Developing, testing and transferring improved Vertisol technology: The Indian experience

L.D. Swindale

International Crops Research Institute for the Semi-Arid Tropics (ICRISAT)
Patancheru, Andhra Pradesh 502 324, India

Abstract
Introduction
Vertisols and their environment
Soil degradation
Traditional dryland farming systems
Prevailing production constraints
Improved cropping systems
On-farm studies of the improved systems
Early adoption by farmers of the new technologies
Conclusions
References

Abstract

Vertisols are widespread in India and are generally used to grow annual crops. In many areas Vertisols are fallowed during the rainy season, which subjects them to soil erosion. Improved cropping systems have been developed for Vertisols in less-assured rainfall zones (less than 750 mm per year), based on post-rainy season production of sorghum or safflower. These systems do not remove the problem of rainy-season fallows. Rotation of crops with grassland or farm forestry is now being studied. Improved cropping systems in assured rainfall zones (more than 750 mm per year) promote cropping in both rainy and post-rainy seasons. The systems are proving successful and are being adopted quite rapidly by farmers, particularly where crops in current demand, such as soybeans, wheat and certain pulses, are included in the systems. The components of the package of practices which are perceived as most beneficial by farmers are double cropping, the use of improved cultivars and the use of fertilizers and pesticides. Land and water management practices are perceived as less attractive because the benefits tend to be long-term and they must be shared with society in the form of erosion control and reduced downstream flooding.

Adequate weed and disease control and the unsuitability of some crops to the practice of dry seeding are technical problems that limit adoption. Lack of inputs, labour, credit and extension services and inadequate marketing and distribution systems are institutional problems. The new watershed-based systems also require more cooperation among farmers than traditional crop production systems.

Introduction

The farming systems of the semi-arid tropics (SAT) are characterised by low agricultural productivity. The soils in these regions often have low fertility and are difficult to cultivate. The rainfall is low, erratic, and highly seasonal, and the socio-economic resources are limited. The current level of crop production in these harsh environments is inadequate to meet the needs of rapidly increasing populations.

Of the major soils of the SAT, Vertisols are some of the most productive for rainfed agriculture. Their high water-holding capacity allows them to compensate better than most other soils for the low and erratic rainfall, which is a major constraint to crop production in the SAT. Because of their high potential to increase productivity and their wide occurrence in the SAT, the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) has given a high priority to the development of improved farming systems for the Vertisols. Recent research has shown that, with proper management, the productivity of many of these soils can be increased several fold, growing two high-yielding improved crops each year instead of the traditional single crop of low yield potential. At the same time, runoff and erosion are reduced. The technology includes a number of interrelated components that enable farmers to improve the workability of the soil and better control the moisture level.

This paper, which draws heavily upon an earlier paper by Virmani and Swindale (1984), discusses traditional and improved farming systems for the Vertisols in the Indian SAT environment. Because maintenance of long-term productivity under intensive farming systems in the tropics is a serious cause for concern, the discussion includes the implications for reducing the current degradation of the soil.

Vertisols and their environment

Vertisols and associated soils cover approximately 257 million ha of the earth's surface (Dudal and Bramao, 1965). They occur extensively in India (72 million ha), northern Australia (71 million ha), Sudan (63 million ha), and Chad and Ethiopia. Vertisols also occur in Central America and Venezuela.

Vertisols and associated soils occupy 22% of the geographical area of India (Murthy, 1981). In the central region of the country, known as the Deccan Plateau, the soils are derived from weathered basalts mixed to some extent with detritus from other rocks. In other areas, particularly in the south, the soils are also derived from basic metamorphic rocks and calcareous clays. In the west, in Gujarat, they are derived from marine alluvium.

Vertisols develop mainly on the gentle slopes-usually less than 3%-of terraces, plains and valley floors in association with vertic Inceptisols, Fluvents, hydromorphic, and salt-affected soils. Occasionally they form on low smooth ridges, but seldom occur on slopes greater than 8%. They are more common at elevations below 300 m, but extensive areas in India are above this altitude (Fitzpatrick, 1980). Vertisols usually have impeded drainage; their parent materials are mostly basic and fine textured, such as basic igneous rocks, limestone, and river or lacustrine alluvium (Young, 1976). Surface gilgai are not common.

