Introduction

Improved oat cultivars have been introduced to Asia over the last 20 years from countries as diverse as Canada, Australia, New Zealand and Europe. The overall impact of the introduction of such oat cultivars on the Himalayan region was reviewed by Dr E. John Stevens et al. in a presentation to the 6^th International Oat Conference, (Christchurch, New Zealand) in November 2000 (The importance of oats in resource-poor environments).

In particular, Stevens et al. (2000) described the impact of introduced new fodder oat cultivars on Nepal. It outlines the highly significant and strategically important role oats have had in developing new livestock feeding regimes across a wide range of ecologies, especially within the poorer regions of the countries bordering the Himalayas. In these areas they are used either as green feed or oaten hay.

Considerable genotype by environment interaction has been noted across latitude, altitude and season sequences. Some cultivars produce significantly betters in certain areas and under certain management regimes.

So far the performance of these cultivars has been poorly documented and only modestly exploited due to limited resources and lack of a regionally co-ordinated approach.

This paper has been drafted to fit into the Integrated Research, Development and Extension Training Approach (IRDET) model developed and demonstrated in Nepal under the FAO-NARC component of Hill Leasehold Forestry Project, now adopted country-wide. It will be explained separately by Dinesh Pariyar (Senior Scientist, Nepal Agricultural Research Council) at the next TAPAFON meeting scheduled to take place in Bhutan, 29 April to 4 May, 2002.

This trial programme network has been designed for use by non-experts in field experimental designs.  Its objectives are easily understood by participating farmers who are themselves experts in their own fields of endeavour.  However, the level of sophistication for any part of this trial programme can be easily increased for sites should an expert field technologist be located nearby to provide necessary technical back-up and training.

Organisation

The Himalayan Fodder Oat trialing network describes a design and implementation system to identify modern fodder oat cultivars that will provide consistent yield and quality performances over a wide range of altitudes, climates, day lengths, fertility conditions, water and crop management systems; and to record disease profiles.  Cultivars from a range of donors’ world wide, including selected landraces or local checks, will be screened.  Sites to be chosen will be representative of the local areas of production.  A participatory system of researcher, farmer and community involvement is intended to build in an effective seed transfer mechanism.

An overall co-ordinator, country leaders and an international technical support group should be established to provide support and essential linkages to plant improvement activities and germplasm access worldwide.

The authors suggest the screening of a genetically diverse set of at least 13 modern fodder cultivars from several countries or continents, together with one or more checks in year one.

This environment has unusual issues for planning small plot field experiments for testing cultivars.  Most potential trial sites are small, where the laying out of conventional small plot randomised “complete block trials” will not be possible.  Nor will it be possible, on several sites, to lay out a complete replicate. On these sites we suggest plot sizes be reduced to ensure a complete replicate can be sown on a single site. Researchers will therefore need to be innovative and be prepared to stretch the boundaries of traditional experimental trial design practices to successfully implement an effective evaluation system for introduced and local cultivars.  Clusters of similar local sites will need to be chosen to lay out a complete field trial. It is not immediately possible, and arguably necessary, in the early stages of this project, due to the large magnitude of differences being searched for, to prescribe a traditional cultivar evaluation design and data analytical programme to identify the cultivars that will provide the consistency of performance required.

A simple definition of “broad adaptation” is to identify high yielding cultivars that perform consistently from year to year over a wide range of environments.  Based on previous experiences the local agronomists, site supervisors and farmers, with basic supporting data, together with their knowledge of traditional cultivars or landraces, will successfully choose the top performing cultivars for use within their communities. 

It is important for the organisers to understand why a particular cultivar has been selected, and to track its performance over time. Organisers will also need to investigate how any data collected may be used to assist germplasm donors identify germplasm suitable for testing in the Himalayan Region from within their breeding programmes.

As much useful data as is practicable should be taken from as many sites as possible. Realistically, the organisers should focus on key cultivar performance indicators such as plant vigour and survival, forage yield and numbers of cuts, quality, maturity, plant diseases, straw height and standing ability if used for hay, or taken through to maturity for grain production.  These performance indicators can be measured subjectively using the quantitative scoring scale and template for note taking described below.  This scale can also be used for making weight assessments if mechanical scales are not available at all sites. 

During the first two seasons of screening, network activities should focus primarily on identifying and recording major and outstanding differences.  At the same time, however, it is hoped that sufficient quality data from enough sites will be obtained. This will enable statisticians to provide training on cost effective methods for conducting and analysing such trial data using clustered sites, and across site comparisons, in future projects.

