Chapter 1: INTRODUCTION

11.2 SUPPLEMENTATION STRATEGIES

11.2.1 Supplementing low-nitrogen fibrous diets

A major factor limiting application of feeding strategies is the difficulty in communication caused by the dispersal of the majority of animals in small herds and in villages, where families may own one to ten animals. Owners of livestock have fed locally available resources to cattle and buffalo according to experience and availability. At times, and in some places, they have had access to the appropriate supplements. However, these resources are rarely combined in the most efficient way.

It is not easy to convince smallholder farmers who own few animals to accept innovations, particularly if they have to purchase ingredients or reagents, since they are usually short of finance. Moreover, if the advice involves arduous or hazardous work (eg. treatment of straw with caustic soda) the farmer is reluctant to put it into practice unless the rate of return is extremely high.

A notable exception to this is seen in the milk cooperatives in India, where the National Dairy Development Board has established feed mills for producing and disseminating concentrate supplements based on agro-industrial byproducts. Farmers are able to buy these 'balanced' supplements because they are paid daily for the milk which they deliver to the cooperative. This strategy has found wide acceptance.

The first step in increasing utilisation of straw by cattle and buffaloes is to supplement them with urea. Although this appears to be simple and farmers often know that it will increase productivity, nevertheless it has not been applied at the village level. Some of the reasons for this are:

Effort: Preparing a urea solution and spraying it onto straw is a demanding and often arduous task. Urea/molasses mixtures and blocks have been more widely accepted, partly because the technology is easier to apply
Risk: Extension workers in villages are reluctant to recommend a strategy with an element of risk. If instructions are misunderstood, and an animal dies, the extensionist will be blamed and will lose credibility. In addition, where deaths occur, the news spreads quickly through village societies with the result that the technology is rejected
Finance: Smallholder farmers almost always have cashflow problems.

Three procedures have been established for providing urea to animals consuming fibrous, low-nitrogen diets. The first is to spray urea on the straw. The other two depend on the free-choice system, in which molasses is used as a vehicle for the urea. One system employs a liquid mixture of urea and molasses (usually at urea concentrations between 8 and 10%). The other method is to make a solidified block containing the same or higher urea concentrations together with molasses and a gelling agent.

Molasses/urea blocks and liquid mixtures are attractive and palatable to ruminants because of the smell and the taste of molasses. These attributes encourage animals to lick these supplements fairly continuously, thus making the ingredients available to the rumen micro-organisms and the animal almost continuously. This is an important feature when animals are consuming fibrous crop residues that are low in nitrogen (eg. rice straw), since the nutrients in the block are available when the basal diet is being fermented. In practice in India, it appears that if buffaloes are given a molasses/urea block continuously, after about a year they adjust their intake to complement the intake of N from the basal diet.

The primary purpose of these supplements is to provide urea and thereby ammonia for the rumen micro-organisms. They can also provide a broad spectrum of trace minerals and major minerals (especially Sand K), and serve as the carrier for a wide range of compounds which could be used to manipulate rumen fermentation (eg. small amounts of protein to provide peptides and amino acids; branchedchain fatty acids, phosphorus, anthelmintics and even chemicals and organic compounds of plant origin).

The use of urea/molasses liquid mixtures as the basis of the diet is discussed in Chapter 8, and their use as a drought feed in Chapter 9. This section is concerned with the background to the development of urea/molasses blocks and their formulation, and appropriate methods for introducing them to village farmers.

Nutrient blocks (modified mud-bricks) based on earth as the carrier and containing salt as the attractant and including urea, bone meal, trace minerals and concentrated superphosphate are at present being tested in research being undertaken in Northern China (Leng 1987) where molasses is unavailable. In this region, because of the highly alkaline soils, there are apparent deficiencies of zinc, selenium, sodium and phosphorus in the diets which are based on crop residues/legumes. This work is demonstrating that blocks can be prepared from a wide range of materials and do not necessarily have to utilise molasses as a carrier and should allow the extension of the concept of the use of blocks as supplements.

11.2.2 Molasses-urea blocks

Background research

Preliminary trials were carried out in India (in project Operation Flood; FAO/IND/78/007), with blocks containing molasses, urea, rice bran, macroand micro-minerals and a gelling agent. The emphasis in this research has been to rapidly move from "in laboratory" feeding trials and then to adaptive research with the small farmer and finally demonstration trials in the villages.

