluctuations in the dry matter require constant care for both feed and faeces samples. Dry matter controls may overcome many obvious mistakes (see Annex 1).

As suggested by Campbell et al (1983) direct determinations of the gross energy contained in the raw material and in the basal diet rather than on mixed feed are advisable. The pooling of faeces of three cockerels saves calorimetry work which can be used to control the gross energy content of the feed more precisely. A balance must be observed between precision of input and output (see Annex 1).

5. Rapid versus classical tests

The classical methods use birds fed ad libitum and 3 to 5 day collection periods. They are tedious. The increased risks of error (feed spillage, variations in dry matter, etc.…) are compensated by the bulk of material analysed. Endogenous losses have a marginal effect on AME values. Rapid tests using birds either fed ad libitum (Kussaibati et al 1983) or force fed (Sibbald) involve a collection period shortened to 48 h (cf. Annex 1). The need for precision must be emphasized and corrections for endogenous losses become necessary.

Figure 4

Figure 4:
Relationship between apparent ME value (Y) and food intake (X) (Guillaume and Summers 1970)

6. TME vs AME : the endogenous factor

AME values vary with the level of energy intake (Figure 4). Sibbald (see 1982) suggests measuring the endogenous energy losses using starved birds, and to correct AME to True ME (TME). Kussaibati et al (1983) discussed the starvation procedure. As proposed by McNab and Fisher (1981) endogenous losses can no longer be ignored in the design of experiments. Its measurement permits data to be converted from one M.E. system to another taking the local conditions into account (birds, climate…).

7. Correction for nitrogen equilibrium

The assay of nitrogen is cheap and may be used for other purposes (cf. AA digestibility). Even if the correction is often marginal with mature cockerels, it adds quality to the data. Prediction of ME values for growing or laying birds can be corrected for a given nitrogen retention (i.e. = 40%) (I.N.R.A. 1984).

8. From an unknown product towards a routine determination

Discussion of the results is very necessary and should involve comparing methods, units and levels of introduction of the same raw material. Experience tends to favour an adapted, simplified procedure. At the same time the regression technique may link ME and the chemical composition of a specific product (Janssen et al 1979; Lessire and Leclercq 1983).


A by-product is even more critical in tests on pigs for at least four reasons:

TABLE 1 Diet composition for pigs: an example from China (from Aumaitre et al 1981)
 Pregnant sowsMilking sowsGrowing pigs
Compounded Feed:% DM eaten
“Rice polishings”413219
Wheat bran5919
Rice bran24157
Protein meals31214
Fresh roughage
Eishornia crassipes and Ipomea batta (leaves)16120

1. Energy system : digestible or metabolizable…BUT!…

The Digestible Energy (DE) system is simple and coherent. Raw material data may be converted from DE into ME using the following equation (Henry and Perez (1982):

ME and DE = MCal/kg ; 1 = gas losses DP = Digestible Protein (N x 6.25) in g/kg 0.07 = Coefficient assuming a 50% N retention and 9 Kcal/g of N in urine (measured).

The validity of these systems is questioned for cereal by-products and fibre rich sources. Pals and Ewan (1978), Taylor and Fisher (1979) (Figure 5) found relatively low Net Energy values for wheat by-products.

Figure 5

Figure 5:
Net/metabolizable energy ratio of wheat by-products in pigs (Taylor and Fisher 1979; Bourdon 1982; Perez et al 1981)

TABLE 2 Influence of dietary composition on the site of absorption and efficiency of utilization of ME in pigs (from Just 1979)
%in diet dry matterIleal digestibility coeff. %Faecal digestibility coeff. %% energy absorbed in the small intestineNet energy % ME
4 % C. Fiber cereals74898364
8 % " " "59817362
5 % C. Fiber straw74848865
9 % " " "63758463
5 % C. Fat animal65827963
27 % " " "69828470
16 % C. Protein68798661
33 % C. "71868354

These observations may be partly explained by the site of absorption (small intestine of hindgut) which vary according to the diet composition (Livingstone et al 1979) (Table 2).