Sophisticated analytical systems are needed to back up decisions taken in the field, for example to quantify Genotype by Environment interactions and study them across agro-climatic/altitude sequences. Given the magnitude of differences expected during the initial stages, however, it is unlikely that subsequent statistical analysis will change the initial choice of cultivars the community representatives will make.  This will probably not be the case, later in the programme, once the initial screening of a wide range of cultivars has been completed. As a result, the input of the statisticians needs to be built into the network from the beginning, and experienced statisticians and agro-ecologists identified and included on the technical support panel.

Seed multiplied and tested under the network will be fed into existing informal systems of seed distribution currently used within the targeted communities. In this way, Plant Variety Protection, royalty collection and seed production systems will not be a village-level issue during the initial pilot stages.  To make this work, local organisers will need to ensure seed from selected cultivars continues to be made available to farmers directly from seed production plots of the tested cultivars. This fulfils a project objective to ensure seed of new cultivars is released unimpeded by compliance and business contract issues. At the same time, a proper varietal maintenance programme will have to be established and managed under the network, by respective co-ordinators, at a national level.

Detail

This paper outlines a trial package developed in consultation with local agricultural experts to ensure relevancy to local situations in target areas.  It is expected, therefore, that this prescription will not fit every site. Local controllers will need to exercise their judgement as to how this prescription is applied.

The package includes a manual/field book with details for selecting trial sites, suggested field trial layouts, sowing and sampling details, sample data recording sheets, seed packets, recording notes and a suggested scale for recording quantitative data.

The first season’s trial seed will be sourced from donors and delivered to the trial site co-ordinators at the next TAPAFON meeting scheduled in Bhutan, for seed multiplication by the respective site supervisors in Bhutan, Tibet, Pakistan, India and Nepal. 

Following the seed multiplication exercise site controllers will arrange for seed to be packaged into (3 m²) plot size seed lots.  Site controllers will deliver seed to the local communities and make arrangements for site selection, sowing, sampling, and data recording.  Frequency of sampling will depend on local growing conditions, and will be the decision of the site controller and community representatives.  It is assumed whole plots will be harvested, weighed and air dried to obtain dry matter yields, and weights recorded or assessed following each harvest.  However, local organisers may decide to sample plots. Either is satisfactory. The number of harvests for each complete trial site cluster will vary according to local farming practices.

Separate seed increase plots of fodder cultivars should also be grown.  These seed production plots should be monitored for grain production capability.  Small samples of seed from selected fodder cultivars will enter the traditional and informal seed distribution network.

Farmer Group Crop Cut Surveys managed by local agronomists trained in the IRDET approach will provide a system of tracking and measuring the post trial performance of cultivars released to the community.

Trial site selections

The following criteria should be applied for site selection:

• sites to be representative of the target growing areas and managed by co-operative owners
• sites to be uniform for soil type and slope
• level sites to be free of small ridges and hollows
• sites to be free from animal foot paths, and historical tree lines.  Site controllers should consider historical uses that may have left fertility or compaction patches within the trial plot area 
• sites to be secured from casual grazing by animals
• sites preferably to be located inside existing forage crops managed by competent, interested and cooperative farmer/owners
• maintenance fertiliser should be applied as per soil analysis
• will be irrigated or rain fed.

Trial content

TAPAFON participating countries will include Pakistan, India, Nepal and China/Tibet.  Invited international participation may include New Zealand and others may follow.

• each participating country may provide 3 cultivars of the 18 total cultivar number.
• provisions in each trial for 18 entries plus 5 local control cultivars. If local oat fodder cultivars are not available, other dual-purpose small grain cereals used also as fodder cultivars such as wheat could be used for local fodder controls.  
• the current proposal is for representatives to bring 3 cultivars of 2 kg per cultivar to the Bhutan meeting