The blocks were fed first to young Jersey bulls and later to lactating buffaloes receiving a basal diet of rice straw. Subsequently, villages with relatively wealthy (able to afford concentrates) or poor (unable to afford concentrates) communities were selected as sites for trials with individual farmers. In the "rich" villages the introduction of the blocks reduced the amount of concentrates required and still maintained milk production. In the "poor" villages the blocks improved the intake of straw by buffaloes and increased milk production (Kunju 1986). Finally, the overall effect of introducing the blocks in villages was assessed.

Growth trials: molasses/urea blocks

In a trial designed to measure the effect of block supplementation on growth rate, two groups of four Jersey bulls (approximately 350kgliveweight) fitted with rumen cannulae were given rice straw and lkg/day of a high-protein concentrate; the experimental group had access to a block containing molasses (55 %), urea (15%), rice bran (18%) and minerals and filler (12%). The protein supplement contained rice bran and cottonseed cake. The in vitro digestibility of the rice straw was 40% and it contained 0.8% N. The results are shown in Table 11.1. Although straw intake was only marginally increased there was a three-fold increase in weight gain when the cattle consumed the blocks. This response was due to the combined effects of the urea providing rumen ammonia and permitting the protein in the concentrate to be used more efficiently.

Table 11.1: Intake of rice straw by Jersey bulls (350kg lzveweight) given 1kg of concentrate/day with (+ Block) 01' without (-Block) access to a urea/molasses block.
	Straw Intake (kg DM/d)	Intake of block (g/d)	LW change (g/d)	Feed costs (Rupees/kg gain)
-Block	6.4	0	220	9.3
+Block	6.8	530	700	3.7
Source: NDDB (Leng and Preston 1983)

In Ethiopia, mature oxen fed a basal diet of wheat straw responded similarly to synergistic supplementation with a urea/molasses block and oilseed cake (Table 11.2).

Table 11.2: Growth rate of oxen (5-6 years old) fed a basal diet of wheat straw and supplements of aqueous urea, urea/molasses (U/M) block (10% urea), 2kg of noug cal~e/day and noug cake (2kg/d) plus block.
	Urea			U-M block	Noug	Noug +UM block
Intake (kg/d)
Straw			4.9	4.3	4.8	4.7
Total			4.9	4.7	6.6	7.1
LW change g/d)		-190		-70	220	570
Source: ILCA unpublished data

The effects in lambs of supplying a molasses/urea block or spraying the urea on the straw are shown in Table 11.3. Data are also given on the interactions of feeding a bypass protein with urea alone or with a molasses/urea block.

Table 11.3: Feed intake and liveweight change of lambs given a diet of wheat straw supplemented with urea (2% w/w) , urea/molasses blocks (block) and/or cottonseed meal (CSM) (150 g/d).
		Intake (g/d)
Supplement	Rumen NH₃ (mg N/litre) #	Straw DM	Block	Total	LW change (g/d)	FCR (g/g DM)
Nil	26	330		330	-53
Urea	237	322		322	-59
Block	262	421	110	498	10	50:1
CSM	209	440		575	38	15:1
CSM+urea	377	440		575	40	14:1
CSM+block	352	480	90	675	90	8:1
Source; Sudana and Leng 1985 # Measured 3 h after feeding

The generalised conclusions that can be drawn from the results of this study are:

Productivity responses to urea by ruminants fed crop residues are often limited by other nutrients that are deficient in such forages. Poultry litter provides a wide range of nutrients and is often highly beneficial in increasing productivity when supplemented together with urea and bypass protein. For example in mature Zebu bulls (5-6 years old, in thin body condition) increased their liveweight to 740 g/day when their basal diet of Teff straw (Eragrostis tef) sprayed with urea (2% w/w) was supplemented with small quantities of poultry litter and oilseed cake (Table 11.4).

Milk production trials

The first studies were made in India (see Leng 1984a) and indicated that providing a urea/molasses block as a supplement to a basal diet of rice straw led to substantial reductions in the amount of concentrate needed to maintain milk yield in buffaloes (Kunju 1986).

Subsequent trials in Ethopia with crossbred cattle given a basal diet of low-N, low-digestibility hay plus 2kg daily of an oilseed-cake meal showed that milk yield and total dry-matter intake were increased when the animals consumed 700g of a molasses/urea (10%) block daily (Table 11.5). Animals that consumed the blocks had higher rumen ammonia concentrations.