These existing discrepancies are not large enough to justify the practical switch to a Net Energy system where experimental error may be larger than measured differences!

2. 30 to 80 kg pigs : the most “usable” test animal

Considering both the economical goal and experimental problems, piglets should be avoided. The adult sow shows varying efficiency in energy digestion depending on its physiological state (i.e. = 84% in gestation and 97% in lactation; Salmon Legagneur 1967). Again methodological difficulties limit the use of adult pigs in digestibility tests.

The age of the growing pig may be selected according to the type of ingredient studied. An older animal is more tolerant to a high fibre diet. Growth must be maintained during the test, thus only balanced diets can be used. A slight feed restriction is worthwhile (90%) i.e. under warm climatic conditions : 80–90 g of feed DM/kg LW 0.75/day.

3. Test diet and experimental design

The utilization of digestibility cages is highly recommendable. However with dry raw materials a “grab” sampling procedure using an indicator may be tried. Cr2O3 is extensively used and criticised as well! McCarthy et al (1974–1977) and Yen et al (1983) suggested Acid Insoluble Ashes as an improved alternative.

The introduction of the raw material follows 3 main procedures:

Raw materialPigsExperimentPeriods (days)level of introd.ResultsReferences
Name% D.M.Nb.L.W kgDesignTechn.Adapt. cagesAdapt. feedCollect% D.M.SampleDigest. coeff. %D.E. kcal/ kg DM
Wheat bran89452simplepure15151097-773450Le DIVIDICH et al. 1976 a)
- raw
 vigna sinensis
 - cooked

427blocksaddition107740raw cooked81


Molasses78430alternate*addition107740-843000Le DIVIDICH et al. 1974
- raw
- cooked

        B. raw

B. cooked


Le DIVIDICH et al., 1976 b) and LE DIVIDICH 1977
- raw
Arrow root
- cooked

alternate*addition107740AR. raw

AR. cooked


- raw
Sweet potato
- cooked

        SP. raw

SP. cooked


** 25
direct 50
by reg.(100) 25
direct 50
by reg.(100)
PEREZ et al., 1981
Meat and94430BlocksDilution107100,
20 10
direct 20 10 direct 20 10
direct 20

* ALTERNATE DESIGN (2 x 2 latin square)

  basal diet (BD)BD + RM
2 to 4 pigs
   collect collect
 Adaptation to cagesBD + raw material (RM)BD
2 to 4 pigs
   collect collect
 10 days7d.7d.7d.7d.

** Direct calculation: Made by differences with the pure Basals diet (level 0) for each level of introduction.Regression to 100 p. 100 is obtained using all the data for a sample. It is valid only ifthere is no interaction BD x RM (i.e. = linearity is true) cf. text.

The variability of the results decreases as the inclusion level rises (Figure 7) and is less for more digestible products. The duration of the collection period also affects variability (Figure 6). The improvement becomes marginal for more than 7 to 10 days.

The number of pigs to be used depends on variability and experimental costs. Four pigs is a compromise often found in the literature. The experimental design can limit the residual variation. A common practice is to use the pig as its own control (Table 3, foot-note). Blocks or latin squares (Velasquez et al 1969) are often used. There is a conflict between the necessity of long adaptation and collection periods, the growth of the animal, the harmful effects of long stays in cages and the desire to test many diets successively. We have tried to describe an average situation in Annex 2.

Figure 6

Figure 6:
Duration of metabolic balance experiments (pigs) (Verstegen et al 1973)

Figure 7

Figure 7:
Concentration of the raw material in the test diet and precision (Bourdon 1982; Perez et al 1981)

4. Precision is still a “leit motif”!

Most of the laboratory work is similar to that of the poultry tests and requires the same care. Some specific additional problems may arise with pigs:

5. In pig digestibility : experience is the key

Standardization is difficult in view of the extreme variety of by-products which might be fed to a pig. There is a strong temptation to abandon the animal model and use a table or a regression equation. Henry (1983) concluded a documented review on the problem of energy prediction from fibre content with this statement.