Trial design and layout

• Time of sowing will be consistent with local farmer practice.
• A field book template will be provided.
• Trial plots will be rectangular, 3 m²  (3 metres by 1 metre).
• Seed will be line sown.
• Sow each plot as 5 single rows, with each row of each plot sown 20 cm apart.
• To ensure consistency in calculating plot area across all participating countries and trial sites, calculate plot area using the formula: 
  Plot length x number of rows x row spacing.
• Where site size is a limiting factor for sowing a complete replicate, the plot size should be reduced by sowing fewer rows per plot.
• If site is variable, longer narrower plots should be sown. 
• Seeding rates will range around 200 seeds for each square metre.
• Therefore total numbers of seeds required for each 3 m² plot will equal    200 x  (plot area of 3 m² ) = 600 seeds per plot. 
• An example of a cultivar with a one thousand grain weight (TGW) of 40 g, the same seeding rate of 200 seeds per square metre equals 8 gms of seed for each square metre. This equates to a commercial seeding rate of 80 kg per hectare, satisfactory for a dry site.  Site controllers will need to make their own judgements on seeding rates for each region. 
• Wetter fertile sites may require a higher seeding rate than the dry sites.
• To ensure accurate seed sowing rates, the authors suggest thousand seed weights for each cultivar entry is calculated prior to packing seed for sowing.
• Alternatively donors provide thousand seed weights and a seed analysis certificate
• Percentage of viable seed could also affect seeding rate.  Increasing the seeding rate can compensate for a low germination percent.
• A simple test for checking seed germination can be done by counting out 100 grains. Lay out the seed in a single row on a wet paper or cloth hand towel, fold towel in a double layer, roll and store in warm place ensuring the towel remains moist but not too moist.  Depending on temperature, germination counts can be done in 5 to 8 days.  Sowing corrections can be calculated from these results.
• A quick method of measuring seed weights is to count out 100 seeds, weigh the counted 100 seeds, and multiply the resulting weight by 10 to obtain the weight of 1000 seeds.  Sowing rates can be calculated accordingly.
• The trial package should include local seed lines for buffer plots sown around the perimeter of the trials.
• There will be one seed packet for one plot.
• Trial entries should be randomised. A simple method to randomise entries is to number discs of similar size and weight, such as metal or plastic bottle tops, number each disc to equal the number of cultivar entries, shake in a container, pick out discs randomly and enter each number immediately and in sequence into the “ENTRY” column of the field book template provided.

Plot sampling

Timing of harvests will be the decision of local controllers and consistent with local harvest or grazing time frames.  Generally our expectation would be at least 3 harvests, more for very fertile warmer regions.  Cutting height will be 10 cm (i.e. the height of a closed fist or extended hand as a simple unit of measure).  If plots are to be sampled, a folding quadrant should be provided.  The size of the sample harvested must be recorded.

Field observations required

Agronomic issues for fodder production: yield, lodging, cold tolerance, disease resistance, drought and/or excessive moisture tolerance. Agronomic performance for seed production: crop management issues, maturity, straw quality, lodging resistance, seed shattering, grain harvesting, threshing, grain quality and storage issues should be monitored.  A field book template is provided. The organisers assume local farmers are experienced in growing and handling forage, fodder and seed crops and will contribute to records of cultivar performances. Additional observations may be recorded by the national co-ordinators.

As a guide, all or some of the following recordings should be made:

• Emergence percentage or seedling survival: count the number of plants per square metre at the two to three leaf stages.
• At the boot stage record:  plant height, number of tillers per plant, number of tillers per square metre, leaf to stem ratio, green forage per square metre, dry forage per square metre, number of days from sowing to 50% flowering, number of days to grain harvest maturity, disease observations.
• Nutritive value: subject to availability of laboratory services, measure crude protein and ME values.

General instructions

• The sample blanks form is for recording general information about the location, site and the experiment. Please provide as much information as possible and clearly indicate the units in which data are taken  
• Data should be reported using the metric system. Provision has been made for recording data using local measures. Supervisors should ensure local measures are converted to metric.
• In the Field Book template be sure to provide details of what is harvested. Record the sample size harvested, whole plot or area sampled from each plot.  
• For recording plant characters please indicate the scale used. There are two main options. The box “ticked” for a favourable expression of a character, or box “crossed” for an unfavourable expression of character.
• For the scale for appraising foliar intensity of cereal disease (Fig. 1) make sure the date and name of the disease is recorded.
• If the appraisal scale (Fig. 1) is used for assessing any other agronomic factor, e.g. lodging, maturity, cold tolerance, fodder quality, and your cultivar choice, make sure the date and name of character is recorded.  In using this scale 1 is used for a favourable result and 9 is unfavourable.  Use 0 for no information.

