Document(s) 15 de 38

Abstract: Considerable international and national breeding effort has been exerted to develop suitable and acceptable rice cultivars for the unfavorable production environments of Eastern India. However, the existing research suggests that farmers are still using their traditional landraces and the production of rice from these ecosystems has increased only marginally over the past decades.

Therefore, increasing the productivity through improved rice varieties remains a challenge. The answer to this challenge might lie in the traditional landraces cultivated by farmers. As landraces often have important traits that give them resistance to many biotic and abiotic stresses, identification of particularly adapted types should help in developing more promising materials for the rainfed lowlands of Eastern India.

To obtain information on (1) the diversity of rice cultivars grown and (2) the reasons for a limited acceptance of modern cultivars, a farm survey was conducted in a typical rainfed village in Orissa. Findings indicate that farmers are using a diversity of landraces in this complex and heterogeneous environment. As many as 12 rice cultivars are grown by a single rice farmer and farmers' have intimate knowledge of the characteristics of rice cultivars which fit well into their specific ecological niches.

Farmers in the study village perceived that improved cultivars perform better under better fertility regimes and when fertilizer is applied. On the other hand, the performance of traditional cultivars is superior under low fertility conditions and they are better in sustaining the soil resources over a long period of time. They also perceived that the performance of traditional cultivars is better under various biotic and abiotic stresses.

Although breeders may use local materials in their breeding programs to develop improved varieties, the local landraces are unlikely to be completely replaced in this rainfed environment. A breeding strategy which helps maintain useful diversity is what is recommended.

Introduction

Increasing rice productivity through improved varieties continues to remain a challenge, especially in rainfed areas. The answer to this challenge might be sought in the traditional landraces grown. Often landraces have important traits that give them resistance to many biotic and abiotic stresses. Identification of such landraces, and elicitation of farmers perceptions about their traits vis-à-vis that of improved cultivars, should generate information that can be used by rice breeders in focussing research on developing promising and potentially adoptable materials (that is, having characteristics preferred by farmers).

The objective in this paper is to provide a picture of the diversity of rice cultivars grown in a rainfed village in Eastern India and to analyze farmers' perceptions about the characteristics of these cultivars (both modern and traditional).

Materials and methods

A census survey was conducted to get information on resource endowment of the households of Garh Madhupur village. The census data formed the basis for a random selection of 50 respondents from 291 households. Two investigators holding masters degree in agriculture were recruited and stationed in the village. This gave the benefit of building excellent rapport with the farmers. Data for the years 1993 and 1994 were collected through direct interviews by using an open-ended format.

Results and discussion

Garh Madhupur is located at a distance of two km on the north side of Daitari-Paradeep express highway and about 120 km away from Bhubaneswar, the capital of Orissa state. The population of Garh Madhupur in November 1994 comprised 2237 persons living in 291 households.

No rainfall data has ever been recorded at Garh Madhupur village. Therefore, the average daily rainfall recorded at Jarka and Barchana centers (7 and 15 km away, respectively, from Garh Madhupur) was used to represent the rainfall pattern in the Garh Madhupur area. The average annual rainfall is 1512 mm with a coefficient of variation (CV) of 12% for 15 years from 1980 to 1994 (Figure 1). On an average, 86% of the total rainfall is received during the months of June to October. The average annual rainy days are 79 with a CV of 14%.

Figure 1: Annual rainfall in Garh Madhupur area, 1980-1994.

Although the extent of average annual precipitation is adequate for farming, its capricious nature is very clear from Figure 1. Four years among the last 15 years were relatively dry, having less than 1350 mm of rains, while three years were very wet, having more than 1700 mm. This erratic distribution often causes drought and/or floods during crop growth period and results in substantial losses for dryland farmers.

Rice is traditionally grown in two well defined seasons, namely kharif and dalua. Of these two, kharif (rainy) is the most important rice season. The kharif rice is the main crop, covering over 85% of the total rice area, and depends entirely on the southwest monsoon. It is sown in June and harvested in October-December, depending upon the duration of the cultivar and topography of the field. The dalua (summer) crop coincides with the dry season and depends entirely on irrigation. The source of irrigation water is tank. The dalua season stretches from December-January to April-May. Farmers grow only high-yielding varieties during this season.

