In order to effectively and efficiently determine the relative value of different varieties of a specific crop species or crop type it is important to evaluate the varieties for the traits of interest in a replicated variety trial. While there is no one "right way" to design and execute a trial, there are a number of parameters that should be considered when setting up and planting a trial. My intent in this brief essay is to give people an appreciation for the way in which the demonstration sites for the participatory farmer selection will be planted, evaluated, and used for selection. I will also address the very important issues of evaluation with a numerical score and how this affects our selection of a variety for advancement and contrast it to our selection within a variety for crop improvement.

Plot Requirements
The prerequisites of any trial are selecting a number of varieties to compare and knowing which traits are the most important for your environment, market, and production system. In order to get the most accurate information from a trial performed in any one location, the person putting in the trial must consider the following three requisites in their design.

1) Replications. All of the varieties being evaluated comprise a subset of the entire trial that is called a replication, or rep for short. There should at least two to four reps in any one trial, with three reps being most common. This is done to ameliorate any "field effects" due to changes in soil type, fertility level, moisture content, and relative solar exposure to name just a few. While you can plant the first rep in any order you want, the subsequent reps must be randomized.

2) Block Design. The overall shape of your trial in the field should be in a block design that is as close to a square (length ~ width) as possible. This minimizes the potential field effects mentioned in #1.

3) Border Rows. Have you ever noticed how much better (and sometimes worse) the outer row of your crop does on the edge of a field when compared to the other rows? This is known as the edge effect. Often the plants on the edge of the field get more sun, nutrients, and/or water as they have less competition for these inputs on the edge. Hence we always plant at least one row (or two) of a standard cultivar of the same crop on all four of the borders surrounding the trial.

Evaluation of the Trial
Trial evaluations often fall into two categories. The informal method of trial evaluation is often done primarily with comments that assess relative performance of entrants. The second type is the much-too-formal type where extensive measurements are made with measurement tools on too many traits. The former is fine for an initial assessment in an unreplicated observation plot and the latter is almost always overkill and a waste of time.
Evaluators should always know their crop well enough to be able to determine which four or five traits are most important and know this traits well enough to score them effectively on a 1 to 9 scale, using all numerals including the extremes! The score of "1"always represents the poorest entrant for a particular trait and the score "9" represents the best entrant for that trait. All other entrants will receive scores that are gradations between the two and for ease the numerals 1, 3, 5, 7, and 9 will suffice. These scores only refer to the performance of each entrant in that field, on that day and in essence will only represent their relative performance for that season and location. When a trial site is stricken with a malady such as disease pressure, insect attack, heat, drought, or cold stress, it often gives the evaluator an opportunity to see which varieties exhibit the highest levels of resistance to that particular environmental pressure.

Disease Field Test / Disease Nursery / Horizontal Resistance
A Disease Field Test is a type of trial where the evaluator has "invited in" a particular disease to determine the relative levels of resistance to that disease among the varieties they are testing in the trial. Disease symptoms are scored based on relative severity of symptoms and sometimes at the rate at which a disease is able to spread through a particular variety. (This technique can also be used for insect damage and subsequent selection for resistance to that insect.)

The primary function of a Disease Nursery is to screen putative disease resistant varieties (and select new versions) for Horizontal Resistance (HR) to a particular disease in the field. HR, while not conferring complete resistance to disease has proven to be the preferred form of disease resistance for organic agriculture, as it is conditioned by a number of genes, giving it a form of disease resistance that is much more genetically elastic than Vertical Resistance (VR). VR, on the other hand, confers a complete resistance (akin to immunity) for the crop variety based on a single resistance gene that matches only one race or strain of the pathogen. (This is why there are disease resistance codes listed in catalogs that read "Resistant to DM 1, 2, 3", which represents three VR genes for Downy mildew). The problem with VR is that eventually a new strain of the pathogen will arise (nature abhors a vacuum) and all current VR varieties will become quite diseased under even moderate disease pressure. In contrast, HR with its genetic elasticity has equivalent resistance to all races of a pathogen and doesn't break down when exposed to a new race of the disease.
Because HR is not complete resistance and is conditioned by a number of genes, there is a wide range in the levels of resistance expressed in individuals of any one variety that has HR. In fact, individuals of a particular variety with varying degrees of resistance can be scored upon evaluation with the 1 to 9 system. This, therefore, allows the evaluator the opportunity to select individuals in a Disease Nursery as parental stock for a new version of the variety with greater HR to a particular disease. In this way it is possible to produce derived versions of a given variety with higher levels of disease resistance by pyramiding the multiple genes conferring resistance from a number of plants.

Restoring Our Seed Demonstration Sites
These trials will be placed both by location and by time of planting to maximize the likelihood of disease outbreak. We will plant a Demonstration Plot Disease Nursery, using tomatoes as an example of selection in a self-pollinated crop and using a yet to be determined cross-pollinated species (we will elicit opinions from Seed Class attendees). These plots will be placed with farmer cooperators in each of three New England states over the next three years. We will hold a summer workshop each year to demonstrate how to identify and select for HR to the native diseases found in each of the given locations.

In each plot we will include 1) a variety of the crop with merit that we wish to improve for HR (the test variety), 2) a variety of the crop that already confers a high level of resistance to the disease we are selecting against (the resistant check variety), and 3) a variety of the crop that exhibits symptoms of extreme susceptibility to the disease (the susceptible check variety). These three varieties of the experimental crop will be planted using the Plot Requirements and Evaluation Techniques stated above. Border rows will be planted with the susceptible check variety to help spread the disease through the plot. Cooperating farmers and project Extension Staff will score plots on the 1 to 9 scale described above at different points in the development of the disease over the course of the season. Selections will be made of test variety plants having higher levels of disease resistance. In self-pollinated tomatoes final HR selections of the test variety can be made later in the season as the rate of outcrossing with susceptible plants will be low. In the cross-pollinated crop that we work with, the selection of individuals from the test variety with HR will have to be relatively early in the season to prevent resistant individuals from intermating with susceptible plants. Therefore when selections for HR are made all other individuals from the test variety, as well as the susceptible check and resistant check varieties will have to be rogued and destroyed. This will insure that only the best individuals with HR from the test variety will intermate. Seed gathered from these HR plants will be used in the next cycle of selection at another Demonstration Plot Disease Nursery in the following year as the new test variety. In subsequent years we will also include the original unselected test variety for comparison to the selected version. This will allow us to assess the progress we've made in our HR selected test variety.

Selection at Demonstration Plots will also consider uniform time of flowering, prolificacy of flowering, timely seed maturity, and seed yield. See future writings on Plant Life Cycles and Reproductive Biology Basics for more information on these traits.

Action Milestones and Outcomes

Concept: Farmers and summer workshop attendees at Demonstration Sites will learn,
1) How to identify diseases symptoms of the pathogen of interest in the field.
2) At what stages in the plant's life cycle is a particular disease best assessed?
3) How to plant a replicated trial & score (1-9 scale) varieties for important traits.
4) What level of HR is meaningful and worthy of selection

Actions: Farmers and summer workshop attendees at Demonstration Sites will act on,
1) Planting trials for evaluation purposes and monitoring on-farm diseases.
2) They will identify, screen, and select for HR to diseases at appropriate levels.
3) They will select for good reproductive characters and fecundity.
4) They will harvest seed of the crops they are working with efficiently and properly.

Outcomes: Farmers and summer workshop attendees will use these skills,
1) They will use on-farm trials for their most important decisions on seed crops.
2) They will use on-farm trials to determine which varieties to select for HR.
3) They will develop their own versions of crop varieties with increased HR.
4) They will improve reproductive traits in their seed crops and increase seed yields.

Copyright 2002 by John P. Navazio