Table 11.5: The effect of providing a urea/molasses block (10% urea) and 1 or 2 kg of noug (Guizotia abyssinica) cake on milk yield, liveweight change and feed intake of crossbred cows fed a basal diet of low-N and low-digestibility meadow hay.
	1 kg noug/d		2 kg noug/d
	No block	+block	No block	+block
Milk yield#	4.2	5.4	5.2	5.4
LW, kg	395	396	336	371
LW change (g/d)	-640	-390	-270	-270

Intake (kg/d)
Hay*	8.7	9.6	8.8	9.3
Block		0.7		0.7
Noug	1	1	2	2
Total^*	8.8	10.1	9.7	10.7
Source: T R Preston, R A Leng and M Nuwanyakpa, unpublished data. #Adjusted by covariance for milk yield on standard diet prior to the experiment and for mean liveweight during the experiment *Adjusted by covariance for mean liveweight during the experiment

There was a significant relationship between milk yield and rumen ammonia concentration (see Figure 11.1).

Figure 11.1: Relationship between rumen ammonia before feeding and adjusted milk yield in crossbred cows fed a basal diet of low-N hay and 2kg of an oilseed cake/day, with or without access to urea/molasses block (Source: T R Preston, R A Leng and M Nuwanyakpa, unpublished data')

Safety in the use of blocks or liquid supplements

At the beginning of the Indian project considerable concern was expressed by the technical advisors about the possibility of over-consumption ofthe block and therefore danger of ammonia toxicity. Initially the blocks were prepared with only 10% urea. Subsequently 15% urea blocks have been used.

When blocks are first introduced to adult buffaloes and cattle on straw-based diets, about 50% of the animals adapt immediately while others take up to 14 days before they consume appreciable quantities. Animals that adapt quickly often consume large amounts of the block, although intake appears to be fairly regular over 24 hours.

Several thousand buffaloes in village herds have been fed urea/molasses blocks containing 15% urea, with considerable increases in productivity. Ammonia toxicity has not been a problem, apparently because of the slow rate of intake. Buffaloes have been maintained throughout pregnancy and lactation on a

diet supplemented with the blocks and have successfully conceived a second time, indicating that there are no ill effects which might appear in the long term from t.he use of the block.

Ammonia toxicity from urea feeding has been overemphasised mainly because of lack of understanding of the principles underlying the efficient utilisation of urea in feeding systems (see Chapter 7). The important point is that when urea is given in a liquid mixture with molasses, only the minimum amount of water needed to dissolve the urea should be added (preferably none if a suitable mechanical mixer is available). Under these conditions the concentration of urea in the molasses can be as high as 20% and the mixture can be offered free choice with safety. As with any new feed, urea/molasses mixtures, either as liquids or as blocks, are best introduced to animals that are not hungry (eg. in the evening after a day's grazing, or after the roughage portion of the diet has been eaten).

Village studies

Both the animal production and socio-economic aspects of the block technology are now being extensively evaluated in villages in India (Table 11.6). In several villages, records have been kept of milk yield prior to and after introduction of a commercial multinutrient block.

Table 11.6: Average daily milk and fat yields from before and after the introduction of molasses/ul'ea blocks to buffalo in villages in the Kaira Milk Producing Union Ltd., Anand, India.
	Milk yield (kg/d)		Milk fat (g/d)
	No block	Block	No block	Block
Village
Alwa	4.8	5.9	330	450
Punadhara	4.0	4.8	270	340
Fulgenamuwada	2.4	3.5	160	280
Hirapura	4.2	5.2	350	480
Banroli	3.6	4.2	270	380
Dehgam	4.3	4.7	310	350
Source: Kunju (1986).

The research indicates that:

Where large amounts of concentrates (46kg/ day) were fed to dairy buffaloes, introduction of the block allowed the farmer to reduce the amount of concentrates fed by 2 to 4kg/day (NDDB 1982-83), thus increasing profitability
Where buffaloes were largely maintained on straw with little supplementation, the introduction of the block resulted in marked increases in straw intake and an increase in milk yield of 50100%. A typical result from data taken from the records of the sales and composition of milk is shown in Figure 11.2. Milk yield was increased from about 1.5 to 2.4 litres/day and, consistent with the effects of stimulating rumen fermentation (ie. increasing microbial protein production and therefore availability of fat and glucogenic energy in the fermentation end-products), the fat concentration of the milk increased.