Importance of traditional cultivars

Table 1: Rice area under local and improved cultivars during Kharif season.
Cultivar details	1993	1994
Local cultivars No. of farmers growing local cultivars No. of local cultivars grown % area under local cultivars	50 33 86	50 33 86
Improved cultivars No. of farmers growing improved cultivars No. of improved cultivars grown % area under improved cultivars	21 14 14	19 11 12

Respondents planted 33 traditional cultivars on their farms during the kharif season. Of these, half a dozen popular cultivars cover more than 50% of the area under traditional cultivars (Table 2). The dominance of these cultivars could be due to their superiority in adapting to the variable production environment of rainfed agriculture.

Table 2: Local cultivars grown during kharif season.
Local name of cultivar	1993			1994
	No. of farmers	Area		No. of farmers	Area
		ha	%		ha	%
Dhinkiasali	25	8.66	13.34	24	10.50	15.74
Chamarmani	18	5.96	9.18	25	7.56	11.33
Kalachaki	18	7.36	11.34	19	7.32	10.97
Khoda	17	5.97	9.20	16	6.00	8.99
Champa	15	4.92	7.58	12	4.24	6.35
Champasali	12	3.62	5.58	12	3.69	5.53
Bankisaru	10	2.18	3.36	12	2.99	4.48
Saruchina	10	3.32	5.11	11	4.07	6.10
Nagara	11	2.85	4.39	9	3.93	5.89
Jiresali	7	1.69	2.60	6	1.31	1.96
Kendumanjia	6	2.37	3.65	3	1.62	2.43
Parbatkaya	5	2.02	3.11	4	1.23	1.84
Chingudi	4	1.42	2.19	4	1.32	1.98
Nimei	4	1.18	1.82	4	0.57	0.85
Kakudimanji	3	0.40	0.62	5	0.91	1.36
Dudhasara	4	0.73	1.12	3	0.81	1.21
Chinamali	4	1.82	2.80	2	1.21	1.81
Matiasaluri	4	1.13	1.74	2	0.61	0.91
Hati Mahal	3	0.91	1.40	3	1.01	1.51
Salua Gaja	3	1.00	1.54	3	0.87	1.30
Betanashia	3	1.96	3.02	2	1.82	2.73
Kalma	3	0.85	1.31	4	0.83	1.24
Bangaramadhavi	2	0.68	1.05	2	0.47	0.70
Ranga Seuli	2	0.45	0.69	1	0.10	0.15
Others	6	1.47	2.26	11	1.74	2.61
Total		64.92	100.00		66.73	100.00

Landforms, soil types and cultivar selection

Table 3: Farmer's classification of and percent area under different landforms, soil types and cultivars on selected farms.
Particulars	Landform: Terraces
	Dhipa (Upper)	Majhili (Middle)	Khala (Lower)
% of total area	35	40	25
Soil type1 Balia (%)2 Kelua (%) Dorosa (%)	24 72 4	38 50 12	7 76 17
Cultivar
Preferred group	Laghu dhan (short)	Madham avadhi dhan (medium)	Bada dhan (long)
Duration (days)	<145	146-165	>165
Common cultivars grown	Jiresali, Chinguda, Champa, Parijat	Dhinkiasali, Kalachaki, Saruchina, CR 1009	Khoda, Champesali, Bangara Madhavi, T 1242
1. Balia, kelua and dorosa are local terms used to represent sandy, clayey and silty or loamy soils, respectively. 2. Percentage calculated on the basis of total area in each landform.

Farmer classification: landforms: Farmers in the study village identified three major landforms: dhipa zami (upper terraces), majhili zami (middle terraces), and khala zami (lower terraces). The lower terraces were subdivided into bahaliapata (very low) and pata zami (even lower) fields.

Dhipa zami or upper terraced land is characterized by drought stress that may occur any time during the season. The freely drained profile of upper terraces lose much of their rain water by run-off and seepage. Majhili zami (middle terraces) is good paddy land and is affected less frequently by drought and/or flood and submergence. The khala zami or lower terraced fields are flood-prone and complete submergence for 10 to 12 days during the periods of heavy rainfall is common. The lower terraces are poorly drained and excess water, rather than water deficit, is the most serious problem. The pata zami is kept fallow during kharif season due to the stagnation of flood water.