“For a given class of feeds the DE value may be estimated (…) from the crude fibre content, but a specific correction factor is needed in order to take into account changes in cell wall constituents”.

The specificity of each by-product on residue is a good reason why the animal test must be performed.


The use of simplified (cereal + soyabean meal) diets has been useful in understanding the general principles of amino acid (AA) nutrition. Formulation of diets by linear programming is efficient for the adjustment of the composition of simple diets using the AA composition of raw materials. However when by-products are used at significant levels there is a strong need for bio-available AA data… We do not rely on gross energy, do we?

1. The practical path through a “jungle” of tests

A number of criteria (PER, NPR, NPU, BV, CS…) have been reported for appreciating the protein value of a feed (see e.g. = the Midlands conference 1977). The results are either partial (Hegedus et al 1983) or too dependent on the technique itself (Coon et al 1980; Bender 1982) and above all the results are difficult to use in practical formulation.

AA chemical composition on the one hand, and AA digestibility on the other hand explain most of the variation in bio-availability between protein sources (Kies 1981; Henry 1983). Although some AA derivatives may be absorbed but remain unavailable at the metabolic level (Larbier and Blum 1979), a review of a wide range of techniques (Larbier 1982) leads to the conclusion that digestibility is the most practical and accurate method.

Twelve years of research on the subject have led us to the same conclusion : digestibility of AA is a necessary first step to improve the efficiency of protein nutrition in monogastric animals.

2. The AA digestibility “CHALLENGES”

3. Towards a simplified but efficient procedure

The following proposals are based upon recent results obtained at AEC*. Over 300 complete digestibility tests have been completed to date (Picard et al 1983; Darcy 1984). Further information may be obtained directly from the authors on request.

Methodology is the goal of this programme. We compared the digestibility of AA of the same raw material samples in pigs and rats (at the ileal level) and in intact or caecectomised cockerels. Test procedures differed: force feeding on cockerels, meal feeding in pigs, ad libitum feeding in rats.

Using 7 common ingredients we observed no essential differences on the average (133 tests) between the profiles of true digestibility of AA in the four animal models (Graph 8). A similar situation is found (Figure 9) when protein free diets are fed, concerning the composition of endogenous losses of AA. These two graphs suggest a relatively similar pattern of AA digestibility among monogastrics. Predictability from one species to another is confirmed when each raw material is considered separately.

* A.E.C. Research centre 03600 - Commentry, France

Figure 9

Figure 9:
Average results with protein-free diets (40 tests)

Figure 10

Figure 10:
Composition of the ileal juice

Figure 11

Figure 11:
Composition of the ileal juice

Table 4 Average ileal juice composition in AA (% of the sum of the analysed AA)
p. 100 of the sum of the analysed AA
 Protein free diet = endogenousVegetable protein standards*
Amino Acidcockratpigcockratpig
Aspartic Ac9.
Glutamic Ac14.314.112.813.714.215.015.314.8

* For animal protein by products : the raw material profile is closer to theileal profile. Then the RM profile may be used instead of the standard

We do not pretend that the accuracy of the complete procedure will be attained. But we think that this simplified test, which can be run together with a ME determination, leads to improved estimation of the economic value of a by-product in foumulation (providing of course that such a residue does contribute a significant part of the protein content of the diet).


“The farthest trip starts by a first step”

Lao Tseu

Digestibility tests provide useful information for more appropriate utilization of by-products by monogastric animals, provided that they are run:

For a given type of by-product, experience should lead to a constant improvement of methods and simplification, rather than reliance on a fixed procedure. But a selected technique must not alter the predictability of the results when the data are compared to table values for other ingredients. The “end of the trip” is to introduce the ingredient into a linear programming process on a coherent technico-economical basis.