The Deccan Plateau, the main area with black soils in India, is a region of low relief with broad ridge remnants of the basaltic plateau, separated by wide, shallow valleys. Vertic Inceptisols of moderate depth and shallower Entisols occur on the plateau remnants. Entisols also occur on the steeper upper slopes of the pediment surface below the ridges, grading through Inceptisols to Chromusterts on the more gentle slopes to the middle and lower pediment. Chromusterts occur on the depositional surfaces below the pediment, sometimes with Pellusterts on the lowest, flattest land surfaces. Hydromorphic and salt-affected soils occur in depressions on the slopes and in the numerous drainageways.

On the eastern fringes of the Deccan Plateau, black soils occur in transported detritus derived mostly from basalts overlying granites and gneisses of the basement complex. Where the basement rocks are soil-forming, they give rise to red brown Alfisols.

The climatic environment, extending from arid through semi-arid to subhumid tropics, is characterised by dry and hot pre-monsoon months (April to June) and dry, mild winters. The mean annual rainfall, equal to about 30-75% of the potential evapotranspiration, ranges from 500 to 1200 mm (Table 1), of which 80 to 90% is received during the monsoon season from June to September.

Soil degradation

In India, Vertisols are particularly subject to soil loss by water erosion under the traditional systems of bare-fallowing during the rainy season. Losses are promoted by the combination of intense storms and lack of plant cover. In research watersheds, introduction of a crop during the rainy season reduced erosion losses from 30 to 60 t ha^-1 year (Binswanger et al, 1980).

Erosion losses during the rainy season are also promoted by the low infiltration rates of Vertisols once the soil profile is filled to field capacity (Table 2). A high percentage of rain falling onto the soil will then be lost as runoff, with substantial risk of erosion.

Assessment of the extent of soil loss is hindered by the variable high intensity storms that cause the major water erosion losses. Losses under traditional monsoon fallowing are substantial; measurements at Sholapur, on land with a 1-2% slope, indicate that 50 cm of soil has been lost by erosion over the past 100 years (All India Coordinated Research Project for Dryland Agriculture, personal communication). The consequences are frightening: in the SAT environment, Vertisols have great potential for productive cropping because their stored water can carry crops through drought periods and support a crop in the post-rainy season. Erosion reduces the long-term productive capacity of the soil.

Table 1. Climatic characteristics in areas of Vertisols and associated soils.

Climatic particulars	Arida		Dry semi-arida			Wet semi-arida
	Rajkot	Bellary	Ahmednagar	Hyderabad	Tiruchi-rappelli	Bhopal	Indore
	(22° 18' N)	(15° 09' N)	(19° 05' N)	(17° 27' N)	(10° 46' N)	(23° 17' N)	(22° 43' N)
Annual mean rainfall (mm)	673.8	518.1	677.3	764.4	867.6	1208.9	1053.4
Annual mean temperature (°C)	26.8	27.6	25.3	25.8	28.9	25	24.4
Maximum summer temperature (°C)	39.4	38.1	39.4	38.7	37	38.4	39.9
Minimum winter temperature (°C)	19.2	17.4	13.1	13.4	21	10.1	9.6
Average minimum temperature (°C)	33.9	22.2	32	20	33	31.5	17.5
Percent of evapotranspiration covered by rainfall	30	29.8	42	63	41	77	58
Number of dry monthsb	10	10	6	8	9	4	6
Number of wet monthsc	1	0	2	3	2	3	4

a. Climatic classification by Troll (1966).

b. When available water (rainfall + water stored in soil by previous rain) covers less than half of potential evapotranspiration (PE).

c. When rainfall exceeds PE.

Source: Murthy et al (1982).

Table 2. Initial and equilibrium infiltration rates of a Vertisol at ICRISAT Center, near Hyderabad, India.

Time from start (fur)	Infiltration rate (mm hr-1)
0-0.5	76
0.5-1.0	34
1.0-2.0	4
After 144	0.21 + 0.1

Source: Virmani and Swindale (1984).

Traditional dryland farming systems

The use and management of Vertisols in India vary widely and depend upon the development of local technology. Flocks of sheep and goats graze on common lands during the cropping seasons, and on stubble and trash during the fallow. Manure is exchanged for feed. In the Indian SAT, the traditional cropping patterns on Vertisols and associated soils can be related to the constraints imposed by the combination of agroclimatic and soil characteristics. On the shallower vertic soils, cropping is confined to the rainy season because the limited water storage capacity is insufficient to support crop growth for very long after the rains have ceased. On Vertisols cropping patterns and land use vary. A few Vertisols are cropped during the rainy season, with the growth period extending beyond the end of the rainy season because long-duration cultivars are commonly used. However, many-perhaps 50% or more (Krantz and Singh, 1975) - are fallowed during the rainy season and produce a single crop grown on stored moisture in the post-rainy season.

