Previous Page Table of Contents Next Page



Two major issues are discussed in this chapter. The first concerns the existing village parameters relating to feeding and feed supply; and the second to the introduction of innovations on farms. The critical role of village/farm surveys will be discussed in relation to both issues.

For unknown reasons, livestock workers have been slower, apparently than workers in other fields such as agronomy, to realize the value of village/farm surveys and on-farm trials. However, there is now a growing understanding throughout the tropical world that for research to be contributing to local development, it needs to be more specifically geared to the particular situation prevailing in a local environment. Earlier, the justification for research attention to animal production has mainly been seen in situation where increasing incomes would lead to greater demands for milk and meat - or increasing incomes have been one of the major underlying premises justifying research objectives.

It is now clear that the spectrum is broader. Particularly, that there are a number of countries where economic growth is stagnating or negative and that in these situations the correct approach will be to maintain and support the traditional systems, which have the merit that they are operating under the existing conditions and on the basis of existing management skills.

There are two more reasons. Numerous attempts have been made to make shortcuts to increased animal production by introduction of the very intensive Western types of animal production, particularly through importation of temperate breeds of animals without due consideration to the nutritional and managerial aspects. It is now clear that no breed can substitute for adequate nutrition and management. And that these problems must be tackled first, if progress is to be expected on the basis of local resources.

Furthermore, with the energy crisis the role of the working animal as a source of renewable energy, independent of oil and spare-parts imports is being increasingly appreciated.

As another reason, it may be mentioned that livestock are seen now as also having a contributing role to play in connection with fuel supply to rural households, either by supplying dung for burning or biogas production or as converters of residues produced in connection with energy production from crops or trees into useful products.

Much of the current thinking may be summarized in the following statement by K jaerby (1983): “The basic premise is that it is easier and potentially more productive to adapt technology to peasant farming than it is to transform peasant farming to inappropriate technology”.

It is on this background that the growing interest in village/farm surveys and on-farm trials should be seen. The approach should be based on the following sequence of priorities:

The following observations are strongly influenced by practical experiences from Bangladesh and to a lesser extent India. However, it is hoped that the underlying principles can be useful in other countries.


Although there is a growing appreciation of the traditional animal production systems, it needs to be stressed that there is a long way from understanding the need for village and farm based survey and trial work till this becomes a regular feature of the time schedule of the animal nutrition worker. Some of the reasons are that most people in developing countries, who have received education to an extent that enables them to consider village-oriented animal nutrition work, come from a social background which is different to the background of the core of the livestock keepers. They are often impractical and their knowledge of local agriculture is poor. This makes rapport and communication with the farmer difficult in the same way as it makes it difficult for the aspiring young research officer to gain the confidence and understanding from his/her superiors. Experienced guidance will be still more difficult to obtain.

On the other hand, expatriate advisers working in developing countries may have a more practical farming background with them from their own country, but the conditions are different and they do not know the local language and customs. Even in the best case it will take time to learn both. Finally, village based work cannot be conducted with the same experimental rigour as laboratory or experiment station work. For scientists giving high priority to such rigour, village based work will not be easy.

Nevertheless, there are examples of successful survey and on-farm trial work being reported (Agarwal and Verma 1983; Breinholt 1982; de Lasson 1981; Dolberg et al 1981a; Jabbar and Green 1983; Helmrich 1983; Ibrahim et al 1984; Sere and Vaccaro 1984) indicating that the problems can be overcome. In this connection there is no doubt that more could be done to make such work professionally rewarding by - for instance - instituting rewards for successful application of scientific principles under practical farm and village conditions. Promotions might also be based on successful conduct of such work.


