Page Banner

United States Department of Agriculture

Agricultural Research Service


item Gealy, David

Submitted to: Government Publication/Report
Publication Type: Proceedings
Publication Acceptance Date: September 26, 2005
Publication Date: September 1, 2006
Citation: Gealy, D.R. 2006. Dynamics of pollen dispersal and confinement in u.s. rice. Proceedings of Workshop on Confinement of Genetically Engineered Crops During Field Testing, Riverdate, MD, September 13-14, 2004. pp. 96-98.

Technical Abstract: Rice (Oryza sativa L.) is arguably, the single most important food crop in the world. In the U.S. it is grown primarily in lowland irrigated areas of Arkansas, Louisiana, Mississippi, Missouri, Texas, and California. Although the U.S. produces only about 3.5 million acres of rice it is typically the world’s third or fourth largest exporter of this crop. Rice has the smallest genome, genetically is the simplest of all major cereal crops (diploid with 12 pairs of chromosomes), and has served as a useful model for crop genomics research because its genes and gene functions have a high degree of applicability to the other cereal crops. Great improvements due to rice breeding, fertility research, mechanization, pest management research, and herbicide-based weed control have been achieved in the U.S. in the last half century. However, weeds remain a major limitation to optimum rice production and economic returns. Barnyardgrass and its close relatives are the most prevalent and economically significant weed species in U.S. rice. Growers apply herbicides to essentially 100% of all rice acres largely because of these weeds. In the southern U.S., weedy red rice (O. sativa L.) is considered to be the most troublesome weed, when present, because it acts like a crop mimic of rice and has traditionally been uncontrollable in rice. California cropping systems are essentially free from red rice due to a highly effective clean seed program and exclusive use of water seeding. Rice and weedy red rice are considered to be the same species and can readily intercross with one another. Several distinctive U.S. red rice plant types are common, especially awnless strawhull, awned blackhull, and awned strawhull. Most of these are tall-statured with medium-grain seed size and shape. Flowering dates for red rice types range from being slightly earlier than most commercial cultivars (often strawhull awnless types) to several weeks later than any modern cultivar (often blackhull awned types). Numerous other red rice types can also be found in low numbers, including short-statured long-grain types, and short-statured awned types, suggesting that these may have been derived originally from natural crosses between red rice and long-grain commercial rice (most prevalent rice in the southern U.S.) or between awned red rice and semi-dwarf commercial rice (introduced into the southern U.S. in the 1970s). Both transgenic (glufosinate-resistant and glyphosate-resistant) and non-transgenic (imidazolinone-resistant ‘Clearfield’ rice) herbicide-resistant rice cultivars have been developed in recent years. With the subsequent marketing of non-transgenic ‘Clearfield’ rice, outcrossing between herbicide-resistant and non-resistant rice and between rice and weedy red rice has been increasingly scrutinized. ‘Clearfield’ rice was first grown commercially in the U.S. in 2002 It has been rapidly adopted by growers due to effective red rice control, and was planted on about 15% and 25% of the rice area of the southern U.S. in 2004, and 2005, respectively. ‘Clearfield’ rice is not grown in California. To date no pest-resistant transgenic rice cultivars have been grown commercially in the U.S. Rice and red rice are primarily self-pollinating because most stigmas (female) are fertilized by pollen (male) produced in the same flower, and pollen shedding usually occurs slightly before or concurrent with flower opening. This is in stark contrast to corn which produces male and female organs on different flowers, easily facilitating cross pollination. Maximum outcrossing between adjacent rice and red rice plants appears to average about 0.2 to 0.7% under field conditions, based on a large number of published, controlled experiments. However, outcrossing is highly variable. In small-scale field tests, the apparent outcrossing rates are sometimes zero, especially if the flowering periods of the two plant types do not overlap sufficiently, or if the total number of seeds sampled was too small to adequately detect very low outcrossing rates. Conversely, in a recent Louisiana report, poor herbicide performance in a large commercial field of Clearfield rice that was heavily infested with red rice, resulted in outcrossing rates of about 3%, even though outcrossing rates averaged over numerous locations were similar to the lower rates indicated above. Interestingly, a landmark report from the 1930s indicated that the average outcrossing rates between rice plants were generally similar to those reported more recently between rice and red rice, and were highly variable as well. Outcrossing between rice and red rice can occur in both directions. In Louisiana and Arkansas studies, outcrossing has usually been much greater when pollen from red rice (tall plants) fertilized stigmas on rice (shorter plants). Thus, pollen produced by the tall red rice plants was more likely to fertilize flowers on the shorter rice plants than the reverse situation. This outcome is consistent with the notion that pollen is more likely to ‘rain’ down (e.g. by gravity) than to be lifted upwardly (e.g. by wind currents) in order to achieve cross fertilization. This directional difference in outcrossing rate may actually reduce the introgression rates of red rice hybrid derivatives into rice fields because nearly all of the relatively large number of red rice hybrid seeds produced on rice plants as a result of fertilization by red rice pollen are removed from the field during harvest (modern combines can remove as much as 95% of the grain from a rice field) and will not impact the field in the future unless the grower unwisely chooses to plant this seed. Conversely, the relatively few hybrid seeds produced on red rice plants as a result of fertilization by rice pollen are likely to shatter from the plants before harvest and remain in the field to cause future problems. Shattering rates of up to 80% are common for red rice seeds. Outcrossing over great separation distances is mitigated by a combination of factors: 1) each rice or red rice flower opens only once, which lasts for about one hour at midday; 2) pollen released into the environment remains viable for only about 10 minutes; and 3) under calm wind conditions, most pollen grains tend to fall near to where they were produced. A complicating factor may actually facilitate both short distance and long distance outcrossing: the stigma can remain viable much longer (i.e. several days) and tolerates greater temperature extremes than pollen. Thus, flowers that do not produce viable pollen of their own (e.g. after exposure to temperatures below 16C during pollen formation; or as in the case of male-sterile plants used in the commercial production of hybrid rice) will be more susceptible to fertilization from foreign pollen, even if it has arrived from at great distance away. Genetic background also can affect outcrossing. Thus, cultivated rices (e.g. O. sativa), which rely entirely on seed production to ensure that uniform plants can be produced for future crops, generally outcross less than wild, perennial species (e.g. O. rufipogon). Additionally, japonica-based cultivars, predominantly grown in the U.S., tend to outcross less than their indica-based counterparts prevalent in more tropical areas of the world. In recent work with imidazolinone-resistant rice cultivars, researchers in Arkansas have shown that ‘CL161’ rice outcrossed substantially more with awnless strawhull red rice than did ‘CL121’. The mechanisms responsible for this difference are not fully understood, but the panicle height differential between rice and red rice (very tall plant) is less for CL161 (taller cultivar) than for CL121 (shorter cultivar), which would tend to place flowers of CL161 in relatively closer proximity than CL121 to red rice flowers. Floral characteristics such as large anthers or lengthened filaments that tend to release great quantities of pollen into

Last Modified: 11/24/2015