Feed Assessment:
Chemical Composition of Feeds

The limitations associated with animal assessments of feed quality mean that a number of laboratory estimates of the chemical composition of feeds have been developed, to provide a clearer description of what nutrients are actually provided by a feed. The main series of chemical analyses that are performed are called "proximate analysis", and this seeks to estimate the different components of a feed. (Note: This analysis can accurately measure content of nutrients in a sample of feed, but cannot measure the amount that will be utilized by an animal). The proximate analysis of feeds is based on the following description.

Moisture Content Dry Matter Content Ash Crude Protein Degradable Protein Digestible Crude Protein Digestible Undegraded Protein Crude Fibre Ether Extract (Fat content) Nitrogen-free Extract References and Further Reading

The moisture or water content of the feed is a key nutrient that is often neglected but is frequently limiting particularly in tropical situations and especially in lactating animals. Much of the animal’s water is likely to come from the feed, particularly when the animal is grazing or browsing lush vegetation.

Once harvested, feeds with a high moisture content are liable to spoil quickly, mostly from fungal contamination. The moulds and more particularly the toxins that are produced by many moulds make the feed unpalatable and can cause illness or even death to both the animals and people handling the feed. On the other hand, very dry feeds, while being stable during storage, are less palatable for the animal and also increase the animal’s requirement for water.

This component of the feed contains all the nutrients (and anti-nutritional factors) in the feed except for the water contained within the feed. They dry matter content of a feed is estimated by weighing a sample of the feed, putting it an oven to dry it (and evaporate off all the moisture in the feed), and then weighing it again. The difference in the two weights before and after drying is assumed to be the weight of water in the feed, while the feed that remains after drying contains all the dry matter present in the feed.

The dry matter content of the feed (g/kg fresh weight) is estimated from:

Dry matter content (g dry matter/kg fresh feed) = (Dried weight/Fresh weight) x 1000.

Typically, fresh grasses consist of between 150 and 200 g/kg dry matter (DM), leaves from fodder trees and silage would consist of 200 to 350 g/kg DM while cereal grains would be 850 to 900 g/kg DM.

Although water is an extremely important nutrient, and it is essential that the animal receives enough of it (either from the feed or separately as water), it should be remembered that the concentration of other nutrients will be lower in very moist feeds, and so more of the feed will need to be fed to provide adequate amounts of those nutrients.

The ash content of the feed contains all the minerals in the feed, but can also contain any soil contaminants associated with the feed as well.

The ash content of the feed is determined by placing a weighed sample of dried feed in a furnace and heating it to 500 0C, typically overnight. At this temperature, all the organic materials in the feed (proteins, carbohydrates, fats and vitamins) are burnt away, leaving just the mineral residue. This residue is then weighed and the ash content of the feed (g ash/kg feed DM) is calculated from:

Ash content of the feed = Weight of ash (g)/ weight of dried feed (kg)

The ash content of a feed says nothing about the quality of the feed’s mineral content and other, much more sophisticated and expensive tests must be done to determine how much of different minerals (potassium, phosphorus, copper, zinc, manganese etc.) is provided by the feed.

The content of organic matter can be divided into the following categories:

• Protein
• Fibre
• Fats
• Vitamins and Minerals

The crude protein content of a feed is estimated by measuring the amount of nitrogen (N) in the feed, and then assuming that:

• All the nitrogen present in the feed is in the form of protein
• All proteins consist of 16% N.

The crude protein content of the feed (g CP/kg feed DM) is then calculated from:

Crude protein content = N content (g N/kg feed DM) x (100/16)

or

Crude protein content = N content (g N/kg feed DM) x 6.25

The two assumptions made in calculating crude protein content are not entirely true, but are reasonable approximations. For ruminant animals (sheep and goats), it is of little importance anyway, since what the animal needs is a supply of nitrogen from which the rumen micro-organisms manufacture protein that is then used by the animal. However, N that is not protein, that escapes degradation in the rumen, cannot be used as a source of protein by the animal.

For pigs and poultry, however, this estimate of protein content is of limited value as what the animals actually require is a supply of amino acids (the building blocks of a protein). Obviously, the higher the crude protein content of a feed, the more amino acids are likely to be supplied by it, but without some estimation of the digestibility of the protein and (ideally) some estimation of the supply of essential amino acids (see section on protein quality for pigs and poultry), the protein content of a feed can be misleading. Leather, for example, has a very high protein content but is almost completely indigestible and so of no use as a protein supplement for livestock.

Estimation of Protein Content

The traditional way to estimate the N content (and thereby the crude protein content) is to digest (dissolve) the feed in acid. This involves boiling the feed in concentrated sulphuric acid until it has completely dissolved. The nitrogen in the solution is then converted to ammonia, which is distilled out of solution. The amount of acid that is then needed to neutralize this ammonia is measured, and from this the amount of nitrogen that was present in the feed is calculated.