7.3.1 The village survey on feed supply

This might also be termed the aggregate survey or the survey setting the framework (context) for animal production. Some key variables are discussed below. Quantity

Most feed grows on land although the water hyacinth is important in some areas together with other water plants. Land can usefully be classified as agricultural, common and homestead land. Each component will typically contribute to livestock feeding. Roadside and embankments are included under the term common land. Figures on size of the areas and the cropping patterns/most common vegetation will help identify the relative quantities and qualities of the feeds available. Based on Helmrich (1983) the following schematic outline can be given;

Table 7.1: Village fodder potential: origin, type and quantity
Origin Type Quantity*
Agricultural land Cultivated fodder  
Crop residues  
By-products from processing  
of harvested material  
Grass and weeds from fallow  
Common land Natural vegetation (grass,  
scrubs and weeds)  
Water plants  
Homestead Grass  
Tree leaves  
Vegetable wastes  

* No figures are given as this is only an illustration.

When presenting the information gathered on quantity, the supply situation will be put into a useful perspective, if figures are given not only in the aggregate, but also as amounts per head of livestock, separated according to different categories (eg: basal feed resource and supplements), to give perspective and relevance to research work in this area. Quality

When information has been obtained on the quantity and origin of the feed, the next question to answer will be about quality. The most important indicator will be the potential rate and extent of digestion (see Chapters 2 and 3). If reasonable values cannot be obtained for these two parameters, voluntary intake will be too low and the material will be unfit as a feed, unless it is upgraded by adequate supplementation and/or chemical treatment.

If there are no facilities to measure rate and extent of fermentation, the nitrogen content may also be used as an indicator of nutritive value. However, this can be misleading as N deficiencies are easy to correct; and a simple analysis for nitrogen does not differentiate between fermentable N for the rumen microorganisms and protein that is likely to escape the rumen fermentation (bypass protein) and contribute directly to the needs of the animal. Seasonality

By establishing seasonality in feed supply, the survey may immediately point at storage and preservation as important research topics. This may not only be with the objective of matching energy and protein supplying feeds. It is possible attention to the seasonality factor also has revealed a need for measures to avoid losses of feeds due to climatic factors such as rain. Ownership/user rights and feed market

Ownership/user rights are other important variables to consider in connection with a survey on feed supply. In the situations where the agricultural land is privately owned, it should not always be taken for granted that feed from such land is only for the owner. There are areas of the world, where crop residues are a part of labour wages or the owners find it too expensive to transport them home and leave it on the field for anybody to pick up. Similarly, it is also seen that livestock graze stubble over all agricultural land after harvest, irrespective of ownership. Or, crop residues may be burned or ploughed into the soil to improve fertility.

With regard to village common land, it will be important to determine who controls it not only in formal but, more importantly, in real terms. It is insufficient to accept a statement that it is - for instance - government land. It is more important to establish whether and from where any opposition may be expected in case any improvement schemes are taken up. Only then, it will be possible to make realistic proposals about how such land can be used better.

The market for feed is linked with the questions of origin and ownership of feed resources. Information on whether the farmer and/or village is a net importer or exporter of feed will also be important. Competition between feed and fuel

The use of dried cow dung as fuel is well known. That this phenomenon is only one step on a road, which ultimately may lead to a competition between feed and fuel is, perhaps, less well understood. Taking the example presented in Figure 7.1 from Bangladesh (Helmrich 1983), it is seen that during phase I until about 1960, wood or wood products were the single most important source of fuel. Thereafter the importance of wood starts to decline and its use becomes insignificant from 1990, as none is available. From 1960, cow dung plays a significant role but agricultural residues, having no use as feed, are more important and from about 1980 agricultural byproducts, which could as well be used as feed, assume an increasing importance.

From 1990 onwards, the use of agricultural byproducts for fuel affects feed supply so markedly that the cattle population starts to decline, which in turn has an influence on dung supply. The underlying cause is the growing human population which leads to an increase in demand for fuel. These issues must be brought to the attention of animal nutritionists so that they can make realistic projections and establish priorities for research. Governments' role

Governments assist in livestock production in various ways by providing veterinary and other services. Governments also decide on the training and allocation of manpower. The scope and intensity of such assistance need to be assessed. For example, Government subsidies for imported feeds affect the use of local feeds.