Michaels (1981) has confirmed that rainy season fallows can be separated into:

· "Dry" fallows, in which the rainfall during the rainy season is unreliable and bare-fallowing is essential to attempt to accumulate sufficient water in the profile to grow a crop on stored water in the post-rainy season (exemplified by Sholapur, Table 3).

· "Wet" fallows, in which the rainfall during the rainy season is adequate to excessive and cropping during this season risks losses from waterlogging and flooding. Because maximum water storage in the profile is assured at the end of the rainy season, crops grown in the post-rainy season on stored moisture are relatively assured, although the productivity is low (exemplified by Hyderabad, Table 3).

Where rainy-season fallow is practiced the common crops grown in the post-rainy season are sorghum, safflower, wheat and chickpea. Examples of multiple cropping systems are intercrops (sorghum/oilseed) or a 2-year rotation of wheat-chickpea or wheat-linseed (Krantz and Singh, 1975). Sorghum may be grown throughout the Indian SAT, but production of wheat and chickpea is largely confined to the northern areas. Less common, but still important, are some specialist crops such as chilliest

Table 3. Reliability of a 90-day rainy season crop on three Vertisol areas (Probability expressed in percentage of years).

	Sholapur deep	Hyderabad deep	Akola medium deep
Annual rainfall (mm)	742	761	840
Probability of emergence before 15 July	65	85	92
Probability of seedling survival	49	76	80
Probability of good growing conditions throughout	33	62	66
Probability of adequate soil moisture for post-rainy season sorghum:
after rainy season crop	60	50	NAb
after rainy season fallow	80	83	NA

a. Available water-holding capacity of deep Vertisols is assumed at 230 mm and of shallow Vertisols at 120 mm.

b. NA = not applicable. The water-holding capacity is far too low to meet post-rainy season sorghum water needs.

Source: Binswanger et al (1980).

Where crops are grown during the rainy season, multiple cropping and sole cropping are practiced with many different crops (Spratt and Choudhury, 1978; Willey, 1981). Sorghum or cotton are commonly grown with pigeonpea as an intercrop, or sorghum in the rainy season may be followed by chickpea in post-rainy season. However, two crops spanning both seasons are grown on less than 10% of the Vertisol area (Krantz and Singh, 1975).

Fields are prepared with a single pointed-stick plough and a blade harrow 'bakhar'. Seed is either broadcast or placed through a seed tube attached to the plough. The only nutrient input is an occasional small application of manure. Cultivars are usually long season, tall, local landraces with characteristically low harvest indexes.

Although productivity is low, it is reasonably stable; use of inputs and loss risks are low. The system was satisfactory while population pressures were also low and population increases could be accommodated by expansion onto unused land. Increasing populations now require increasing productivity per capita and per unit of land.

Prevailing production constraints

The Indian farmer's traditional system has several major problems and limitations. The poor internal drainage of the heavy soils can severely restrict operations during the rainy season, especially if rainfall is excessive and/or slope of the land is minimal. The cultivation equipment is not versatile for operations under difficult soil conditions. The crop cultivars, although resistant to some pests and diseases, have limited yield potential and little ability to respond to inputs such as fertilizers. The farmers' crop options are often limited by the use of long duration cultivars. Animal production is a minor activity.

Improved cropping systems

Virmani et al (1982) have classified the Vertisols of central India into two broad production zones on the basis of annual rainfall (Figure 1):

· Unassured rainfall zone, including the drought-prone areas of Maharashtra and north Karnataka, which receives erratic rainfall ranging from 500 to 750 mm, equal to 40-48% of annual potential evapotranspiration.

· Assured rainfall zone, extending from Hyderabad to Gwalior in central peninsular India, which receives mostly assured rainfall ranging from 750 to 1250 mm, equal to 43-77% of annual potential evapotranspiration.

Different strategies are needed in each zone. In the unassured rainfall zone sorghum or safflower grown in the post-rainy season as sole crops have been most successful. Improved cultivars used with moderate doses of fertilizers and good weed control increase yields significantly (Rao and Rao, 1980).