Farmer classification: soil types: The existence of locally recognized soil types is well documented by many researchers (eg. Barrow, 1987; Bellon and Taylor, 1993; Dvorak, 1988). Garh Madhupur farmers divide their soils into three major categories: balia (sandy), kelua (clayey) and dorosa (silty or loamy). Similar classification by the farmers was also reported by Fujisaka et al (1991) in Cuttack district of Orissa State. Balia (sandy) soils have lower water holding capacity and farmers claim that it suffers from drought earlier than any other soil types. Dorosa soils are characterized by high water holding capacity and are usually found in middle and lower terraced fields. Kelua (clayey) soils fall between balia and dorosa in terms of water holding capacity and are dominant and distributed in all landforms.

Farmer classification: rice cultivars: Farmers divide rice cultivars into three groups: short (laghu dhan), medium (maadham avadhi dhan) and long (bada dhan) duration cultivars. Farmers' definition of short (less than 145 days), medium (146-165 days) and long duration (more than 165 days) does not necessarily match with that of plant breeders. In this paper, we use farmers' classification.

Farmer practice: matching of cultivars to environments: Generally, long duration tall cultivars are sown in lower terraces. These are indica cultivars having photosensitive characteristics. These cultivars thrive well under ill-drained or submerged conditions. Moreover, long duration cultivars are slow-growing, have good tillering abilities and adjust well to changes in climatic conditions, thereby providing low but stable yields.

The long duration cultivars are sown in early June. The crop matures and becomes ready to harvest after the cessation of monsoon rains and recession of water from the field by December, a relatively dry month of the year. Moreover, the abundant sunlight and cool nights, during the period of reproductive stage, provides optimal conditions for grain filling and better milling.

There are only a few improved cultivars that are grown in lower terraces. Farmers perceive that improved cultivars do not perform well in flood-prone fields. It has been also reported by scientists that flood in many parts of eastern India restricts the adoption of high-yielding varieties (Muralidharan et al. 1988).

Medium duration cultivars are grown mostly in middle terraces. Cultivars such as Parbatkaya and Nimei (duration 146 to 155 days) are grown in fields that are located in the upper portions of middle terraces or in fields having sandy soils, whereas cultivars such as Dhinkiasali, Kalachaki and Saruchina (duration of 155 to 165 days) are grown in fields having kelua and dorosa soils.

Medium duration cultivars are sown in early June after the completion of sowing of longer duration cultivars in lower terraced fields. These cultivars such as Kalachaki and Saruchina are valued for eating qualities by the Garh Madhupur farmers and are usually retained for home consumption.

There are not many medium duration improved cultivars grown in the village. Improved cultivars such as CR 1009 and Swarna are always planted (90% of area) in clayey or loamy soils. Farmers preferred to grow improved cultivars on these soil types as they believe these varieties need soils with higher fertility and moisture holding capacity.

The short duration cultivars can be further subdivided into two groups, the first group having a duration of less than 120 days and the second group having a duration from 120 to 145 days. Early maturing short duration cultivars are usually sown in the uppermost terraces to escape late season drought. Most of the improved cultivars adopted by the Garh Madhupur farmers belong to this category. Important among them are Parijat, Annada and Pathara. The second group of cultivars such as Chingudi, Tike Narda, and Champa on the other hand, are allocated to middle and lower portions of upper terraces.

Number of rice cultivars grown

Farmers in the study village almost invariably grow more than one rice cultivar on their farms, with the number ranging from two to more than 10 (Table 4). More than 70% of the respondents cultivated two to five cultivars, where as 20% sample farmers went for six to eight cultivars.