7.3.2 Animal production systems

The notion that livestock are an integrated part of the total farming system in developing countries is not always correct. This has been illustrated in Bangladesh with data on land ownership and livestock keeping (Table 7.2). Only 23% of the households owning draught animals were in the category of having from zero to 0.20 ha of land. Relatively speaking it is clear that the draught animals belong to the holdings with land. The situation for the other categories of livestock is different. Of the households owning milking cows, 44% had between zero and 0.20 ha of arable land. For young stock, goats and poultry, the figures were 50%, 49% and 57%, respectively.


Figure 7.1

Source: Helmrich (1983)

Table 7.2 Relationship between total arable land ownership and livestock keeping
land, ha Draught cattle Milk cows Young stock Goats Poultry
House-holds Animals House-holds Animals House-holds Animals House-holds Animals House-holds Animals
13 24 20 21 35 38 49 94 177 228
27 49 50 55 61 70 49 84 192 274
63 115 32 38 52 68 53 93 153 240
58 116 39 53 46 64 33 57 105 204
Above 1.6
14 35 18 29 16 26 14 28 22 49
175 339 159 196 210 266 198 356 649 995
Chi square 230.40 124.73 98.55 41.03 171.22
Pearson's R 0.39 0.28 0.26 0.12 0.33

* Number of households in this size group

Source: de Lasson (1981)

It is clear that a substantial part of the livestock, even among the households with draught animals, is kept by households with insufficient home-produced feed. From that point of view, it may be safe to assume that about 50% of the livestock in this case is not integrated into the farming system as such livestock have to rely on common land for feed supply. Such information is important as it can be used to identify the existing animal production systems.

It is clear that at least two systems exist: one that is integrated with farming and one which is not. To devise useful strategies for the section of livestock owners not having sufficient land, organizational issues (cooperatives, credit and marketing), including control and increased productivity of common land, may be the first constraint to increased animal production.

Feed can also come from private land, owned by individual landowners or it may be imported. For any of these sources to be exploited by livestock keepers who have no land, an organization needs to be formed, which will ensure the transfer of feed to the livestock keepers.

Transfer of feed to livestock keepers without land can be done through the established technology of feed mills. However, for this to be economical, it often presupposes large animal units and productivity per animal usually has to be at a high level.

There are traditions for extensive grazing of stubble after harvest across fields, irrespective of animal or land ownership. Productivity per animal is low in these systems. There is also the tradition of common land grazing with low productivity per animal. It is quite possible that a more intensive exploitation of village common land is a way by which the landless livestock keepers could derive more benefits from animal production. However, the organizational and technological issues have hardly been touched, although this approach might represent an alternative to the high-technology feed mills and their dependence on imported feeds.

7.3.3 Introduction of innovations Dealing with farmers

To be able to evaluate the impact of an innovation, it is useful to start by collecting some base-line data. There is now sufficient evidence to postulate that farmers can be found all over the developing world, who are willing to collaborate with researchers. A successfuloutcome depends on the researcher.

An introductory survey is a useful tool to identify collaborating farmers for on-farm testing of innovations. In selecting such farmers it is important to appreciate that whereas it may be possible to maintain statistical rigidity, when selecting farmers for the base-line survey, this may not be the case for on-farm trials. It is not all farmers, who are interested in hosting an innovation trial. The pragmatic procedure is to include only motivated farmers.

The foremost condition for successful work is to establish a good relationship with the farmer. It is imperative that the farmer be regarded as a collaborator, who has specialized knowledge and who can be helpful in establishing research priorities. The farmer must not be treated as an object, and a farm survey/trial should not be considered in the same way as a controlled experiment station trial, where the researcher decides when a variable has to be changed. It is quite possible that farmers will agree to be grouped into various treatments. It is also possible to arrange with them that animals are weighed on particular weekdays at times suiting the farmer, although this may mean odd working hours for the researchers such as early morning or late evening. Agreements can also be reached about feeds kept aside in separate heaps for weighing till the recorder reaches the village, when due respect is given to feeding times.

If during the course of the survey an extraordinary situation arises, which causes the farmer to deviate from the agreed procedure, this will have to be accepted by the researcher. Such situations may be caused by a daughter's marriage and suddenly the researcher may find that supplements are not being fed any more or - even worse - the animals are being sold.