Sowing 3 to 4 weeks before traditional dates (into late September) was found to increase average yields from 770 to 1870 kg ha^-1 in a long-term experiment by Randhawa and Venkateswarlu (1980). Improved use of available water was considered to be the reason for the increase.

Figure 1. The Vertisols areas of India where rainfall is assured and unassured

Cropping in the post-rainy season still leaves the land susceptible to erosion during the early rains. Various land treatments such as land smoothing, guide bunds or broadbeds-and-furrows, and short duration cover crops such as mung beans, have been recommended by various authors, but with limited success. Current research has concentrated on rotating crops with grassland or pasture, with buffer grass and stylo as the main grass and legume components, but this practice has not yet attracted much farmer interest. The same is true of rotation with farm forestry. Farmers prefer to seek access to irrigation water.

For the assured rainfall zone ICRISAT has developed an improved farming system, the key elements of which are:

· Growing the same crop in both rainy and post-rainy seasons. Each of the two crops is more productive than the previous single crop.

· Use of improved cultivars and improved cropping systems which increase the number of options for the farmer, including sole crops, sequential crops, and intercrops.

· Use of fertilizers, usually N and P. and less commonly Zn.

· Introduction of improved management techniques.

The key to the success of the system is improved management: the basic elements are:

· Shaping land to promote disposal of excess water by introduction of broadbeds-and-furrows and grassed waterways.

· Rough ploughing land immediately after the previous crop is harvested when there is still a little moisture remaining in the soil.

· Completion of seedbed preparation after the first pre-monsoon rain, which always occurs between harvest in January or February and onset of the southwest monsoon in June.

· More precise placement of fertilizer and seed.

These developments have required a great deal of research and planning. Prediction of areas where the rainy-season rainfall is assured and where dry seeding can be used, selection of improved crop species and cultivars, selection of the best combinations for improved cropping systems, and development of simple bullock-drawn equipment have all been significant tasks.

An important aspect of the improved system is that it contains a package of options for the farmer. While the physical productivity gains are greatest when all options are introduced (Table 4), the farmer has the option of selecting only those components which are desired or possible.

The basis of the technology developed at ICRISAT is a system of semi-permanent graded broadbeds-and-furrows laid out on a gradual slope (usually 0.4-0.8%). Each bed is slightly raised, acting as a 'minibund' for good moisture conservation and erosion control. The broadbeds are adaptable to a range of sowing arrangements to accommodate different crops; the number of rows per bed can vary from one to four, giving effective row arrangements from 150 to 30 cm. The furrow is shallow but provides good surface drainage to prevent waterlogging of the crops growing on the bed. Excess water is drained through a system of field drains and grassed waterways.

Table 4. Synergistic effect of variety selection, soil management, and fertilizer application in a maize/pigeonpea intercropping system on a Vertisol at ICRISAT (1976-77).

Treatment	Yield (kg ha)
	Local maize variety	Improved or hybrid maize variety	Pigeonpeaa
Traditional inputs and management	450		320
Improved soil-, and crop management alone	600		64
Fertilizer application alone	1900		452
Improved soil-crop management and fertilizer	2610		837
Traditional inputs and management		630	500
Improved soil-, and crop management alone		960	640
Fertilizer application alone		2220	540
Improved soil-crop management and fertilizer		3470	604

a. Pigeonpea variety was the same in ale experiments.

The two major cropping systems that have been developed at ICRISAT to utilise both the rainy and post-rainy seasons are:

· a "sequential" system of rainy-season maize or sorghum (two rows per broadbed) followed by a post-rainy season chickpea (four rows per broadbed); and

· an intercrop system of maize or sorghum and pigeonpea (one row of pigeonpea at the middle of the bed and one row of cereal on either side).

Maize has been the better cereal to use in these systems because it avoids the late-season disease problems of sorghum.

The yields of these two systems at ICRISAT over 4 years from operational scale watersheds of several hectares are given in Table 5. Improved seeds and adequate fertilizers were used as part of the technology. Both systems substantially outyielded the traditional rainy-season fallow system growing only a post-rainy season crop of chickpea or sorghum without the benefit of raised beds or improved seeds and fertilizers.