Table 4: Access to landforms and number of cultivars grown.
Number of cultivars grown by each farmer	Number of farmers cultivating:
	Only 1 landform		Only 2 landforms		All 3 landforms
	1993	1994	1993	1994	1993	1994
1 2 3 4 5 6 7 8 9 10 or more	1 2 4 2 2 1 0 0 0 0	0 4 4 1 2 0 1 0 0 0	0 5 2 7 7 2 1 2 0 0	0 4 6 7 4 2 1 1 0 1	0 1 1 0 3 1 2 1 2 1	0 0 1 1 3 3 1 1 1 1
Average number of cultivars grown by each farmer	3.42	3.42	4.38	4.35	6.42	6.33

To the extent that one of the reasons for varietal diversification is to match the varietal requirement to a specific environmental niche, varietal diversification can be expected to increase with environmental differentiation. To examine this hypothesis, farmers were divided into groups operating within one or more sub-ecosystems as defined by farmers. Detailed environmental characterization to cluster "homogeneous" sub-ecosystems was beyond the scope of the study. We simply used farmer classification of upper terraces, medium terraces and lower terraces as the basis for sub-ecosystems classification. Data in Table 4 lend some support to the hypothesis that varietal diversity increases with an increase in environmental heterogeneity. The average number of cultivars planted by farmers operating land in all three sub-ecosystems is highest at six. Farmers operating in two sub-ecosystems grow four cultivars on an average, while farmers having land in only one sub-ecosystem grow only three cultivars.

In addition to niche matching, farmer interviews indicated that risk reduction was also a major reason for cultivar diversification. Nearly all farmers reject exclusive reliance on a single cultivar having specific characters which may prove disadvantageous under the unreliable climatic and hydrological conditions which characterize the rainfed lowlands. Staggering of labor demand and differential end uses were also identified by several farmers as reasons for growing multiple varieties.

The implication of farmer preferences for varietal diversity is that the chances of wider adaptability of a single improved cultivar in rainfed lowlands may be limited. A set of improved cultivars having different traits may be more acceptable to farmers. Adequate consideration should be given to this point in our efforts to develop improved cultivars for rainfed lowlands.

Farmer's perceptions on traditional and improved cultivars

Soils and fertilizers: The adoption of improved varieties seems to be positively correlated to the fertility status of the soil. Ninety percent of the farmers said that they grow improved cultivars on their most fertile piece of land (Table 5). Ninety-four percent of the farmers applied chemical fertilizers to improved cultivars. Besides this, more than 90% of the farmers thought that traditional cultivars did better on low fertility fields and did not erode the fertility status of their fields.

These observation suggest that: (1) farmers perceive that improved cultivars perform better under better fertility regimes; (2) improved cultivars perform better only when chemical fertilizers are applied; (3) the performance of traditional cultivars is superior to that of improved cultivars under low fertility conditions; and (4) traditional cultivars are better in sustaining the fertility of soils.

Most of the farmers in rainfed rice growing areas are subsistence farmers with limited financial resources for purchased inputs such as fertilizers. Their perception that modern varieties need more fertilizers may be a factor constraining the widespread adoption of such varieties. It would be useful to establish whether or not improved cultivars perform equally well under low fertility situations. If modern varieties are not inferior in low fertility situations, this message needs to be communicated to farmers in an appropriate way. On the other hand, if traditional varieties are found to be superior under low fertility conditions, breeders need to consider performance under low fertility conditions as a selection criteria.

Table 5: Farmer's perceptions on traditional and improved cultivars.
Perception	% of respondents
	Traditional	Improved	Both	Don't know
Soils
Most fertile land is allocated to	8	90	2	0
Usually fertilizer is applied to	0	94	4	2
Can do better in low fertile land	92	4	2	2
Soil fertility not eroded quickly by	94	0	2	4
Biotic stress
Tolerance to pest	88	2	10	0
Tolerance to disease	82	2	4	6
Management
Needs better management	0	94	6	0
Grown in fields nearer to house	40	52	8	0

Tolerance to pests and diseases: Farmers have identified Dhinkiasali as the most resistant cultivar to insect pests, followed by Kalachaki and Nagara. Moreover, farmers also perceive that these cultivars are tolerant to diseases. When it comes to the traditional and improved cultivars, more than 80% of farmers responded that traditional cultivars are more tolerant (Table 5).

In high rainfall and flood-prone areas of Eastern India, the severity of pests and diseases and difficulty in controlling them by chemical treatments highlight the need for breeding cultivars for multiple resistance.

