CORN PERFORMANCE TRIALS
by David Morris, Secretary Ontario Corn Committee

The year 2004 will mark the 68th year that hybrid corn performance trials have been conducted in this province. The first preliminary tests were planted in 1937 at Guelph, Ridgetown and Woodslee and included 40 hybrids. Hybrid corn was new to Ontario, and farmers wanted to know if this new development was really an improvement over the open-pollinated strains they had been using. The rest, as they say, is history.
As the corn industry in Ontario evolved and spread across the province, so did the activities of the Ontario Corn Committee (OCC). In 2003, a total of 410 different hybrids were tested and trials were conducted at 21 locations. Testing methods and locations have changed to reflect current realities and the climatic diversity of this province as much as possible. The OCC continues to refine its procedures to improve the reliability of the trials and the accuracy of the data. However, the system is not perfect, nor does it match everyone's needs exactly, so questions still arise about what the OCC does or doesn't do.

Q: Why doesn't the Corn Committee manage its corn the same way that farmers do?

A: It comes down to the difference between growing corn commercially and doing scientific research. Your goal is to make a profit - ours is to generate reliable, unbiased data that you can use to compare hybrids. Although our results are intended for farm use, we cannot get reliable data using the same approach as a corn producer uses to grow a field of corn. For all hybrids to be treated fairly, they must all have the same growing conditions. Although our trial co-ordinators choose the most uniform sites they can find, no field is totally uniform, and hybrids in different parts of a field inevitably are subjected to different growing conditions. Usually, these differences are insignificant and are overcome through the use of randomized, replicated trials. To accommodate the number of hybrids that we test (up to 120 per trial) we must use small plots and small plot equipment. Otherwise, the amount of land required would be mind-boggling. As it is, it takes 1.5 to 2 acres for each trial, and finding a sufficiently uniform piece of land of even that size is a challenge in some areas.

Q: Why don't you report the results from all of the trials that you harvest? Farmers have to live with low yields too.

A: It's not a question of whether the overall yield level is high or low. The OCC has no qualms about publishing the results of a low-yielding trial, provided that it is uniformly low. However, it will not publish the results of a trial if it believes that the data are misleading. The Committee believes that "Bad data is worse than no data." Under extremely stressful conditions, especially drought, otherwise small differences in soil conditions become magnified and hybrids begin to perform differently in different parts of the test. When this variability becomes too great, replication is not sufficient to compensate for it, and the data become unreliable.

Q: Why are the trials thinned to a uniform population? Farmers have to live with what comes up, why don't you?

A: To compare the genetic yield potential of hybrids fairly, they have to be grown at the same population. If the population of a plot is too high or too low, we have to discard it. And if there are too many "holes" in a test, we have to discard the whole test. To minimize this risk, we over-plant and thin to 30,000 plants per acre (ppa). If emergence problems are widespread, trial co-ordinators have the option of thinning a trial back as low as 26,000 ppa to save the test. We can't rely on planting the same number of seeds per plot because of the risk of variable pest damage and because of differences in the quality of seed that we receive. For example, seed of experimental hybrids is often produced in winter nurseries in the southern hemisphere. The quality of that seed is often marginal and in cool wet springs, its emergence suffers more than that of commercial seed.

Q: Why aren't there trials in my area?

A: It's a matter of limited resources, both financial and human. The trials have to be located reasonably close to someone with the expertise, equipment and time to manage them properly. All but one of our trials are managed by personnel within the University of Guelph system, at Guelph, Ridgetown, Huron Research Station and Kemptville. The Ottawa trial is managed by researchers from Agriculture and Agri-Food Canada. This dictates the areas within which trials can be located. The OCC would like to have trials in other areas (e.g. Central Ontario) but has been unable to find anyone with the expertise who is willing to conduct trials there. The cost for the existing co-operators to do so would be prohibitive.

Q: Why do you not publish the results for all hybrids that you test?

A: The OCC does not want to burden you with data for hybrids that you can't buy or that aren't adapted where they were tested. Most companies enter some experimental hybrids each year in anticipation that they may be marketed the following year. Sometimes, one of these new hybrids is entered in a zone where it is not adapted. Sometimes, the company decides to not market the hybrid, ever. In either case, publication of the results would be in no one's best interest. Often, a company decides it is not ready to proceed with marketing, in which case, it has the option of deferring publication of the results for one year. In the case of older hybrids, companies have the option of dropping them from the publication if they are being removed from the marketplace. This may be done for a variety of reasons, such as limited seed supply or poor performance. When a company decides to not publish any of the results for an established hybrid, that hybrid may not be entered again.

Q: Why doesn't the OCC test herbicide-tolerant hybrids (ie. Round-up Ready or Liberty Link) in separate trials treated with the corresponding herbicide?

A: Because conducting separate trials would increase the cost of the testing system, the OCC wants to be certain that they are justified. Researchers within the University of Guelph have completed the first year of a study looking at the effect of herbicide treatment on corn yield. Preliminary results of this study suggest that with proper herbicide selection, application and timing, one can use conventional herbicides on herbicide-tolerant corn with no significant adverse effect on yield. The OCC is awaiting additional data before making a decision.