The good performance of the intercrop is worth noting. When the two improved systems were compared, the intercrops gave only a little less maize and rather more pulse than the sequential system, and gross returns were similar. The intercrop may be attractive in practical terms in that both crops are planted in one operation at the beginning of the rainy season. This avoids a possible problem with the sequential system in which the post-rainy season crop has to be established at the end of the rains when the upper soil layers may have dried out, and when the farmer has a peak labour demand to harvest his rainy-season crop. This is one of the reasons why the intercropping systems have given more stable net returns than the sequential system in these operational watersheds (Ryan et al, 1980).

Economic analyses of the results from 1976 to 1981 at ICRISAT have shown that the improved technology based on maize intercropped with pigeonpea can increase profits by about 600% compared with the traditional system based on rainy-season fallow followed by post-rainy season sorghum and chickpea. The improved system has generated profits averaging Rs. 3650 ha^-1 year^-1 over 5 years, compared with only Rs. 500 ha^-1 year^-1 from the traditional system. These profits represent a return to land, capital and management; the cost of implements has been deducted (Ryan and Sarin, 1981).

Table 5. Grain yields (kg ha^-1) from an intercrop system and a sequential system compared with traditional rainy-season fallowing from Vertisol operational scale watersheds at ICRISAT.

	1976-77	1977-78	1978-79	1980-81	Mean
Maize/pigeonpea intercrop system
Maize	3291	2813	2140	2918	2791
Pigeonpea	783	1318	1171	968	1060
Maize-chickpea sequential system
Maize	3116	3338	2150	4185	3197
Chickpea	650	1128	1340	786	976
Traditional fallow and single post-rainy season crop
Chickpea	543	865	532	596	634
Sorghum	436	377	555	563	483

On-farm studies of the improved systems

On-farm studies of the improved systems developed at ICRISAT began in 1981. The objectives were to:

· verify whether the ICRISAT experience could be replicated in farmers' fields;

· evaluate the performance of the various technology options;

· test the ability of delivery systems to support demands of the improved systems and to utilise the increased production; and

· study the technical and economic performance of the options under farmer conditions.

The initial on-farm trials were conducted in one village with a soil type similar to that at ICRISAT. It was conducted by farmers under ICRISAT supervision and monitored by the state Department of Agriculture. Later, the trials were expanded to 28 locations involving 1406 farmers in four states: some trials were supervised by ICRISAT, others were supervised by the state Departments of Agriculture or other agencies and monitored by ICRISAT, and others were handled without any direct ICRISAT involvement.

Von Oppen et al (in press) analysed the ICRISAT-managed trials in the states of Andhra Pradesh, Karnataka and Madhya Pradesh (Table 6). They found the results to be consistent over time. The improved cropping systems yielded 3000-4400 kg ha^-1 against 500-700 kg ha^-1 with traditional systems. Average gross returns were four to five times those of traditional systems. Except for the first year in Madhya Pradesh where some farmer recommended cropping systems failed, marginal rates of return were 106 to 381%. Generally a cereal/pigeonpea intercrop performed better than a cereal-chickpea sequential crop. At all locations the improved systems showed a coefficient of variation of gross profits lower than those for the traditional systems. This indicates reduced risk with the improved technology.

Table 6. Economic performance of Vertisol technology at ICRISAT collaborative on-farm test sites: 1981/82 to 1983/84.

Test site and year	Improved technology		Traditional technology		Marginal rate of return
	Operational cost	Gross profits	Operational cost	Gross profits
	(Rs ha-1)	(Rs ha-1)a	(Rs ha-1)	(Rs ha-1)	(%)
Taddanpally, Andhra Pradesh
1981/82	1181	3055	595	1625	244
1982/83	1035	3957	448	1722	381
CV of gross profits (%)		42		50
Sultanpur, Andhra Pradesh
1982/83	1062	3576	448	1722	302
CV of gross profits (%)		37		50
Farhatabad, Karnataka
1982/83	1194	3323	1142	2186	-b
1983/84	1226	4494	1188	2207	-b
CV of gross profits (%)		23		31
Begumgunj, Madhya Pradesh
1982/83	2348	1172	866	786	26
1983/84	2321	2743	1250	1611	106
CV of gross profits (%)		76		89

a. Profitability is measured in gross profits. Indian Rs. 13 = Us $1.00.

b. The differences in operational cost are too small to get a meaningful value for marginal rate of return.

Source: Von Oppen et al (in press).