Figure 11.2: The effects of introducing a molasses/urea blod~ (20% U1·ea) to a milking buffalo in a village in Mehsana, Gujarat, India. It took approximately 8 days before the buffalo consumed significant amounts of the block (500 g/d). Milk yield and fat content increased after· a time lag of 7-10 days. The basal dlet was straw from millet with a "handful" of green forage (Source: G Kunju, A Dave and R A Leng, unpublished data).

The increase in milk fat when the nutrient block was provided to buffaloes is consistent with an increased efficiency of synthesis of microbial cells in the rumen. The increased availability of microbial protein and fat relative to acetogenic nutrients (acetate and propionate) accounts for the im proved fat deposition in milk. In previous cakulations (see Table 4.4) the effect of creating an efficient microbial ecosystem by providing the deficient nutrients in a straw-based diet was to increase intake from 10 to 14kg straw / d and microbial cell production in the rumen from 0.83 to 2.33kg DM/d. Because microbes are 10% lipid this would increase the availability of long chain fatty acids by 150g/d or about the same increase in fat yield in milk brought about by supplementation with a molasses/urea block.

So far, the research and development activities with urea/molasses blocks have been mostly confined to tetlH'red animals. For ruminants grazing on communal lands and housed and fed crop residues at night, the block may be fed in the pens or sheds. However, when the pasture is dry and low in N it is preferable that the blocks accompany the animals to the grazing areas so that they are always available.

11.3 Restricted suckling

In Chapter 2, attention was focussed on the interaction between systems of calf rearing and the use of crossbred cows for milk production. Crossbreeding native cows using imported semen of specialised dairy breeds is the strategy most frequently advocated for establishing a dairy industry in developing countries. As part of the "technological package" it is usually advocated that calves are removed from their dams within a few days of birth and bucket-fed, either with cow's milk or with a milk substitute. This policy is being increasingly questioned (Preston 1977) as being inappropriate under the conditions of feeding and management in most developing countries. Restricted suckling appears to be a more viable option (Preston 1983a) because it results in increased production of both the dam and the offspring (Table 11.7, Table 11.8, Table 11.9 and Table 11.10).

Table 11.7: Effect of restricted suckling on milk yield of cows. Comparisons were mostly during early lactation (8-12 weeks) or until the calves were weaned
Breed	Reference		Bucket	Calf	Total	Milk yield (no calf)
Lactation yield (kg)
Cross breed	1		910	560	1470	218
Holstein^b	2		3424	-	3424	2340
Holstein^b	3		1598	-	1598	1463
Daily milk yield (kg/d)
Holstein		1	6.2	6.9	13.1	10.7
Holstein		2	7.8	6.8	14.6	9.7
F1 (Holst.xZebu)b		3	3.9	6.6	10.2	6.3
Sahiwal		4	4.3	2.7	6.9	2.7
Creole		5	7.9	2.7	10.6	8.8
Hereford x Holstein		6	4.5	3.9	8.4	4.9
1. Alvarez et a1. (1980) 2. Paredes et a1. (1981) 3. Ugarte and Preston (1975) -4. Ugarte and Preston (1972) 5. Khan and Preston (1985) 6. Gaya et a1. (1977). a.Milking without the calf was in the same herd but in different years b. Yield from 8 week to end of lactation (suckling only during the first 8 weeks).

Table 11.8: Friesian/Sahiwal cattle in Malaysia were milked by hand with the calf at foot (Normal System) or by machine without calf stimulation ("Improved" System). Total milk produced per lactation was much higher in the former system, mainly because lactation was prolonged in the hand milking/restricted suckling system.
	Normal System	"Improved" System
Milk, kg/lactation#	1860	1410
Lactation, d	330	229
Calving interval, d	438	419
Source: Cheah and Kumar (1984). # Not including milk consumed by calf.

Table 11.9: Mastitis incidence in crossbred Holstein and Brown Swiss/Zebu cows was reduced when they were milked with the calf given restricted suckling.
	No suckling	Restricted suckling
No. of cows	45	47
Mastitis incidence (%)#
Negative	54	77
Suspicious	13	14
Positive	32	7
Clinical cases	7	0
Lost quarters	2	0
Source: Alvarez et al. (1980). # California Mastitis Test.