Natural calamities, such as cyclones and droughts, can also interfere. The farmer may face a shortage of feed and accept emergency feed distributed by the Government or start to buy feed. Although such deviations may defeat the purpose of the survey, it cannot help if the researchers start to argue with the farmer.

The farmer and his family have to live from farming and have no permanent salary and they have to exist within the social structure of the village. Instead of criticizing the farmer, it is more useful for the researcher to reflect on his/her own reactions to social pressure and salary cuts.

It must be understood from the initialstage that the trial must in no case be detrimental to the income of the farmer. Firm guarantees should be given before undertaking any trial and compensation be paid should any difficulties be encountered. Size of sample

The size of the sample may be determined in consultation with a statistician. However in many situations such ideal requirements cannot be met for want of money, qualified manpower, transport, etc. In this situation, it may be useful for the researcher to recall that the human brain is the most important research tool after all and that it is important not to allow the lack of conventional research gear to stifle one's determination to get into the villages and be exposed to the environment.

Even with one farmer important observations can be made. Thus the first observation on the high level of straw intake by the native cattle of Bangladesh (Mould et al 1982) was made on a village farm.

There have been numerous research reports discussing the possibility of using rice straw for animal feeding, often concluding that the high silica content is a major limitation. In this connection it is sobering to get into the villages and realize that irrespective of the composition of rice straw, it is the staple feed of millions of animals and it has been so for years and probably centuries.

Agarwal and Verma (1983) have described an on-farm trial carried out by Jackson in the hills of Uttar Pradesh in India in a cluster of 4–5 villages. Twenty-seven calves, a mixture of males and females, and cattle and buffaloes, were used and allocated in blocks to treatments of untreated rice straw and a small supplement of grass (control); the control plus 30 g/d of a commercial mineral mixture; or rice straw ensiled with urea accordingto the procedure described by Dolberg et al (1981a). Liveweight gains (Table 7.3) were derived as the differences between initial and final weight or by regressing liveweight on time on experiment. There were no statistically significant differences between the treatments when the former method of calculating liveweight gain was used; whereas use of regression analysis revealed an obvious advantage for the animals fed the urea-ensiled straw.

Table 7.3 On-farm trials involving the feeding of urea-treated straw
Farm diet Farm diet + mineral mixture Farm diet of urea-treated straw  + mineral mixture
DM digestibility of straw (%) (nylon bag) 32 32 57
Nitrogen content of straw 0.5 0.5 1.3
Ca content of straw 0.5 0.5 0.6
Average daily weight gain (kg)
By difference 0.12a 0.17ab 0.26b
By regression analysis 0.13a 0.17a 0.26b

Source: Jackson, M.G(personal communication) NOTE: Values followed by different superscripts are significant at P< 0.05.

It is important that the research worker undertaking village work is alert to such phenomena and uses the correct analytical methods.

It is likely that village-oriented work will throw up a number of issues requiring the use of appropriate research tools. As stated earlier, the correct approach is not to stay out of the village for want of tools, but rather look to other disciplines for techniques which will overcome the problems. Time scale

It is difficult to give firm rules about the correct time scale. The objective of the particular research work will obviously be one important determinant. For measurements of feed intake, 3 weeks may be enough; whereas the effect of a particular supplement on reproductive parameters such as age at puberty or calving intervals will have to be studied over several years.

In general, the view taken here is that while the specific research objectives may vary, village surveys and on-farm trials should become a permanent feature of animal production work.

In most developing countries, relatively large livestock farms are found belonging to universities, research stations and Government department farms. All too often, applied research is confined to such farms and researchers rarely move outside them. This is in spite of the fact that only a small fraction of the national livestock production will ever take place on farms in this category.

