| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Donald Danforth Plant Science Center, St. Louis, MO 63132
b Monsanto Company, St. Louis, MO 63167
c Monsanto Company, Coalinga, CA 93210
d Qualls Ag Labs, Ephrata, WA 98823
e Chesterfield, MO 63017
* Corresponding author (Philip.j.eppard{at}monsanto.com)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
500000 kernels). Increasing temporal separation reduced the distance required to achieve genetic isolation. Outcrossing was <0.01% at 500 m when source and receptors flowered at the same time, whereas this level of confinement was achieved at 62 m or less when 2 wk of temporal separation was used. No outcrossing was detected at 750 m and 2 wk of temporal separation. This is the first practical evaluation of time and distance acting together to achieve genetic purity in maize.
Abbreviations: gdu, growing degree units • PMGF, pollen-mediated gene flow • RM, relative maturity
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Inadvertent release of plant genetic traits may occur through escape of viable seeds into commerce, or through transfer of the traits in pollen to other maize plants. The latter is called pollen–mediated gene flow, and may be evaluated by observing outcrossing from a source field to receptor plots. Pollen–mediated gene flow is of concern due to the characteristics of the maize plant and its pollen.
Maize is an open pollinating, anemophilous crop that produces abundant pollen. The tassel of a single modern hybrid maize plant may produce three million pollen grains, a number greatly in excess of that needed for fertilization (Uribelarrea et al., 2002). However, the pollen of maize is among the largest and heaviest of the grasses, with a diameter of about 90 µm. For comparison, the pollen grains of ragweed (Ambrosia spp.) and Timothy (Phleum pratense L.) are 20 and 34 µm in diameter, respectively (Raynor et al., 1972). Most maize pollen is dispersed by gravity downward from the tassel, falling in the vicinity of the originating plant. Bateman (1947) found pollen deposition at 27 m was <1% of that close to the source plants. Raynor et al. (1972) estimated that 98% of maize pollen remains close to the originating plant, and that <1% would be found beyond 60 m. Jarosz et al. (2003) estimated that 95% of the pollen produced was deposited within 10 m of the plot. While maize pollen may fall quickly from the atmosphere, the great abundance of pollen produced and its wind-borne nature create the risk of low levels of gene flow to neighboring maize crops.
There have been a number of studies of natural PMGF in maize, where outcrossing is measured between a pollen source and phenotypically or genetically dissimilar receptors. Bateman (1947), working in the United Kingdom, used a flint-kernel pollen source about 0.001 ha in size, and evaluated flinty kernels arising from cross pollination of the source hybrid into sweet maize. Gene flow was 70% next to the pollen source, and <1% at the 23-m distance. Jones and Brooks (1950) used a yellow kernel pollen source, outcrossing onto white kernel receptors in Oklahoma in 3 yr. The source plot was about 3 ha in size, and 3-yr average outcrossing was 28.62, 1.19, and 0.20% at 0, 200, and 500 m from the source plot, respectively. Using the same system, Jones and Brooks (1952) detected levels of gene flow from a 5-ha source plot ranging from 25% next to the source to 3.1% at 125 m. More recently, Jemison and Vayda (2001) in Maine used the Roundup Ready (Monsanto Company, St. Louis, MO) trait as a genetic marker to evaluate gene flow downwind from a source of about 0.3 ha in size. In 2 yr, average outcrossing was observed to be 1.34, 0.48, and 0.39% at 30, 35, and 40 m downwind, respectively. Luna et al. (2001) studied gene flow in Mexico, using yellow or purple kernel source plots and white kernel receptors. Source plots were 0.4 ha in size. They found only one positive (outcrossed) kernel each in plots at the 100-, 150-, and 200-m distances, and none at greater distances (300 and 400 m). Klein et al. (2003) studied gene flow in two maize fields in France, using 0.04-ha source plots homozygous for blue kernel color inside yellow maize fields. The farthest distance studied was 50 m from the edge of the source plot, and a low but undefined level of outcrossing was observed at that distance. Ma et al. (2004) studied outcrossing from 0.07-ha plots of yellow-kernel maize into surrounding white maize in multiple sites in Canada during 3 yr. They found that outcrossing downwind was as much as 82% in the row immediately adjacent to the pollen source, and declined to <1% within 37 rows (28 m).