(These will need to be re formatted and added to if necessary during the meeting as a discussion/working group)

Himalayan fodder oat trial network

General notes to be taken

Soil type: (Supervisor to complete)

Chemical fertiliser:

Sowing date:       Early?          Normal         Late

Soil conditions at sowing:                     Moist?            Dry        Normal        

Seed emergence:                               <80%           80-90%          >90%          

Method of sowing:                       Line sowing              Broadcast          Drill             Other

Local major cultivars:

Proportion of crop seed saved for seed:           

Chemical control of insects

Chemical control of weeds

Mechanical control of weeds

Hand weeding

Disease development:   None            Trace            Moderate        Heavy

Insect damage:            None             Trace            Moderate       Heavy 

Weed problem:           None              Trace            Moderate       Heavy  

Bird or animal damage: None           Trace             Moderate       Heavy  

Other observations:

Weather:                      Normal         Abnormal   

Rainfall mm:     

Field Book

Supervisor/recorder:     ___________________________

Location details:         ___________________________

Altitude                                    Longitude                    

Latitude                                    Aspect            

Harvesting details:

Product harvested:        fodder          grain                         other

Plot size: _______m²     Local measure: _______________      Area of plot sampled: ________

Foliar disease: a scale for recording quantitative data

Recording plant diseases

The following scale (Fig. 1) represents a relatively simple and standardised method for recording severity details of foliar diseases (especially rusts
(Fig. 2 & 3), mildew, blotch and barley yellow dwarf virus (Fig. 4) across environments, by different controllers.  This scale is more appropriate for crops taken beyond panicle emergence and this will be the case for crops harvested for hay or fodder cultivars taken through to seed for a farmer's own use.  The following scale was used in part for managing and reporting on CIMMYT nurseries and adapted by plant breeders and agronomists for computerising observational data.

Figure 1:  Scale for appraising foliar intensity of cereal diseases.

To apply this scale, especially for rusts, mildews and blotches, grasp the plant half way up.  Use the scale 5 for lesions located at this point but not above.  Lesions above this point are given values from 6 to 9.  Lesions below this point are given values 1 to 4.

Therefore the score values will go like this:

1 =Resistant: Possibly a few isolated lesions on bottom leaves only.

2 =Resistant: Scattered lesions on bottom leave only.

3 =Resistant: Light infection of lower third of plant, most leaves infected at moderate

to severe infection.

4 =Moderately Resistant: Moderate infection of lower leaves; scattered to light infection extending to leaves immediately below mid point of the plant.

5 =Moderately susceptible: Severe infection of lower leaves; moderate to light infection extending to the mid point of the plant; infection does not extend above mid point of the plant.

6 =Moderately Susceptible: Severe infection of lower third of plant; moderate degree of infection on middle leaves; scattered lesions above the mid point of the plant.

7 =Susceptible: Severe infection on lower and middle leaves; infections extending to the leaf below the flag leaf, or trace infections on the flag leaf.

8 =Susceptible: severe infection on lower and middle leaves; moderate to severe infection of upper third of plant; scattered lesions on flag leaf.

9 =Very Susceptible: Severe on all leaves and the spike may be infected to some degree.

0 =No scoring possible due to necrosis or as a result of other factors such as missing plots.

To simplify this scale plant breeders and agronomists may choose to use a shortened version of the scale, numbers 1, 3, 5, 7, 9.  The full or shortened version can be applied to other characters such as height, adaptability, maturity, and general appearance.  Farmers could be trained to use this scale for scoring their yield assessments, for example.  This shortened scale may be particularly useful for recording barley yellow dwarf virus, a disease that frequently is not uniformly spread through out trial plots or farmers crops.

Figure 2: Crown rust.	Figure 3:  Stem rust.
Figure 4:  Barley yellow dwarf virus	Figure 5: Loose smut.

Acknowledgments

We acknowledge the assistance provided by Dr Stephen Reynolds (FAO), Mr Dinesh Pariyar (Nepal), Dr Bimal Misri (India), Dr S.M. Rafique (Pakistan), Professor Kent Eskridge, University of Nebraska, the New Zealand Seed Technology Institute and Crop & Food Research (New Zealand).

References

Stevens E.J. Wright, S.C.; Pariyar, D.; Shrestha, K.K.; Munakarmi, P.B.; Mishra,C.K.; Muhammad, D.; Han, J. 2000: The importance of oats in resource-poor environments. Proceeding of the 6^th International Oat Conference, Christchurch New Zealand, November 2000. 74 p.

Instructions for the management and reporting of results for wheat program international yield and screening nurseries.  International maize and wheat improvement centre. Londres 40, Apdo. Postal 6-641, Mexico 6, D.F., Mexico.