When we compared the amino acid composition of the excreted ileal juice, we found no significant differences whatever raw material was fed or whatever animal species was used. Therefore the large discrepancies which exist between the AA digestibility profiles of the 7 usual raw materials might be explained by the AA content of these raw materials and the amount of ileal juice excreted.

With respect to by-products, these may be compared to a “standard curve” (STD on Graphs 10 and 11) of average ileal juice composition in AA for the 7 usual raw materials. One can see on Figure 10 that wheat bran yields the same profile. Animal by-products (Figure 11) show some differences for proline, glycine, alanine (meat and bone meals) or serine, proline and cystine (feather meal) suggesting that part of these proteins may completely escape digestion.

A consistent AA profile of the ileal juice (with adjustments for animal meals) is of great importance for simplifying tests on AA digestibility. The only critical point which remains is the measurement of the average amino-nitrogen excreted at the ileum level for a given raw material.

Figure 12 illustrates the close linear relationship existing between the apparent digestibility of nitrogen and that of the sum of the analysed AA (70 points per line). For cockerels faeces have to be treated using the Terpstra and Dehart (1974) technique for uric acid elimination.

Figure 12

Figure 12:
Regression between AA and N apparent digestibility

The prevailing conclusion to date is that for AA digestibility the major critical step is to measure the global amount of proteic material excreted at the ileum end.

Thus, a simplified procedure, may be recommended for AA digestibility (Annex 3):


Aumaitre, A., Brette, C., Dagorn, J. and Legault, C. 1981. Compte-rendu de la mission effectuée en République Populaire de Chine du 5 au 22 novembre 1981. Document interne - I.N.R.A. (in French), 49 pages.

Bender, A.E. 1982. Evaluation of protein quality. Methodological considerations. Proc. Nutr. Soc., 41 (3), 267–277.

Bourdon, D. 1982 . Essai de digestibilité des farines de viandes chez le porc. Résultats non publiés (in French).

Campbell, G.L., Campbell, L.D. and Blair, R. 1983 . Calculation of metabolizable energy for ingredients incorporated at low levels into a reference diet. Poultry Sci., 62, 705–707.

Coon, C.N., Johnston, J. and Kelley, K. 1980 . The effect of temperature on feed consumption and protein efficiency ratio of broiler chicks fed two meat and bone meal samples of different protein quality. Nutr. Rep. Int., 22 (6), 907–915.

Darcy, B. 1982 . Transit et digestion dans l'intestin grêle chez le porc. In Proc. Digestive Physiology in the pig. Versailles 27–29 Octobre 1982. Ed. I.N.R.A. Publ., 1982, pp. 45–59 (in French).

Darcy, B. 1984 . Recent development in amino acid digestibility measurement. In Proc. Symposium on amino acid nutrition N.F.I.A. Chicago - April 30th - May 3rd, 1984 (in press).

Darcy, B. and Rerat, A. 1983 . Digestion and absorption of the hydrolysis products in the small intestine of the pig. In Proc. IVth Int. Symp. Protein metabolism and nutrition. Clermont-Ferrand, 5–9 Septembre 1983. Ed. I.N.R.A. Publ., 1983, pp. 233–244.

Farrell, D.J. 1980 . The “rapid method” of measuring the metabolizable energy of feedstuffs. Feedstuffs, 52 (45), 24–26.

Fuller, M.F. and Cadenhead, A. 1969 . The preservation of faeces and urine to prevent losses of energy and nitrogen during metabolism experiments. In Energy metabolism of farm animals. Ed. Oriel Press Ltd. Newcastle-upon-Tyne, England, pp. 455–460

Fuller, M.F. and Livingstone, R.H. 1982 . In: Annual report of studies in animal nutrition and allied sciences. Ed. Rowett Research Institute, Scotland, Vol. 39.

Guillaume, J. and Summers, J.D. 1970 . Maintenance energy requirement of the rooster and influence of plane of nutrition on M.E.. Can.J.Anim. Sci., 50, 363–369.

Guilloteau, P. and Toullec, R. 1980 . Digestion of milk protein and methanol-grown bacteria protein in the pre-ruminant calf. 1. Kinetic and balance in the terminal small intestine and faecal balance. Reprod. Nutr. Develop., 20 (3A), 601–613.

Hegedus, M., Bokori, J. and Hadobas, K. 1983 . Protein quality of mixtures of animal protein meals. An. Feed Sci. and Techn., 9 (2), 79–87.

Henry, Y.M. 1983 . Principles of protein evaluation in pig feeding. In Proc. 34th Annual Meeting of the EAAP, Madrid, Spain, 3–6 October, 1983.

Henry, Y.M. and Perez, J.M. 1982 . Les systèmes d'évaluation de l'énergie dans l'alimentation du porc. Les dossiers de l'élevage, 5 (1) and (2) (in French).

I.N.R.A. 1984 . L'Alimentation des Animaux monogastriques: porcs, lapins, volailles. Ed. I.N.R.A. Publ. Versailles, France. 296 pages (in French).

Janssen, W.M.M.A., Terpstra, K., Beeking, F.F.E. and Bisalsky, A.J.N. 1979 . Feeding values for poultry. Ed. Spelderholt Institute for poultry research. Beekbergen - The Netherlands, 59 pages.

Jonsson, G. and McNab, J.M. 1983 . A comparison of methods for estimating the metabolizable energy of a sample of grass meal. Br. Poultry Sci. 24, 349–359.

Just, A. 1979 . Influence of dietary composition on site of absorption and efficiency of utilization of metabolizable energy in growing pigs. In Energy Metabolism. L.E. Mount. Ed., Butterworths, London, England, pp. 27-30.

Kaduskar, M.R., Netke, S.P. and Kolte, G.N. 1981 . Effects of high inclusion rates of salseed meal in the diet of chicks. Br. Poultry Sci., 22, (6), 503–509.

Kies, C. 1981 Bioavailability: a factor in protein quality. J. Agric. Food Chem., 29 (3), 435–440.

Kussaibati, R., Prevotel, B. and Leclercq, B. 1983a . A simplified rapid method for the determination of real apparent metabolizable energy values of poultry foods. In Proc. 4th European Symposium on poultry nutrition. Tours, October 17th–20th, 1983, France (in press).

Kussaibati, R., Prevotel, B., and Leclercq, B. 1983b . Factors affecting endogenous energy losses in chickens - assessment of the methods of estimation. Nutr. Rep. Int., 27 (20) 221–230.

Larbier, M. 1982 . Les méthodes bioloques de mesure de la disponibilité des acides aminés chez le poulet. In La biodisponibilité des acides aminés. C.A.A.A.Ed. I.N.A. Paris. pp. El-E38 (in French).

Larbier, M. and Blum, J.C. 1979 . Valeur nutritionnelle de la -N-acetyl-lysine comparée à celle de la L-lysine monochlorhydrate chez le poulet. Arch. Geflugelk., 43, 238–241 (in French).

Le Dividich, J. 1977 . Feeding value of Canna Edulis roots for pigs. J. of Agriculture of the University of Puerto Rico, 61 (3), 267–274.

Le Dividich, J., Seve, B., Canope, I. and Kempf, H. 1974. Essai d'utilisation de fortes proportions de mélasse de canne à sucre dans les rations du porcelet sevré à 5 semaines et du porc en croissance-finition. In: Journées Rech. Porcine en France (1974). Ed. I.N.R.A.- I.T.P. Parish 199–207 (in French).

Le Dividich, J., and Seve, B. 1975 . Valeues énergetique et azotée de Vigna Sinensis pour le porc: effet de la cuisson et comparaison avec le tourteau de soja. Ann Zootech., 24 (1), 13–20 (in French).