Distance alone is only one possible barrier to PMGF. The successful transfer of genetic traits in pollen also requires that viable pollen fall on receptive silk. Most silks are exposed within 5 d after silking begins (Carcova et al., 2000), and are quickly fertilized by the enormous amount of pollen released by surrounding plants. Silks that are not fertilized begin to senesce within about 8 d of emergence from the husk (Bassetti and Westgate, 1993a; Anderson et al., 2004). Basal cells of the aging, unfertilized silk begin to lose turgidity and collapse, which restricts pollen tube growth and prevents fertilization (Bassetti and Westgate, 1993b). Unfertilized, receptive silk is available for only a brief time, as is viable pollen. The majority of pollen produced by a maize crop is shed within a 5- to 8-d period; low levels of early and late pollen release may extend the total pollen shed period to 14 d (Ogden et al., 1969; Jarosz et al., 2003). Successful fertilization requires that male and female flowers be active at the same time. Such flowering synchrony—commonly called nicking—is critical to seed set within a particular crop (Westgate et al., 2003), and also to PMGF between two maize crops (Ma et al., 2004). In the latter case, immigrant pollen must contact a silk that has not yet been fertilized by indigenous pollen grains. Because of the need for receptive silk for genetic transfer, gene dispersal via PMGF cannot be directly equated with the physical dispersal of pollen (Feil and Schmid, 2002).
The deliberate disruption of flowering synchrony between plots or fields, with the goal of enforcing genetic isolation, is usually done by displacing planting dates and is called ‘temporal separation’ or ‘temporal isolation’. Temporal separation is a well-established mechanism for genetic isolation in maize (Sprague and Dudley, 1988, p. 12), but has not been extensively quantified in the same fashion as distance separation. Ma et al. (2004) noted the importance of site factors such as drought on synchronous flowering and thus on gene flow, but did not explicitly study deliberate temporal separation.
Time and distance are among the most common, effective, and easily utilized measures for genetic separation of maize plants (Lamkey, 2002; Aylor et al., 2003), and meet the requirements for redundant systems of biological confinement in which "two or more safety measures are applied ... each with fundamentally different strengths and vulnerabilities, so that the failure of one will be balanced by the integrity of another" (Anonymous, 2004). Acting in conjunction, time and distance are both simple systems that may be expected to provide robust confinement of genetic traits in maize.
In practice, time and distance are deployed in such a fashion that an increase in temporal separation is often used to reduce isolation distance requirements. The USDA guidelines for isolation of maize producing pharmaceutical products call for 1600 m (1 mile) distance separation for fields planted at the same time, but only 800 m for fields planted 4 wk or more apart (USDA-APHIS, 2003). However, little empirical work has been done on gene flow restricted by distance and time together. The interaction of the two factors, in either a general or a statistical sense, and their effect on PMGF has not been evaluated. Such information is needed to validate existing guidelines, and could also be used to develop more precise schemes for deploying time and distance together to achieve desired levels of genetic purity.
