Oilseed rape Recommended List trials are undergoing a revamp. But will planned and recent changes help farmers get better information for their money? Robert Harris reports
Untreated oilseed rape trials are to be dropped from the Recommended List (RL) testing programme from this autumn to help make better use of limited funds.
Other moves designed to target effective spending include limiting the time a variety can spend in RL trials to five years and a more rigorous selection process introduced last autumn to ensure only varieties with real potential are trialled.
“The aim is to target trial resources as effectively as possible,” says RL trial manager Jim McVittie. “It is all part of a wider HGCA review that will be published this autumn.”
The whole testing system costs about £400,000 a year, mainly co-funded by HGCA levy payers and breeders. With the number of new varieties rising year on year, the budget is under pressure.
Dropping untreated trials makes sense, says Dr McVittie. “There were too few to deliver useful data, especially given the marked difference in disease threat between regions.
“We shall instead fund more treated trials, including two more National List 2 trials and three more RL trials in England. We shall also collect more light leaf spot data in observation trials in Scotland, rather than trying to measure yield differences.”
The selection process was rejigged last autumn to help address rapidly rising yield trends. The key to this was raising the hurdle for varieties aiming to become RL candidates.
“We are seeing a yield increase of 3-4% a year,” says Dr McVittie. “Promoting varieties to be an RL candidate on the basis of how they compared with the mean of the top-three recommended varieties in their class didn’t make sufficient allowance for this. We were probably testing material that didn’t have much chance of being recommended.
“We are now comparing performance with the mean of the top-five recommended and candidate varieties instead. Including newer varieties will raise the gross output bar and only the best varieties will get through.”
Agronomic merit will still be assessed, he adds. “Some varieties may not quite achieve the necessary gross output, but other characteristics might be valuable.”
Once on the RL, most varieties will be limited to a five-year testing programme, says Dr McVittie. “We’ve done our job by then – people know how older varieties perform on farm.
“That doesn’t mean they won’t stay recommended, but resources will be better used testing new material. Breeders can appeal where they think a variety deserves to be re-entered into trials.”
Matthew Clarke, Dekalb‘s UK oilseed rape breeder, questions the use of existing candidate varieties as a benchmark for successors.
“These varieties have not proved their consistency. They have only been in HGCA trials one year, so you are setting a high target on limited information.”
Mr Clarke, who also represents plant breeders on the Oilseeds Crop Committee of CEL (the HGCA subsidiary that manages the RL), is more sympathetic towards the five-year testing programme.
“In general, it’s sound, though we need to ensure varieties that the market really wants can remain in trials longer, particularly if there is no obvious replacement.
“Varieties such as Apex, Castille and Excalibur could have been at risk. So we hope CEL will listen to breeders’ representations – once varieties are removed from trials, their yield data can plummet as control varieties change.”
On balance, the ending of untreated trials is a gain, especially for the north, he believes. “There was not enough data, so channelling funds to collect more information through observation work to use in the selection process makes sense.”
DSV sales director Mike Mann agrees. “They were of little relevance – no-one grows oilseed rape without fungicides.”
He welcomes the tougher selection process for would-be candidate varieties. “RL trials are expensive to run, so it’s sensible to weed out poorer performers at the start.”
However, he also welcomes the further revision in the selection process to provide a safety net for varieties that, while not having the highest gross output, still have plenty to offer. Candidate varieties aiming for first-year recommendation will now be compared against the mean of the three best RL varieties of their type, rather than the best only.
“The old method could exclude good varieties because of a sudden yield hike. Last season Compass was compared with PR46W21, but because it’s gross output was 2% less, it wasn’t recommended – even though it would have been ahead of all other hybrids on the list.
“This year, it will be assessed again, being compared with PR46W21, Dimension and Flash, so hopefully should make the RL.”
Richard Elsdon, technical manager at United Oilseeds, would like more assurances that dependable varieties will be readmitted to RL trials after five years.
“More than anything farmers want reliability and repeatability. It would be wrong to remove a variety just because it had been around for a long time – it might still have plenty of genetic merit.”
He believes untreated trials should have been retained, and has little faith in disease observation work. “The current scoring system for disease doesn’t really work.
“PR46W21 has a treated gross output of 111. It’s stem canker rating is 4, yet it maintains an untreated yield of 93. Excel, on the other hand, has a treated gross output of 100 and a stem canker rating of 9. But its untreated yield drops to 84. On that basis, PR46W21 should have been caned out of sight.”
Philip Marr, technical development manager for Masstock SMART Farming and an Oilseeds Crop Committee member, reckons the main problem with the RL is that it is too big. “Five varieties for each region would fit the bill.
“Removing a variety from testing after five years might help. Most farmers are not using the RL for older varieties – they know how to grow them. But why not go further. Some varieties, such as Hornet and Canti-CS for the north, hardly feature, so why keep them on?”
He’d also revamp the disease ratings. Although from next autumn, all north varieties will need a minimum 6 rating for light leaf spot to be considered for recommendation, few farmers in Scotland grow varieties scoring less than 7.
“It should be introduced for the east and west region too – the disease is spreading south quickly.”
He’d like to see stem canker ratings tightened too. “This disease is spreading north – we need to be raising the bar.”
What else should change?
Trials protocols are being kept under review, says Dr McVittie. “We are always looking for improvements, to make things more realistic.”
One area of keen interest is establishment – currently all plots are ploughed. “This no longer reflects most grower practice, so we are looking at Claydon and subsoiler plot drills this autumn.”
Mr Clarke welcomes that. “Given the increase in min-till establishment, that’s the way we should be going. We also need to review how we establish populations – many farmers are now using wider rows.”
Seed rates are a continual source of discussion, with critics pointing out that conventional and hybrid rates of 100 and 70 seeds/sq m respectively are well above farm practice.
“Establishment was very good last year, and it’s fine in hindsight saying we should cut. But perhaps we could leave final seed rate to the discretion of the plot manager on the day,” says Dr McVittie.
Mr Mann says excessive seed rates in RL trials can prevent hybrids from expressing their vigour. “We recommend 50 seeds/sq m in practice, and 30 is not uncommon.
“Farmers know there is a yield advantage, and a recent Openfield survey quizzing 160 farmers concluded the average advantage was 180-190kg/ha, and that hybrids will deliver a lot more from poorer land.”
Dr McVittie defends the heavy use of fungicides, designed to keep infection levels below 5%. “Our rationale is to measure genetic potential. Agronomists decide how much to spend in practice.”
Mr Elsdon is less convinced. “RL trials should reflect farm-type application rates. An optional 3 litres/ha of Bravo followed by Proline at 0.7litres/ha in late autumn and again at stem extension is unrealistic.
“Rather than seeking maximum genetic potential, applications should reflect what the top 5% of farmers would apply. The huge anomaly between trials and on-farm yields suggests levy payers are not getting the information they need.”
Vigour is one key attribute missing from the RL system, says Mr Marr. “The important thing is speed of leaf development, which is governed by temperature.
“Excel only needs 85 day degrees to produce a new leaf during the autumn – which typically means it will have one true leaf six days after emergence, and will continue to produce leaves quickly. Flash, on the other hand, takes 140 day degrees, so will take almost twice as long to get to the same stage.”
“Unlike CEL, I’m not surprised Excel has sold in large quantities in the north, even though it is not recommended there. Others might look better choices on paper – until you try drilling them in September.”