Most of the livestock production takes place in private livestock units, small and large, and will continue to do so. Measured on any product (milk, meat or draught power) the private units account for the major part (usually over 90%) of total national production. As the funds, which go into training of livestock research workers are spent with the objective of increasing national production, national questions must be addressed. As nobody plans to stop research work in animal nutrition, it is only logical that village based work becomes a permanent feature of such work and not only a brief interlude. This is also the only way dynamics over time can be established. The control group

The question to be discussed in this section is, how to measure the effects of innovations? Participating farmers may be allocated to control and experimental treatments, but after a short time the control farmers may also want to switch to the experimental treatment. There may be two reasons for this. One may be a positive effect on the animals; another is that the treatment input is provided free of cost, and the control farmers want their share as well. In this case, some indication of real impact may be obtained by returning to the village some months after the trial is over and making a survey on whether the farmers still continue with the new practice.

In situations where all participating farmers have at least two animals of the same category, the problem of the control group may be overcome by putting one animal into the control group and one into the treatment group. This will also minimize the management factor. A third way is to use the base-line data collected through the introductory survey ( If this has been done for a sufficiently long period, the researcher will have a good foundation for judging when performance is good due to the treatment and when it is poor. What can be measured?

A number of parameters can be measured, although it may be useful to start with only a few. Some of them are feed intake, growth rates, working hours, age at calving or work, calving intervals and lactation length.

Feed samples can be collected and taken to experimental stations for determinations of in vitro or nylon bag digestibilities. Content of nitrogen and other nutrients can also be measured. An assessment of draught power can be made in villages by recording the nature of the work and the time spent in the activity; and as the objective is usually to compare one treatment against another, absolute figures are rarely required. Equipment

The required equipment will vary with the objective. Patience, common sense and endurance are important qualities in the researcher. A small pocket balance with a capacity of 25 kg may be all that is needed, if the objective is to weigh new-born and young sheep and goats. In many areas this will also be sufficient for new-born calves. Feed can also be weighed with such a balance. A piece of rope and a basket or a bag on which to place the animal or feed during weighing can be found in most villages or their markets.

Growth of larger animals can be measured in terms of increments of chest girth, measured by a tape. It is not necessary to transform these into weights, as the objective is simply to have comparative data; and in any event, the girth: weight relationship differs according to breed.

In Bangladesh a cycle rickshaw was remodelled and used to carry a weighing scale from village to village. Transport facilities in the form of bicycles, bus tickets, motorbikes or cars will be needed. Manpower and cost

There are many examples of investigator assistants with from 5– 10 years of school education, recruited from within the village, who have done an excellent job. People with higher education can also be used and it is possible to identify a number of problems for graduate student projects within this type of work.

The first point, which needs appreciation in considering cost is that by spending money on getting into the villages, the researcher is saving on other accounts. Money is not needed for experimental animals, buildings, pens, offices, machinery and feed. Management of the animals is also free of cost. These points are very important for the research worker, who is short of funds.

There may be good reasons to conduct animal trials simultaneously on farms and in research stations; one need not exclude the other. However, if the arguments advanced in support of village-based work (7.3.3) are accepted, it seems logical to ask the question as to whether it would not be correct to allocate funds for this type of work, and to give this higher priority than the work to be carried out on large livestock farms (university, research station and Government farms), which almost all as a rule operate at a loss without contributing much to increasing national livestock production.

It is though-provoking that in situations where this applies, the only positive function left for these units is a social welfare function, which they serve by providing employment and some produce (eg: eggs, milk and meat) for the surrounding community.


It is concluded that a number of factors have to be considered for successful execution of village surveys and on-farm trials. However, the most important is the attitude of the researcher. There are sufficient examples from developing countries to demonstrate how such surveys can be conducted and that farmer participation is not a problem.

Finally, such research conducted on the farm is the best guarantee that the objectives are sound, and that the findings will be of interest to the farmer. This also facilitates the success of the extension.

There is no doubt that more could be done to stimulate such work by Governments and other high level authorities. Surveys and on-farm trials must, ideally, be viewed as tools to establish two-way communication with livestock keepers. New ideas can be tested and new research objectives identified. They should therefore become permanent features of animal nutrition work.

Previous Page Top of Page Next Page