To receive weekly email notification when the latest issue of the Bulletin is online, click on this link and fill out the form.

Pollination Problems in Corn

September 7, 2001
I know it's a bit late to talk about this because most of the corn in the state was finished pollinating more than a month ago, but it is a problem whose scope is just now being assessed. Without question, poor kernel set in corn is a problem in Illinois. Although it has been seen in most areas and on more than one hybrid, certain genetics and certain areas have experienced more of the problem. I'll address the problem using questions and answers.

Did weather during pollination cause this problem?

Little in the record of weather and crop development this year suggests that we should have anticipated this problem. It was warm in July, but the high temperatures were only in the mid-90s (at Urbana there was 1 day at 96 and 2 at 94, with an additional 9 days with the high temperature between 90 and 93), and the average high temperature, average low temperature, and growing degree-day accumulations were all about average for the month. Rainfall was spotty, but most areas received some in early July, and many areas got some later in the month as well. So it will hardly go in the record book as a poor month for pollination, especially where soil moisture was adequate.

What did cause it if not the weather?

Agronomists are still piecing this together, but we think that there was a rather unusual combination of genetic hybrid traits, timing and duration of pollination, and insect feeding that resulted in the severe lack of kernel set in some fields. Most of the worst fields are in areas where very large numbers of Western corn rootworm adults were actively feeding on silks and perhaps pollen. The earliest-pollinating hybrids were finished pollinating in many areas by the time of maximum pressure from WCR and Japanese beetle, and most of those fields have good kernel numbers. Where pollen shedding and silking were late for any reason--genetics, later planting, delayed or uneven emergence--then the insects gravitated to those fields and probably ate silks off before they could receive pollen. There is also some anecdotal evidence that pollen shed, especially if it took place just after mid-July, may have been speeded up, such that it had peaked before silk emergence began and may have ended before all silks emerged. Many observers noted that there seemed to be less pollen left on the plants after pollination than is normally the case. Because winds were generally light during much of July, this may indicate that there was less pollen production than normal. And in some cases there appear to be more than normal numbers of tassels that seem not to have "opened up" to release pollen.

Was there a genetic problem with pollen supply?

Rumors have been circulating rapidly about a "male sterility problem" with some hybrids. Mixing male-fertile and male-sterile versions of the same hybrid has been a common practice in the corn seed industry for years. While it would probably be easier to simply sell hybrids with all of the plants male-fertile (able to shed pollen), it has been known for many years that male-sterile plants will produce grain yields that are several percent greater than yields from male-fertile plants of the same hybrid. Reasons for this increase in yield potential are not well understood, but the fact that the plant does not have to spend energy to produce pollen and the slightly less shading from male-sterile tassels seem to boost the energy (sugar) supply to the ear and kernels at the time the plant is usually limited in its energy supply. The amount of mixing of male-sterile seed varies with hybrid, depending on how much pollen the fertile tassels shed and on other factors such as the general ability of the hybrid to pollinate well. But 50% male sterility is not uncommon. In the high-oil Topcross system, in fact, only 8 to 10% of the plants shed pollen for the other 90 to 92% of the plants, all of which are male-sterile. If by mistake all of the plants in a field were male-sterile, of course, then few kernels would develop. But because the hybrids most affected this year have pollinated successfully in other fields, it is clear that male sterility is not the problem.

Did herbicides cause some of this?

In general, the answer to this question is no. Bob Nielsen at Purdue pointed out in his article that he saw this problem in fields where Lightning had been applied, but the majority of fields with poor kernel set in Illinois were not Clearfield hybrids, and many did not have Lightning or related herbicides applied. In fields where early applications of herbicides may have stunted plants and delayed development even slightly, it is possible that the timing of pollination was delayed enough to contribute to the problem. We do not have evidence that herbicides directly affected tassel development or pollen-shedding ability.

What hybrids were most affected?