Table 11.10: Growth rate of calves and conversion of milk into liveweight gain are improved when calves are reared by restricted suckling (RS) rather than with milk from a bucket (artificial rearing AR)
		Calf growth (g/d)		Conversion#
	Ref.	RS	AR	RS	AR
Crossbreeds	1	464	277
Holstein	1	770	500	7.8	8.0
Sahiwal & AIS crosses*	2	552	370	5.0	9.0
Creole	3	317	413	8.4	9.3
HerefordxHolstein	3	497	353	7.8	1l.4
Buffaloes	4	463	330	6.2	8.5
# Milk consumed by the calf/weight gain (kg/kg) * Australian Illawarra Shorthorn. 1. Alvarez et al. (1980), Velazco et al. (1982b). 2. Khan and Preston (1985). S. Gaya et al. (1977). 4. T R Presion (unpublished data).

Most calves in developing countries are expected to obtain most of their nutrients from the cheapest and most available resources such as low-N pasture and crop residues. In this situation, restricted suckling allowing the calf to consume a small amount of milk, which bypasses the rumen completely and is a balanced combination of essential amino acids, glucose and long-chain fatty acids-enables the calf to grow at a satisfactory rate on basal diets which, if fed alone, would not support maintenance. The actual cost of allowing the calf this vital supplement is very small because the calf is only allowed to suckle after the cow has been milked. Restricted suckling also appears to stimulate the cow and may increase the total amount of milk produced by the cow (Table 11.7, Figure 11.3).

Figure 11.3: Holstein cows m Venezuela milked by machine and with restricted suckling of their own calves after milking gave more milk and lost less bodyweight after calving than cows whose calves were removed after 3 days and reared artificially (Source: Velazco et al. 1982a).

In tropical climates, the most appropriate animals for producing milk are buffaloes and crossbred cows, since they have the necessary reproductive capacity and the tolerance to environmental stress. In these situations, milk must be produced against a background of moderate nutrition (eg. basal diets of crop residues and low-N pasture) and a hot, often humid climate where there is usually a continuous challenge from vector-borne diseases. The proportion of 'indigenous' genes (usually Bos indicus) in the crossbred cow almost certainly should never be less than 50%. A characteristic of this type of animal is that both milk 'letdown' and persistency of milk production are poor compared with specialised dairy breeds from temperate areas unless the calf is present at the time of milking.

The contrast between the systems for rearing calves in industrialised and developing countries must be understood. In the former, artificial rearing is the system of choice, but this is conditioned by the availability of milk substitutes formulated from a variety of products including skim-milk powder, buttermilk powder and tallow, and which can be purchased at less than the farm-gate price for whole milk.

These liquid feeds are supplemented with, and gradually replaced by, high-quality dry feeds based on cereal grains and oilseed and animal byproduct meals. Feeds such as straw and dry pastures are never an important source of nutrients during the calf's early growth period.

In addition, the specialised dairy breeds let down their milk as readily to machines as to calves and milking systems have been devised that enable a high output of milk per unit of labour (which is expensive). Thus, artificial rearing is "appropriate" in the industrialised countries.

11.3.1 Conclusions

For developing countries, the advantages of restricted suckling of calves are many and include:

Milk 'letdown' is stimulated whether the milking is done by hand or machine
Milk yield is increased by up to 30% (Table 11.6 and Table 11.8)
Cows that continue to suckle their calves lose less bodyweight after calving (Figure 11.3).
Growth rate and health of the calf are improved considerably in comparison with animals given the same amount of milk from a teat or a bucket (Table 11.10)
There is little or no effect on reproductive rate (Table 11.11).

Table 11.11: Effects of restricted suckling (RS) or artificial rearing (AR) of calves on time to first oestrus and inte1'Calving intel'val of the dam.
		First oestrus (d)		Calving interval (d)
Breed	Ref.	AR	RS	AR	RS
Holstein	1			352	352
F₁(HolxZebu)	1			336	343
F₁(HolxZebu)	2	-	84	-	350
Holstein	3	89	90	365	373
HolxZebu	4				380
F₁ HolxZebu	5			399	422
Hol/SwissxZebu	6	-	-	474	416
Holstein	7	65	66	392	394
1. Ugarte and Preston (1972) 2. Veitia and Simon (1972) 3. Ugarte and Preston (1975) 4. Fernandez et al. (1977) 5. Ugarte and Maldonado (1979) 6. Alvarez et al. (1980) 7. Paredes et al. (1982)

11.3.2 Theoretical aspects of restricted suckling

The reason for the improvement in milk yield and liveweight gain of the cow is not readily explained unless there is a contentment aspect of suckling or the hormonal balance in the cow is changed such that it partitions nutrients into milk/body weight gain more efficiently. It is possible that the inefficiency of dairy cows that are stressed by the removal of the calf is associated with an increased glucose utilisation in tissues, which, in turn, affects the balance of nutrients available for milk production.