The studies reported here quantify PMGF generated by open-pollinating source plots of about 1 ha in size. This situation is intended to model natural gene flow from a seed production field, which is a likely source of unwanted PMGF in the production of regulated maize. A practical understanding of the relationship of time and distance in this model system is a step toward establishing realistic standards based on both factors acting together. Plot size, wind strength, and turbulence (Aylor and Parlange, 1975; Ma et al., 2004), atmospheric humidity (Luna et al., 2001; Aylor, 2004), or environmental conditions specific to the study locations in California and Washington may be expected to influence the level of gene flow. Thus, standards inferred from these data should be only cautiously applied in other conditions.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Trials in California were placed on Nord loam (coarse-loamy, mixed, superactive, thermic Cumulic Haploxerolls). The soil type in Washington was a Quincy sandy loam (mixed, mesic Xeric Torripsamments). A comparison of important attributes of the two locations is in Table 1.
|
Experimental Procedures
Experimental Designs
Source plots were 0.4 ha to 1.2 ha in size, and receptor plots were four rows wide by 6 m long (18–23 m2). Receptor plots were located at distances from the source plot, and were planted at weekly intervals before and after the planting time of the source plot. Preliminary experiments were performed in the same locations in 2000. These experiments studied distance isolation alone using somewhat different techniques than used in 2001 and 2002, and are reported in detail in the supplemental data (Crop Science website, http://crop.scijournals.org/).
Diagrams of experimental designs, experimental details, and wind results for each site are found in Fig. 1 through 6 . All source plots were bordered by a strip of male-sterile maize about 12 m wide, modeling likely production practices for open-pollinated maize with regulated genetic traits. Beyond the male-sterile border, the areas between the plots were left fallow, allowing pollen to move unhindered from the source to receptors.
|
|
|
|
|
|
Receptor plots were composed of a blend of commercially–available yellow kernel hybrids of different relative maturities (RMs). The receptor hybrids where chosen attempting to bracket the flowering of the pollen source hybrid, and thus to represent a population of commercial fields of somewhat different RMs. Receptor hybrids chosen were 92- to 99-d RM in 2001, and 94- to 107-d RM in 2002. Approximately equal numbers of kernels of each receptor hybrid were used in the blended seed planted in the receptor plots. Small plots (two or four rows by 3 m) of receptors planted as individual hybrids (not as a blend) were placed inside the pollen source to observe flowering of the individual hybrids. Outcrossing results from these hybrids gave an empirical measure of flowering synchrony with the source, and provided a baseline of outcrossing at very short distance. All seed was supplied by Monsanto Company or by Holden's Foundation Seeds. Seed used is shown in Table 2.
|
All distances were measured from the outside of the pollen source plot, which was construed to include the male-sterile borders.
The effect of temporal separation on gene flow was studied by displacing the planting dates of sets of receptor plots from the planting date of the source plot. Temporal separations were imposed on a weekly basis, and are expressed as, for example, –1 wk or +1 wk (1 wk before or 1 wk after the source plot planting, respectively). Temporal separations used for each experiment are shown in the experimental designs (Fig. 1, 3, and 5). Growing degree units (gdu) were calculated for each temporal separation interval, using base 10°C (Sprague and Dudley, 1988, p. 616–617).
Wind Determination
Wind velocity and direction were recorded by on-site weather stations during pollen shed for all trials. Data logger models used were CR510 (California) and CR10 (Washington), both from Campbell Scientific (Ogden, UT). Model 034A wind sensors (MET ONE Instruments, Inc., Grants Pass, OR) were used in both locations. Wind data are illustrated with diagrams (Fig. 2, 4, and 6) constructed using S-Plus6 software (Insightful Corporation, Seattle, WA). Each diagram shows wind direction by time in 15-min intervals. Wind force is grouped in three categories: light, defined as <5 km h–1; moderate, defined as 5 to 10 km h–1; and strong, defined as >10 km h–1. Wind gusts frequently exceed average wind strength in a corn canopy (Shaw et al., 1979), so grouping average wind speed into categories is a concise yet meaningful way to express the likelihood, direction, and timing of gusts in a qualitative fashion. Most pollen release occurs during the morning hours (Ogden et al., 1969; Miller, 1985; Jarosz et al., 2003). The morning and early afternoon are when most viable pollen grains are dispersed, so the time period from 0600 to 1800 (local time) is shown (Fig. 2, 4, and 6). Wind is normally shown for the days around peak pollen shed. In Washington 2001, those days covering the emergence of receptive silks in the 0-wk planting are also shown.