This method is relatively simple, but it is time consuming and uses a number of hazardous chemicals. Other techniques have therefore been developed, either to measure the N content in the feed directly, or to predict the N content of the feed by measuring something else, and then relating it to measured N content. These predictions are then used to avoid the necessity of measuring N directly. This is the basis for measuring the near infrared reflectance spectrum (NIRS) of a feed, but again, this is a technique that requires extremely expensive equipment that is not usually available in the tropics.

Degradable Protein

When a sheep or goat eats any protein, it first goes into the rumen, where it is attacked by the rumen micro-organisms. (See under Nutrition for more information). How much of the protein gets degraded by the micro-organisms depends on how fast the protein is broken down by them, and on how fast feed leaves the rumen. If the feed is broken down quickly, and feed flows out of the rumen only slowly, then most if not all of the protein will be broken down (degraded) in the rumen. If the feed is broken down slowly, and feed flows out of the rumen more quickly, then little if any of the protein will be degraded.

Degraded protein will only be of any use to the animal if the rumen micro-organisms are able to build it up into microbial protein. This microbial protein will then eventually leave the rumen and be digested, absorbed and used by the animal. The rumen micro-organisms will only be able to use degraded protein to make microbial protein if they have enough energy. They will get this energy from fibre (if it can be degraded), but also from sugars and starch in the diet. They will not get any energy from fat or oil in the diet, as micro-organisms are unable to ferment fat to any great extent.

Estimation of Degradable Protein

The common way to estimate how much and how quickly a protein will degrade in the rumen is to take samples of the feed and put them in porous bags. These are placed in the rumen of an animal (usually sheep, cow or buffalo) that has been operated on so that it has a hole (with a cap on it) in its rumen. The bags are taken out at different times, and the amount of protein that is left in the feed that is still left in the bag is measured. From the amount of protein that is left in the bag at different times, the rate (and extent) of protein degradation can be calculated.

Protein degradability is usually expressed using three terms, a, b and c.

a is the proportion of protein that is almost immediately degraded, and so is likely to be degraded in the rumen no matter what the outflow rate is.

b is the proportion of protein that is potentially degraded, depending on the relative rates of degradation and outflow.

c is the rate of degradation of b.

To calculate the proportion of protein that will be degraded, you also need to know the outflow rate from the rumen (k). For most tropical livestock, being kept at or near maintenance, this is likely to be 0.02/h. In other words, 2% of the material in the rumen will flow out into the animal’s lower gut every hour.

The proportion of a feed’s protein that is degradable can be calculated from the equation:

Degradability = a + bc / c + k  

and since usually k = 0.02

Degradability = a + bc / c + 0.02

The amount of degradable protein in a feed can be estimated by multiplying its degradability by its crude protein content.

Protein digestibility for pigs can be estimated by feeding them known amounts of the feed under investigation and then collecting all the faeces they produce. The feed and faeces are then analysed for protein, and by knowing how much protein has been fed, and how much protein has been excreted, the digestibility of the protein is calculated from:

Since this method is expensive and time consuming, laboratory tests have been developed to predict protein digestibility. The feed can be mixed with digestive enzymes (pepsin and pancreatin, which are available commercially). These will digest the digestible part of the protein, and the feed/enzyme mixture can then be filtered. The digested protein, being by now soluble, will be filtered off with the enzymes. The undigested protein will remain and can be measured and from this, the protein digestibility can be estimated.

This method is also expensive and uses chemicals that can be hard to obtain for some laboratories. A good prediction of protein digestibility can also be achieved (for tropical forages) by estimating the proportion of protein (N) that is water soluble. This is done by putting the sample of feed through three 30 min cold water wash cycles in a washing machine, and then measuring the amount of N that is left in the residual material.

Digestible Undegraded Protein

The protein that is not degraded in the rumen may be digested and absorbed by the animal. This protein is called digestible, undegraded protein (DUP). There are times (particularly in early lactation) when animals may need some DUP to meet all their needs for protein, if the rumen micro-organisms are not able to make enough protein for the animal.

To estimate how much undegraded protein is supplied by a feed, subtract the degradable protein content away from the feed’s crude protein content:

Undegraded protein = Crude protein – Degradable protein

Not all of this undegraded protein will be digested by the animal. With most feeds, however, 95-98% of the undegraded protein will be digested by the animal and if there is no other information available then this is probably a reasonable estimate to use.

To estimate the amount of indigestible protein there is in a feed, a feed might be heated with acid detergent, and the nitrogen content of the residue determined. This gives an estimate of the feed’s ACID DETERGENT INSOLUBLE NITROGEN (ADIN) content, and is taken to be an estimate of the N fraction that will not be digested in the gut.

To estimate the indigestible protein content, therefore, the ADIN content is multiplied by 6.25 (see estimation of crude protein for an explanation).

Indigestible protein (IP) = ADIN x 6.25.

The digestible, undegraded protein content of the feed may then be calculated from:

DUP content = CP content – Degradable protein content – IP

(Where CP is crude protein).

This is an estimate of the indigestible (to pigs and poultry) and either indigestible or only slowly digestible (to sheep, goats and other ruminants) fraction of the feed. However, it is a poor estimate, and most feed assessments would use other tests to estimate the fibre fraction of the feed, with different tests being applied for ruminant and monogastric animals.