While this problem did occur in more than one genetic background, its incidence was considerably greater in one set of hybrids than in others. If you are unsure about the hybrids you planted, walk into your fields to see if you have a problem. In many cases, the companies who sold seed of the most-affected hybrids have already examined many of the fields to see if there is a problem. A different but related type of damage that we're seeing on a different group of hybrids is a rather unusual abortion of what look like large but empty kernels, often scattered on the ear. We have seen this before, and it appears that some genetic types may begin development of kernels even without pollination and fertilization. Such kernels develop a seed coat and may show a slight amount of starch deposition, and they usually are fluid filled soon after pollination. We think that the endosperm, which is diploid before pollination, might start to grow in these "kernels" but stops when no embryo is developing, leaving behind a "shell" that eventually empties of fluid and appears as an empty seedcoat.

What do we do now?

Scout! There is clearly no way to recoup lost kernels, but it is certainly better to know whether there is a kernel set problem before the combine pulls into the field. This problem is not uniformly distributed in most fields, and there may not be an obvious pattern to its severity across a field. Because WCR adults in some cases ate silks after pollination was completed, the lack of silks on the ear may not predict the severity of the problem. You simply need to peel back husks to see how many kernels there are. In some fields, there will be normal-looking rows of kernels, but they may be only 5 to 10 kernels long, with the rest of the cob blank. In other cases all of the kernels will be scattered; and I have seen some full-sized cobs with hardly any kernels at all. Keep in mind that scattered kernels get larger than normal, but they also tend to dry down more slowly, and they are often damaged during harvest due to their size and moisture. A final hint: as harvest approaches, leaves and leaf sheaths of plants that have few kernels developing almost always turn purple. This is due to the buildup of sugars that the plant generates but that have no place to go if there are few kernels.

In general, much of the crop in Illinois is in fairly good shape, and the crop is reaching physiological maturity (black layer) rapidly in central and southern Illinois. Harvest is getting under way in the southern half of the state, with average to above-average yields being reported. The slightly cooler weather of recent days will slow drydown some, but it appears that much of the crop will reach combine moisture without the lodging that was so widespread last year. This is not altogether positive--to some extent, we have better stalk quality this year because the ears are not as large and so did not draw on the stalk as much. Statewide, we expect a little better than average yields (the August crop report had Illinois at 146 bushels per acre), but as usual this will comprise yields that are higher than average in some places and lower than average in others.

Soybeans Near Maturity

Although soybean plant height is less than normal in many areas due to dryness during stem growth, in other areas where rainfall was plentiful the crop is of average height or taller than average. Height is not correlated well with yield in soybean, but where crop height was decreased more than 8 or 10 inches below expectation (which in my estimation is about 36 inches, give or take), it is likely that pod number is decreased as well. Dryness has probably reduced pod and seed number in some areas, but the crop overall looks fairly good.

I've answered some questions about yield estimation this year. While the "standard" method of counting pods and seeds and converting this to seeds per acre, then dividing by expected seed size (number of seeds per bushel) to give expected yield may work well in some cases, I tend to favor a less laborious and more subjective method. This consists of simply pushing a group of plants sideways and observing general podding density, seeing whether plants are in good condition to fill pods, and looking over the top of the crop before "guessing" final yield. If one assumes that an average expected yield for a field might be, say, 45 bushels per acre, one can also use that as a factor in the "guess." It may not be very scientific, but yield estimation in soybean is usually quite inexact, probably because plants are so variable and also because conditions during the last week or two of seed filling make so much difference in final yield.

Soybean harvest is just beginning in southern and central Illinois, and I have heard few yield estimates so far. A continuation of good weather will be a great boost to both corn and soybean and to wheat planting as that gets under way later this month.--Emerson Nafziger

Author: Emerson Nafziger

The Pest Management and Crop Development Bulletin
Executive Editor: Kevin Steffey, Extension Entomologist

Subscription information: Phone (217) 244-5166 or email
Comments or questions regarding this web site: