During the summer months it’s quite common to find the labs in the Crop Genetics department at the John Innes Centre fairly empty. Although many of us might wish we were on sunny beaches enjoying the sun, the reality is that we’ll generally be found in the less tropical climate of East Anglia attending to our field trials. Trials are very important for plant geneticists and pathologists as they provide a chance to test crop varieties in a natural environment, and also allow data on yield performance, disease resistance and stress tolerance to be collected on a larger scale than in a glasshouse or polytunnel.
But what does running a field trial involve?
Whilst the activity in the plots usually begins to pick up during spring, trials can actually be a year round activity. Generally seeds must be bulked at least 4 months before they’re due to be planted to make sure there is enough to fill a plot. Winter cereal varieties such as winter wheat or barley are generally sown in the field during October to December, so that they’re exposed to the cold temperatures they require to germinate, a process which is called vernalization. Spring cereals can be sown from January to April however, as they don’t need such cold temperatures in order to grow.
Once the seeds have begun to emerge from the soil, irrigation can be set up to ensure that the trials have enough water and fertiliser can be applied to make sure the plants are as healthy as possible. Then depending on the type of trial, plants may be sprayed with herbicides to prevent unwanted weeds, or fungicides which prevent fungal diseases. The time it takes to collect the required data also depends on what type of trait is being investigated. For example, height measurements may be taken after a few weeks of growth, whereas yield data might be gathered at harvest maturity which can be 3 or 4 months after initial planting.
After all the required traits have been recorded (and the weather is co-operating) it’s then time to harvest, either by hand or mechanically using a harvester. This means that yield data can be recorded and the seed can be used for further experiments or analysed for nutrient, fungus or toxin content. It’s also time to start doing some statistics with the data that has been gathered. Whilst two varieties might look different in terms of a specific trait, it’s important to know whether this difference is statistically significant- is the difference real or is it purely down to chance? The results can sometimes be unexpected, but give an indication of which lines would be useful to investigate further. Once this is decided it’s usually time to start planning for next year’s trial!
As you generally only get one chance a year to gather trial data it’s important to plan properly. However, sometimes even the best planned trial can be changed due to problems with drilling, poor seed or even animal damage. For example having control lines within a field plan, such as including dwarf varieties, means it is easy to check the location of specific lines within a field. I’ve found this particularly useful when plots aren’t where I expected them to be due to drilling issues! Using replicates or split plots within a trial can also be helpful, especially if seeds don’t establish well in a certain part of the field, meaning that you don’t completely lose data for that line.
Multi sites and environment
In lab experiments you would usually want replicates to be as similar as possible in conditions to the original. This is not true of field trials. You instead aim to have trials in a wide variety of environments to prove that the effects you see are universal. However running field trials is costly so there is a limit between the ideal number of locations you’d test and what you can actually perform. This year I have 9 different trials, with 5 in Norfolk to allow me to monitor them easily. The others are in Suffolk, Dorset and Ireland. This geographical spread, as well as a showing reproducibility, can also act as a failsafe if the weather conditions cause problems in one area. For example the disease that I work on requires rainy weather to thrive. Thus if Norfolk has a dry summer I may have poor data, however additional trials in different weather conditions can compensate for this.
People usually react with shock when I reveal that I can go weeks without using a PCR machine! But the equipment favoured by the field scientist is often very different to those that you would associate with a scientist. My most important piece of equipment is my clipboard; it is so much easier for data collection than having to write on the ground or leaning on the back of a passerby. Other important parts of key field gear include wellingtons, sunhats and suncream depending on the weather. Beyond this, the equipment you need depends on what you are working on, from a simple metre sticks for height measurements to combine harvesters for measuring yield. As for me, my favourite piece of field equipment is my field cushion, as it has prevented my knees being completely destroyed when taking measurements at ground level.
What do we do in the field?
Chris: My field trials are about how the fungus Zymoseptoria tritici spreads in the crop canopy and how this relates to differences in yield. So my main task in the field is scoring the disease. This is done by inspecting leaves at different heights in the canopy and scoring the lesions formed by the fungus. This process is complicated by the presence of other disease and abiotic damage so you need to be trained in what you are looking for. In addition to this a wide variety of physiological traits are being measured in the plots to identify differences in the physiology of the lines used. At the end of the experiment, both diseased and undiseased plots of these lines will be harvested allowing for comparisons of yield between the lines. Across my different sites the combination of the disease, physiology and yield data should allow us to understand the septoria-yield trade off better.
Rachel: My trials look at the resistance of a heritage barley variety to Fusarium Head Blight (FHB) which is a major disease of cereals caused by toxin-producing Fusarium fungus. Resistance to FHB has been linked to plant height, with taller varieties being more resistant. However, tall varieties are less favoured in agriculture, so I’m aiming to see if FHB resistance and a shorter height trait can be combined into one variety. This involves spraying the crops with Fusarium to generate disease, and then scoring for disease resistance, height and other important traits. My trial will be hand harvested and then the grain analysed to determine the toxin content of each line to see if the disease resistance score correlates with a reduction in toxin production.
By Chris Judge and Rachel Goddard, both 3rd year PhD students in the department of Crop Genetics