Estimation of Crude Fibre

A fat free sample of the feed is boiled with first acid and then alkali of defined concentration. It is then filtered, dried and weighed, and the weight of the residue (as a proportion of the original sample weight) is defined as the crude fibre content of the feed. The boiling with acid and alkali is supposed to remove the digestible fractions of the feed (protein, starch, sugars, vitamins and minerals) and leave the fibre fraction (the cell walls of plant products). However, variable proportions of the cell walls are dissolved during this extraction process, and so this estimation of fibre is imprecise. This also has consequences for the estimation of the starch-containing component of the feed, which in the case of cereals is the most important component of the feed (see the section on Nitrogen Free Extract). The large and unpredictable errors associated with the estimation of crude fibre (and thereby nitrogen free extract) have therefore resulted in alternative methods of fibre characterization being developed. Estimates of crude fibre contents of feeds should be treated with the utmost caution.

The detergent system of fibre analysis

This system was developed by Peter Van Soest, and so is also referred to as the Van Soest system. It was originally developed to estimate the fibre content of ruminant feeds. A sample of the feed is boiled with detergent, which is designed to dissolve the protein, sugars, oil and any other material in the feed apart from the cell walls. The milder detergent is ‘neutral detergent’, and the residue from this analysis is ‘neutral detergent fibre’ (NDF), which comprises all the cell walls. More stringent conditions, using acid, give ‘acid detergent fibre’ (ADF), which is an estimate of the less digestible fibre component. More stringent conditions still give ‘acid detergent lignin’ (ADL or lignin). This is supposed to be an estimate of the indigestible fraction of the feed.

The analyses are quite robust, and give a useful estimate of the fibre content of the feeds. However, the estimate for lignin is poor. Even in samples of tree fodder, the lignin content is low, and the errors involved in estimating the lignin content of the feed are often greater than the lignin content itself. Another disadvantage of these analyses are that they are relatively expensive to do, and use a range of chemicals that are difficult to obtain in some countries and/or are hazardous to handle.

Estimation of non-starch polysaccharides (NSP)

This estimation of fibre content (dietary fibre) was developed for use in human nutrition, but is more suitable for pigs and poultry than the detergent system. It is defined as the proportion of food that is not digested by enzymes secreted in the (human) gut, but which may be digested by micro-organisms in the gut. A number of different methods have been developed to estimate NSP as the definition of NSP has changed (to include or exclude different chemicals). The official method of analysis involves the use of enzymes to digest the protein and starch in the feed. The NSP fraction is then precipitated with 95% alcohol and weighed. Earlier methods (such as the ‘Englyst’ method) will give lower estimates of fibre content because Englyst NSP does not include ‘resistant starch’, which is now included in the definition of ‘dietary fibre’.

 

This is a measure of the fat or oil (lipid) content of the feed. Fats and oils are extremely rich sources of energy, although because they impede microbial fermentation, ruminant diets should be limited to about 4% fat.

Estimation of fat

The sample of feed is put into a porous ‘thimble’ made of parchment and plugged with cotton wool. It is then put into an apparatus (called a Soxhlet apparatus) that allows petroleum ether to be continuously added to the apparatus, siphoned off and re-evaporated so that it returns to the apparatus. While in the apparatus, it soaks the feed, and the oil in the feed then dissolves into the petroleum ether. After the ether has been siphoned off, it drops down into a flask. As the ether evaporates again, it leaves the oil in the flask. After several hours of this, the amount of fat in the feed is determined either by the increase in weight of the flask, or in the decrease in weight of the feed.

This estimation is prone to errors for most typical livestock feeds, as the oil content is so low that it is very difficult to accurately weigh the amount of oil that has been collected (or has been lost from the feed). In addition, included in the oil content are all compounds that are soluble in petroleum ether, and this includes waxes, pigments and fat soluble vitamins. However, as a technique it is reasonably simple (if rather expensive) to do and does provide a reasonable estimate of the fat (or oil) content of the feed.

This fraction consists of such things as the water soluble vitamins, and also any other feed component that has not been accounted for elsewhere, but it principally comprises of starch and sugar (in other words, apart from the oil, the main energy sources in the feed). It often (certainly in the case of cereals) constitutes the greatest proportion of the feed. It is therefore a crucially important feed fraction, but the method used for assessing it is extremely poor and for this reason a number of techniques have been developed to improve the estimation of, particularly, the starch and sugar fractions of the feed.

Estimation of NFE

The main problem with the estimation of NFE is that it is not directly determined, but only calculated by difference. The NFE content of the feed is that part of the feed that is not accounted for by ash, crude protein, ether extract and crude fibre, and is calculated using the equation:

NFE (g/kg DM) = 1000- (ASH+CF+EE+CP)

Where ASH, CF, EE and CP are, respectively, the ash, crude fibre, ether extract and crude protein contents of the feed expressed in terms of g/kg DM.

Any errors in the estimation of ash, crude protein, crude fibre and ether extract will therefore be included in the estimation of NFE.