Management: Farmers in the study village as well as rice experts (Singh et al., 1994) believe that timely completion of various crop management operations is relatively more crucial for improved cultivars than for traditional cultivars. Forty-seven out of 50 respondents felt that improved cultivars need better crop management than traditional ones. Fifty-two percent of farmers reported that they grow improved cultivars in those fields that are relatively nearer to the house so that they can monitor the crop very closely and frequently (Table 5). In addition, fields nearer to house may receive a larger quantity of organic manure due to lower travel time. It would be useful to study farmers' crop and fertilization practices to shed further light on why farmers grow improved varieties closer to the house.

Summary and implications

There is a noticeable tendency among farmers to grow more than one rice cultivar on their farms. The number ranges from two to more than 10. Diversity is higher among farmers with more heterogeneous environments (soil types and landforms). Cultivar diversification is an important practice followed by rainfed lowland farmers for a variety of reasons, such as environmental matching, risk reduction, avoiding labor bottlenecks and obtaining a range of product qualities.

Farmers perceive that improved cultivars perform better under better fertility regimes and when fertilizer is applied, while traditional cultivars perform better under low fertility. Farmers' perceptions on the suitability of improved cultivars to fertile fields and need to apply chemical fertilizers to grow improved cultivars may limit the adoption of improved varieties. If improved varieties do indeed perform poorly under low fertility conditions, it would be appropriate to expand breeders' selection criteria to include performance in low-input situations.

Most farmers believe that traditional cultivars are more tolerant to pests and diseases than the improved ones. Susceptibility of available improved varieties to biotic stresses may be a factor limiting their adoption. Therefore, opportunities for exploiting resistance available in traditional cultivars to breed improved cultivars with multiple resistance should be explored.

Farmers believe that improved cultivars demand timely operation and management to realize their full potential. If this is the case, flexibility in crop management operations would be an important trait to select for.

In areas with unreliable climatic and hydrological conditions, the key element in farmer survival strategies has been the ability to match the diverse genetic material with the agro-hydrological conditions existing on individual rice fields. The availability of diverse genetic material and farmers' sound knowledge of landforms, soil types and hydrology gives them immense ability and flexibility in dealing with the risky production environments of rainfed lowlands. In risky production environments, breeding for wide adaptability may not be a very appropriate strategy. What is needed perhaps is to develop a range of improved cultivars with different characteristics so that farmers can choose combinations with traits that are most appropriate for their field conditions. Breeding efforts could perhaps be made more efficient by using farmers' knowledge about appropriate plant types and agro-hydrology through a more direct involvement of farmers in the breeding process.

Although breeders may use local materials in their breeding program to develop improved varieties, the local landraces are unlikely to be completely replaced in this rainfed environment. Maintenance of a degree of diversity of traditional cultivars is an important survival strategy of rainfed farmers. A breeding strategy which helps maintain useful diversity of cultivars is recommended.

References

Bellon, M.R. and J.E. Taylor, 1993. Farmer soil taxonomy and technology adoption. Economic Development and Cultural Change, 41:764-786.

Dvorak, K.A., 1988. Indigenous soil classification in semi-arid tropical India. Economics Group Progress Report No.84. ICRISAT, Patacheru, Andhra Pradesh, India.

Fujisaka, J.S, K.T. Ingram, and K. Moody, 1991. Crop establishment (Beusani) in Cuttack District, India. IRRI Research Paper Series No. 148. International Rice Research Institute, Manila.

Govt. of Orissa, 1993. Orissa Agricultural statistics 1990-91. Directorate of Agricultural and food production, Orissa, Bhubaneswar.

Muralidharan, K, J. E. Shinde, I.C. Pasalu, S. Venkataraman, A.S.R. Prasad, M.B. Kalode, A.V. Rao and A.P.K. Reddy, 1988. Rice in Eastern India- a reality. Oryza, 25: 213-245.

Singh, R.K., V.P. Singh and C.V Singh. 1994. Agronomic assessment of `beushening' in rainfed lowland rice cultivation in Bihar, India. Agriculture, Ecosystems, and Environment. 51:271-280.

Footnotes:

Document(s) 15 de 38