The improvement in utilisation of feed by the calf, however, may be explained in terms of the balance of nutrients available to the calf from milk (Table 11.12).

Table 11.12: The theoretical balance of nutrients available to a ruminant calf receiving 6 litres of milk containing 4% fat, 4 % protein and 4 % lactose. The milk is either suckled (restricted suckling RS) or fed from a bucket (artificial rearing A R). It is assumed that either 50% (R₅₀) or 100% (R₁₀₀) of the milk enters the rumen and its components are fermented.
		Source of nutrients
			AR
Nutrient	Origin	RS	R₅₀	R₁₀₀
Protein	Diet	240	120	-
(g/d)	Microbial	-	39	78
	Total	240	159	78
LCFA	Diet	240	240	240
(g/d)	Microbial	-	7	1
	Total	240	247	254
Glucose	Diet	270	135	-
(g/d)	Microbial	-	11	22
	Synth.#	-	50	100
	Total	270	196	122
Acet+Butyr (MJ/d)	Rumen	-	1.0	2.0
# Synthesized from propionate

Milk, when it is sucked from a nipple, or from the treat of a cow, is channeled by the reticular groove reflex directly to the abomasum avoiding the developing rumen. Drinking milk from a bucket, however, is less effective in activating the reticular groove and a large proportion of the milk probably enters the rumen.

Rumen development in the calf commences almost immediately the calf begins to chew long roughage (grass) and at 3-5 weeks of age it can be well developed and have an active fermentative digestion. Milk entering the rumen is fermented and thus potentially creates an acid rumen environment which slows its development.

The rate at which milk proteins are degraded in the rumen is not known but is probably high; lactose will be fermented rapidly and totally: fat will be hydrolysed to glycerol (which is fermented) and LCFAs which are unchanged.

To illustrate the effect of milk entering the rumen or bypassing the rumen on the balance of nutrients available to the calf a fermentative balance is presented in Table 11.12. The assumptions are:

All milk protein and lactose entering the rumen is fermented efficiently into VFA and microbes
One third of the milk sugar fermented is converted to microbial cells and two thirds to fermentation end-products (VFA, CO₂ and methane)
One sixth of the milk protein fermented is converted to microbial cells and five sixths to VFA, CO₂ and methane
That microbial cells are 60% protein, 10% fat, 20% polysaccharides and 10% ash
VFA are produced from milk solids in the ratios; 60 Acetic, 30 propionic and 10 butyric acids
That all the propionate is synthesised into glucose, one mole of propionate giving rise to 0.5 mole of glucose.

The values in Table 11.12 are calculated assuming an intake of 6 1itres of milk/day having 4% fat, 4% protein and 4% lactose. These calculations indicate that feeding milk from a bucket as compared to restricted suckling of the cow decreases the availability of glucose precursors and amino acids relative to long chain fatty acids and acetogenic VFA in the nutrients absorbed by the calf. From previous discussion this should lead to an increased heat increment of feeding and a lowered efficiency of con version of milk to liveweight gain.

Such an elevated heat production in hot climates could lead to a much reduced intake of solid feed.

The practice of artificial rearing is accepted in temperate countries and widely advocated in the tropics even though traditionally, farmers have used restricted suckling. On theoretical grounds and in practice, this system is inefficient and is therefore inappropriate for the majority of small-scale farmers.

Table 11.4: Effects on liveweight change of Zebu bulls of supplementing their basal diet of Teff straw (+ 2% urea) with either poultry litter (500 g/d) and/or noug cake (Guizotia abyssinica) (500 g/d).
		Supplements
	Basal diet	Poultry litter	Noug cake	Poultry litter + Noug cake
LW gain, g/d	-0.25	0.32	0.67	0.74
Source: M Hiwot, T Tadesse and T R Preston, unpublished data.

Chapter 11 NUTRITIONAL PRINCIPLES: SMALLHOLDER LIVESTOCK SYSTEMS

11.1 Introduction