Overall the on-farm trials gave substantial increases in productivity and rates of return and testified to the overall viability of the improved systems. However, the wide range in rate of return from 25 to over 2000% indicated a need for closer monitoring of the components of the technology and particularly of the crop combinations suggested by the farmers. Other components which need attention are:

· contribution of broadbeds-and-furrows to long-term increases in productivity and erosion control;

· efficiency and utility of the wheeled tool carrier and other implements;

· methods of controlling the pigeonpea pod-borer (Heliothis armigera);

· dry seeding; and

· cropping system rotations.

Early adoption by farmers of the new technologies

Farmers do not all adopt new technologies, even when they appear well-suited to their conditions, and those who do so, adopt the technologies piecemeal and at different rates. Two years after ICRISAT managed on-farm trials at Begumgunj village in Madhya Pradesh, Foster et al (1987) surveyed the response by farmers (Table 7). Double cropping, the most important innovation, has been well adopted. So have its economically most productive components: rainy-season soybeans, improved cultivars and increased use of chemical fertilizers and pesticides. Components contributing to improved workability or water management were not adopted.

Similar results were obtained in separate studies conducted as part of a large watershed development project carried out near Indore in Madhya Pradesh by a British technical team in association with the All India Coordinated Research Project on Dryland Agriculture (AICRPDA) and the College of Agriculture at Indore (Raje, 1983). The area under double cropping increased over the 5-year period of the project (1975/76 to 1979/80) from 190 to 1224 ha and the cropping intensity from 103 to 139% without the addition of any irrigation.

The problems and constraints to adoption of parts of the package as perceived by farmers were both technical and institutional. Prominent among the former were:

· Weed control imposes problems in several of the double cropping systems tried.

· Pest and disease controls were not fully integrated into the initial packages of practices and hasty ad hoe solutions created bad impressions among the farmers.

· Failure of late season rains or labour bottlenecks created problems with sequential post-rainy season crops. Farmers were not always willing to accept intercropping as a substitute because post-rainy season cereals (wheat in the north and sorghum in the south) are important for subsistence food and forage.

· Farmers perceive dry seeding as a risky practice because early season rains are erratic in some regions, the practice is not suited to all the crops that farmers wish to grow, and interactions with pests, weeds and diseases are sometimes adverse.

· Lack of bullock power. Many small farmers do not have bullocks and rental of bullocks is not common.

· The wheeled tool carrier (WTC) that was part of the ICRISAT package was too expensive. The WTC is a form of intermediate technology between traditional animal-drawn implements and tractor-drawn modern equipment. Lack of credit, lack of bullock power, small size of holdings and the intermediacy of the WTC all mitigate against its use.

Table 7. Use of components of the double cropping technology package in Begumgunj, Madhya Pradesh, by 18 watershed and 7 non-watershed farmers in 1986/87.

Practice	18 watershed farmers			7 non-watershed farmers
	Number using before 1982a	Adopting during field trials	Number using in 1986-87	Number using in 1986-87
Rainy season soybeans dryland	4b	14b	13c	4c
Dryland double cropping	Probably none	17	9+4d	1+3d
Summer ploughing	18	-	18	6
Improved drainage furrows	0	18	2	0
Broadbeds	0	18	0	0
Dry rainy season sowing	0	8	1	0
Improved seed	3	13	16	4
Use of chemical fertilizer	4	11	15	5
Using recommended dose of fertilizer	-	-	4	1
Mixing seed and fertilizer	All who use fertilizer at seeding time
Row seeding rainy season crop	1	14	14	5
Chemical plant protection	1	6	7	6
Use of wheeled tool carrier	0	18	0	0

a. ICRISAT field trials began in 1982.

b. Includes wet and dryland.

c. Including those growing soybeans on land that can be irrigated, 23 of 25 farmers grew soybeans in 1986-87.

d. The second number indicates the number who planned to double crop but had to fallow in the post-rainy season because of a moisture shortage.

Source: Foster et al (1987).

Prominent among the institutional constraints were:

· Supply systems for fertilizers and agricultural chemicals are weak in the dryland areas.

· Credit needs to be expanded. Short-term credit should embrace both rainy and post-rainy seasons. Medium-term credit is needed to enable bullocks and WTCs to be purchased and long-term credit is needed for land development. Weakness of cooperatives, a subsidy orientation among farmers and a poor record of credit repayment-aided and abetted by politicians-contribute to the lack of both inputs and credit.