Le Dividich, J., Canope, I., Hedreville, F. and Despois, E. 1976a . Possibilités d'emploi aux Antilles du son fin de blé dans l'alimentation du porc en finition. In Journées Rech. Porcine en France, Ed. I.N.R.A.-I.T.P., Paris (1976), pp. 25–28 (in French).

Le Dividich, J., Aumaitre, A.L. and Seve, B. 1976b . Utilization of tropical plants rich in starch for pig feeding. in Proc. First Int. Symposium Feed Composition, July 11–16, 1976, Ed. Utah State U., Logan, U.S.A.., pp. 125–132.

Lessire, M., Leclercq, B. and Conan, L. 1982 . Metabolizable energy value of fats in chicks and adult cockerels. Animal Feed Sci. and Technology, 7, 365–374.

Lessire, M. and Leclercq, B. 1983 . Metabolizable energy content of meat meal for chicken. Arch. Geflugelk., 47, 1–3.

Lessire, M., Leclercq, B., Conan, L. and Hallouis, J.M. 1983 . Etude méthodologique de l' énergie metabolisable des farines de viande. In proc. Workshop on Animal Protein and fat. Ed. N.R.A. - Paris, 27–28 Oct. 1983 (in French).

Lipstein, B., Kary, S. and Hurwitz, S. 1982 . The nutritional value of activated sludges for poultry. Nut. Rep. Int., 25 (5), 829–836.

Livingstone, R.M., Atkinson, T. and Baird, B.A. 1979. The effect of wheat bran in the diet of growing pigs on feed digestibility sites of nutrient absorption and nitrogen retention. Animal Prod., 28 (3), 437 (58).

Lopez, J.D., Carmona, J.F. and Pascual, J.L.M. 1981. Evaluation of brewer's dried grains in the diets of laying hens. Animal Feed Sci. and Technology, 6 (2), 169–178.

Low, A.G. (1980). Nutrient Absorption in pigs, J. Sci. Food Agric., 31, 1087– 1130.

McCarthy, J.F., Aherne, F.Z. and Okai, D.B. 1974 . Use of HCl insoluble ash as an index material for determining apparent digestibility with pigs. Can. J. Anim. Sci., 54, 107–109.

McCarthy, J.F., Bowland, J.P. and Aherne, F.X. 1977 . Influence of method upon the determination of apparent digestibility in the pig. Can,. J. Anim. Sci., 57, 131–135.

McNab, J.M. and Fisher, C. 1981 . The choice between apparent and true metabolizable energy systems. Recent evidence. in Proceedings 3rd European Symposium on poultry nutrition. Edinburgh 26 –29th Oct. 1981. Ed. PRC Roslin Scotland, pp. 45–55.

Midlands Conference 1977. New concepts for the rapid determination of protein quality. Nut. Rep. Int., 16 (2), 157–226.

Ngou Ngoupayou, J.D., Maiorino, P.M. and Reid, B.L. 1982 . Jojoba meal in poultry diets. Poultry Sci., 61 (8), 1692–1696.

Pals, D.A. and Ewan, R.C. 1978. Utilization of the energy of dried whey and wheat middlings by young swine. J. of Anim. Sci., 46 (2), 402–408.

Perez, J.M., Castaing, J., Grosjean, F.M, Chauvel, J., Bourdon, D. and Leyillet, M. 1981 . Valeur énergetique de deux types de manioc et utilisation comparée dans les régimes du porc en croissance et du porcelet: synthèse des résultats d'une étude concertée I.N.R.A.-I.T.C.F.-I.T.P. in Journées Rech. Porcine en France (1981). Ed. I.N.R.A.-I.T.P. Paris, pp. 125–144 (in French).

Picard, M., Bertrand, S., Duron, M., and Maillard, R. 1983 . Comparative digestibility of amino acids using five animal models. In Proc. 4th European Symposium on Poultry Nutrition. Tours, Oct. 17th–20th 1983. France (in press).