· Extension services are sometimes, but not always, inadequate. Suitably trained extension personnel may not be available or their abilities to train farmers may be inadequate. Not all farmers are responsive or willing to cooperate.

· Local marketing and distribution systems may be unable to cope with the increased production of less-favoured crops such as sorghum or maize. Some on-farm trials failed because local authorities did not anticipate the increased supplies of coarse grains, and catastrophic price reductions occurred.

The demand for farm labour increased with the improved technology by 300 to 400 hours ha^-1. Even in rural India, where unemployment rates are high, the technology faced labour bottlenecks for weeding during harvest and during the mid-season harvesting and planting of sequential crops.

The components that farmers have been less willing to adopt all relate in one way or another to land and water management: improving soil filth or reducing puddling when the soil is wet, improving water infiltration into the soil, reducing runoff, and improving drainage. To make these improvements, the farmer must change his style of farming: he must grade and shape the land, install grass-protected drainageways, use more efficient but more expensive equipment, and ensure that field drainage is connected to community drainage channels and the regional drainage system. Short-term economic benefits of these improvements, although significant, are less than the benefits of using improved seeds and fertilizers. Furthermore, the benefits from reduced erosion and water control accrue to society as much as to the farmer, perhaps even more so.

It is not difficult to understand that the farmer is reluctant to adopt these practices. If society is to share the benefits it should also share the cost. Greater efforts by local or state government are required. Tax revenues may appropriately be used to ensure the adoption of these valuable practices. Government must also undertake public works needed to improve community and regional drainage canals.

Generally, the components adopted from the packages of practices were those that were already known or in use to some extent. The on-farm trials raised farmer awareness of these practices. Hence those practices such as improved drainage furrows and dry seeding prior to the rainy season at Begumgunj (Table 7), which have been adopted to a limited extent, may spread over time or if a special extension effort is made, particularly if the perceived technical problems can be overcome and the most important institutional constraints are removed. It is also important to remember that when farmers are attracted to a new innovation, for whatever reason, they will themselves find ways to make it succeed.

Conclusions

Vertisols are widespread in India. They occur in arid, semi-arid and subhumid climates and generally have potentials far above their present use in traditional agriculture. These soils are generally used to grow annual crops. In many areas the soils are fallowed during the rainy season and are then subject to serious erosion from high intensity storms and the lack of plant cover. Grazing and farm forestry are minor activities at present.

Two broad production zones have been delineated on the basis of annual rainfall; the boundary lies approximately at 750 mm. Improved cropping systems have been devised for both zones. For the drier zone, use of improved cultivars of sorghum and safflower grown in the post-rainy season with the addition of chemical fertilizers improves yields and profits, and resists drought better than traditional cropping systems. Unfortunately the soils remain susceptible to erosion during the early rains. There seems to be a place for grassland or farm forestry in rotation with annual crops, but this has not yet received adequate research attention.

In the more assured rainfall zone, systems for cropping the soils in both the rainy and post-rainy seasons have been successfully developed. Farmers are adopting double cropping and the related use of improved cultivars and fertilizers, particularly in areas where the crops produced-soybeans, wheat and certain pulses-are in demand or receive effective government price support.

Other constraints to increased use of double cropping are both technical and institutional. Weed, disease and pest control are the major technical problems, while input supplies, labour, credit and extension services are the major institution ones.

Farmers have been reluctant to accept improved tillage practices and land treatments that improve surface drainage. Farmers seldom perceive drainage as a problem. Furthermore the economic benefits are long term and thus unattractive compared to the use of improved cultivars, fertilizers and even pesticides. Reduced soil erosion and improved flood control are benefits that accrue more to society as a whole than to the individual farmer.

References

Binswanger H P, Virmani S M and Kampen J. 1980. Farming systems components for selected areas in India: evidence from ICRISAT. Research Bulletin no. 2. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, India. 44 pp.

Dudal R and Bramao D L. 1965. Dark clay soils of tropical and subtropical regions. FAO Agricultural Development Paper no. 8. FAO (Food and Agriculture Organization), Rome.

Fitzpatrick E A. 1980. Soil classes of the world. Vertisols. In: E A Fitzpatrick, Soils: their formation, classification and distribution. Longman, London and New York. pp. 274-279.

Foster J H. Kshirsagar K G. Bhende M J. Rao V B and Walker T S. 1987. Early adoption of improved Vertisol technology options and double cropping in Begumgunj, Madhya Pradesh. Economics Group Progress Report No. 77. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, India. 81 pp.

Krantz B A and Singh S. 1975. Cropping patterns for increasing and stabilizing agricultural production in the semi-arid tropics. In: International Workshop on Farming Systems, 18-21 November 1974, ICRISAT, Hyderabad, India. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, India. pp. 217-248.

Michaels G. 1981. The determinants of kharif fallowing on the Vertisols in semi-arid tropical India. PhD thesis, University of Minnesota, St. Paul, Minnesota, USA.

Murthy R S. 1981. Distribution and properties of Vertisols and associated soils. In: Improving the management of India's deep black soils. Proceedings of the Seminar on Management of Deep Black Soils for Increased Production of Cereals, Pulses, and Oilseeds, New Delhi, India, 21 May 1981. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, India. pp. 9-16.

Murthy R S. Bhattacharjee J C, Landey R J and Pofali R M. 1982. Distribution, characteristics and classification of Vertisols. In: Vertisols and rice soils of the tropics. Twelfth International Congress of Soil Science, New Delhi, India, 8-16 February 1982. Indian Society of Soil Science, New Delhi, India. pp. 3-22.

Raje S R. 1983. Operational research: the Indo-UK experience at Indore. In: Dryland agricultural research in India: thrust in the eighties. All India Coordinated Research Project for Dryland Agriculture. Hyderabad, Andhra Pradesh, India. pp. 171-186.

Randhawa N S and Venkateswarlu J. 1980. Indian experiences in the semi-arid tropics: prospect and retrospect. In: V Kumble (ed.), Proceedings of the International Symposium on Development and Transfer of Technology for Rainfed Agriculture and the SAT Farmer, 28 August - 1 September 1979, ICRISAT Center, India. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, India. pp. 207-220.

Rao V R and Rao M S R M. 1980. Rainfed agriculture. In: Research on soil and water conservation in semi arid deep black soils. Monograph no. 1. Central Soil and Water Conservation Research and Training Institute, Bellary, Karnataka, India. pp. 77-143.

Ryan J G and Sarin R. 1981. Economics of technology options for Vertisols in the relatively dependable rainfall regions of the Indian semi-arid tropics. In: Improving the management of India's deep black soils. Proceedings of the Seminar on Management of Deep Black Soils for Increased Production of Cereals, Pulses, and Oilseeds, New Delhi, India, 21 May 1981. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, India. pp. 37-57.

Ryan J G. Sarin R and Pereira M. 1980. Assessment of prospective soil-, water-, and crop-management technologies for the semi-arid tropics of peninsular India. In: Proceedings of the International Workshop on Socioeconomic Constraints to Development of Semi-Arid Tropical Agriculture, ICRISAT, Hyperabad, India, 19-23 February 1979. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, 324, India. pp. 52-72.

Spratt E D and Choudhury S L. 1978. Improved cropping systems for rainfed agriculture in India. Field Crops Research 2:103-126.

Troll C. 1966. Seasonal climates of the earth: the seasonal course of natural phenomena in the different climatic zones of the earth. In: E Rodenwaldt and H J Jusatz (eds), World maps of climatology. 3rd ed. Springer-Verlag, Berlin. pp. 19-28.

Virmani S M and Swindale L D. 1984. Soil taxonomy: an aid to the transfer of improved farming systems for Vertisols in the semiarid tropics. Soil Taxonomy News 7:3-9.

Virmani S M, Sivakumar M V K and Reddy S J. 1982. Rainfall probability estimates for selected locations of semi-arid India. 2nd ed. Research Bulletin no.1. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, India. 180 pp.

Von Oppen M, Ghodake R D, Kshirsagar K G and Singh R P. (in press). Socioeconomic aspects of transfer of Vertisol technology. Presented at the Workshop on Management of Vertisols for Improved Agricultural Production, ICRISAT Center, Patancheru, Andra Pradesh, India, 18-22 February 1985. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, India.

Willey R W. 1981. A scientific approach to intercropping research. In: Proceedings of the International Workshop on Intercropping, ICRISAT, Hyderabad, India, 10-13 January 1979. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics), Patancheru, Andhra Pradesh, India. pp. 4-14.

Young A. 1976. Vertisols. In: Tropical soils and soil survey. Cambridge University Press, Cambridge and New York. pp. 180-191.