Picard, M., Bertrand, S., Genin, F. and Maillard, R. 1984 . Digestibilité des acides amines; intérêt de la technique du shunt ileo-rectal chez le porc. J. Rech. Porcine en France, 16, 355–360 (in French).

Ravindran, E.T., Kornegay, L.M., Potter, L.M., Webb, K.E., and Parsons, C.M. (1983). True metabolizable energy values of cassava tuber and leaf meals for poultry. Trop. Agric. (Trinidad), 60 (2), 82–84.

Salmon-Legagneur, E. 1967 . Nutrition of the lactating sow. in Proc. of Symposium on the nutrition of sow. Nottingham, England, 1967, pp. 43–51.

Schang, M.J., Sibbald, I.R. and Hamilton, R.M.G. 1982 . Comparison of two direct bio-assays using young chicks and two internal indicators for estimating the metabolizable energy content of feeding stuffs. Poultry Sci. 62 (1), 117–124.

Sibbald, I.R. 1982 . Measurement of bioavailable energy in poultry feeding stuffs: a review. Can. J. Anim. Sci. 62 (4), 983–1048.

Tanksley, T.D. and Knabe, D.A. 1982 . Researchers outline amino acid digestibilities of some high protein feedstuffs and use in swine diet formulation. Feedstuffs, 54 (53), 16, 29–31.

Taylor, A.J. and Fisher, C. 1979 . Thermogenic effect in pigs of diets containing wheat or wheat feed. in Energy metabolism. L.E. Mount. Ed. Butterworth, London, England, pp.31–35.

Terpstra, K. and Dehart, N. 1974. The estimation of urinary nitrogen and faecal nitrogen in poultry excreta. Z. Tierphysiol. Tierernaehrg. U. Futtermittelkde, 32 (6), 306–320.

Terpstra, K. and Janssen, W.M.M.A. 1975 . Methods for determination of the metabolizable energy and digestibility coefficients of poultry feeds. Rep. N0101.75 Nutrition Dep. Spelderholt Inst. Poult. Res., Beekbergen, The Netherlands.

Velaquez, M., Ly, J. and Preston, T.R. 1969 . Digestible and metabolizable energy values for pigs of diets based on high-test molasses or final molasses and sugar. J. of Anim. Sci., 29 (4), 578–580.

Verstegen, M.W.A., Mount, L.E. and Close, W.H. 1973 . The optimum duration of metabolic balance experiments with groups of pigs. Proc. Nutr. Soc., 32 (2), 72A.

Vohra, P. 1972 . Evaluation of metabolizable energy for poultry. World Poultry Sci. J., 29 (2), 204–214.

Waldroup, P.W. 1982 . Determining the nutritive value of new improved ingredients. Feedstuffs, 54 (42), 38–43.

Yen, J.T., Tess, M.W., Pond, W.G. and Dickerson, G.E. 1983 . Digestibility and metabolism of dietary nitrogen and energy in contemporary, genetically lean and obese pigs as estimated by total fecal collection and acid insoluble ash. J. of Anim. Sci., 56 (2), 426–430.

Zebrowska, T. 1978 . Determination of available amino acids in feedstuffs for monogastrics. Feedstuffs, 50 (53), 16–17, 43–44.

Zombade, S.S., Lodhi, G.N. and Ichhponani, J.S. 1977 . A note on reassessment of metabolizable energy content of rice bran for poultry. Indian J. Anim. Sci., 47 (1), 51–52.

Zombade, S.S., Lodhi, G.N. and Ichhponani, J.S. 1979 . The nutritional value of salseed (shozea robusta) meal for growing chicks. Br. Poultry Sci., 20 (5) 433–438.

Annex 1

The poultry method :the ME route of a sample S (knowing X is the “goal concentration” :i.e. acceptable by poultry in the feed)

1. -OFFICE : Identification - origin and characteristics - Timing and Protocie

Test diets


* only if force feeding

  1. Homogeneisation of sample S
  2. Preparation of a Basal mix (B) i.e = pure maize
  3. Assays on S and B :
  4. Gross energy of the diets will be calculated from measured values for S and B Controls of the diets may be performed for humidity, N and Gross Energy and if a consistent difference appears, test diets must be re-manufactured.



Annex 2: The pig method: the ME/DE route of a sample S (X is the “goal concentration” acceptable for a pig feed)

1.- OFFICE : Identification - origin and characteristics - Timing and Protocole


  1. Homogeneisation of sample S
  2. Proparation of a Basal Diet (B) rich and balanced (growth)
  3. Assays on S and B = Dry matter (2 rep.) - Nitrogen (3 rep.) - Gross Energy (5 rep.)


Test dietSB
(4)(2x)(1-2x) (if possible)


S and B will be given separately and mixed in the pig feeder.

A constent laboratory control of Dry matter and Nitrogen content is nacessary during the test.

× must be ≥ 30 p. 100 of lngested dry matter

3. - CAGES

DILUTION technique

4–8 pigs/treatment - Block design

ADDITION technique

4–8 pigs/treatment - Alternate Latin square design

7d. adapt. 10 days adapt.cages to feed 7d. collect (ct. table 3)
ANTI PARASITE TREATMENT 2 MEALS + 1 WATER MEAL PER DAY 2 periods : B and B + S, for each pig Constant Intake = 90 g/kg L.W0.75/day B= constant Intake with or without S Control of feed refusals and feed humidity (and N content for Wet products) Collection of faeces twice daily pig per pig - weighing and storage - 18°C


5. -OFFICE :

D.E. of S is calculated directly for each dilution (diets 2, 3, 4) considering the energy value of B (diet 1). The result are compare and if they do not differ extrapolated to 100 p. 100 by regression

D.E of is calculated is calculated by difference between the B-period and the B + S period.


Annex 3:

Simlified procedure for AA digestibility

1. -RAW MATERIAL SAMPLE S simple methodcompl. test
• Analysis of Its Nitrogen concentration$ In  
Than either - Amino acid analysis of SN2.2652.265
  -Adjustment of a table value*AA 12.2
• On normal cockerels (see produce anneze 1)Bran g.4848
• Measurement of nltrogen excreted (after uric acid elimination)N mg10871087
• For a protein free control - endogenous lossesAA mg-5860
• For S fed pure or diluted with starch (to reach 15 p.100 crude protoin)LYS. mg264261
• Apparent ntrogen digestibility = A.N.DN mg347347
  (uric ec-) 
• Apparent AA digestibility = A.A.A.D.AND%68.168.1
A.A.A.D. = 0.81 A.N.D + 0.168 (ct. graph 12)   
• Apparent (i.e.) lysine digesibility = A.LYS.DA.LYS.D%  
LYS excreted = EM Ingested × (1 - 0.81 A.N.D. - 0.168) × 0.073(1)54.6D54.1
  (1) for AA Ingested :
  5860 mg
0.073 : coefficient from table 4 -(2) for AA Ingested :
∑ AA IngestedD : may be different from N × 6.25 for some raw materials.N × 6,25
4. -TRUE AA DIGESTIBILITY (necessary for rapid tests)mg/24th350278
.Endogenous AA losses = EAAmg/48h700556
EITHER AA value from analysisE.LYS  
or Endogenous N (uric ecid-) measured × 6.25 × 0.8   
or 350 mg/cock/24 h(1)69.9  
.EAA can be calculated from tthe composition given table 4.T.LYS.D 66.3
.true AA digestibility = T.M.D   
  Underlines date are
  measured. Other are
where AA excreted = coeff. table 4 × [sum; Ingested × (1 - 0.81 A.N.D - 0.168)]calculated.

Previous Page Top of Page Next Page