Sampling
All primary ears were evaluated from the entire plot area in each receptor plot (4 row by 6 m), which typically had 150 to 200 primary ears. Purple xenia kernels were counted and recorded on an individual kernel basis. Percentage outcrossing was calculated using the total number of kernels evaluated in each plot, which was estimated by counting the number of ears evaluated, multiplied by the average number of kernels per ear for each year, location, and planting time. Where high levels of outcrossing were observed, such as in the plots inside the pollen source, outcrossing was estimated visually as a percentage of each ear surface occupied by kernels showing the purple xenia effect.
Statistical Analysis
Analysis of variance was applied to each experiment to evaluate the statistical significance of the main effects Distance and Time, and the Distance x Time interaction. Only data for Distance and Time combinations with nonzero outcrossing were included in the ANOVA. For comparison across sites and years, the 95% upper confidence limit of percentage outcrossing was calculated at selected distances. There is approximately 95% confidence that the true level of outcrossing would be less than the 95% upper confidence limit for any specific location and distance (Johnson et al., 1993). SAS version 9.1 (SAS Institute, Inc., Cary, NC) was used for the data analysis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In California, plant stand was reduced in the earliest planting (–2 wk) in 2001 by damage from darkling beetles (Blapstinus spp.) and beet armyworm (Spodoptera exigua Hubner). This damage decreased the number of kernels evaluated, increasing the effective limit of detection to 0.0008% outcrossing. In California in 2002, damage due to bird feeding was noted, but was largely confined to early plantings of the 2-row receptor plots inside the pollen block. Plots at farther distances, where precise detection of low levels of gene flow were critical, were not affected. No other adverse pest or environmental effects occurred.
Flowering synchrony for the individual receptor hybrids within the pollen source in the California trials are shown in Tables 3 and 4. The best flowering synchrony in 2001 was obtained from the +1-wk planting, which was 160 gdu after the source plot. Peak silking coincided with 50% pollen shed, around 2 September, and outcrossing to receptors within the source plot was 49 to 70%. In the 0-wk planting, silking was slightly earlier than pollen shed, and outcrossing was 37 to 54%. There was little overlap in flowering between the earliest planting (–2 wk or 348 gdu) and the source, but gene flow was higher than might have been expected, with up to 15.6% noted. Apparently, low levels of early silking or pollen shed were not captured by the flowering observations, but were sufficient to cause measurable outcrossing. The latest planting, +2 wk (304 gdu), gave only marginal nick and gene flow, since pollen flow was terminated on 7 September, just at the beginning of flowering of that planting. In 2002 (Table 4), silking of the 0-wk receptors took place at the beginning of pollen shed, and outcrossing was 32 to 56%. The +2-wk planting was 335 gdu later than the source, flowered after peak pollen shed, and had 3.8 to 18% outcrossing. Other plantings showed increasingly poor nick and lower levels of outcrossing, although some gene flow was detectable at even the longest interval and the poorest nick.
|
|
In California, the highest outcrossing outside the pollen source itself was 0.7 and 0.6% at 24 and 32 m in 2001 and 2002, respectively (Tables 3 and 4). Outcrossing declined to 0.002% or less at 750 m, when source and receptor plots flowered at the same time. Gene flow was reduced by both distance and temporal separation, so that 2 wk and 750 m reduced outcrossing below the limit of detection (0.0002%) in both trials (Fig. 1).
Results of the Washington trial in 2001 are in Fig. 3 and Table 5. The design of this experiment was constrained by the comparatively small field size available, and the pollen source was located on the western side of the field, to allow as much distance as possible to the farthest receptors in the predominant downwind direction. Temporal separation intervals were only enforced after the planting of the source plot (+1 and +2 wk), and this separation represented only moderate crop development, with about 80 gdu accumulated each week. There was some degree of flowering synchrony for all plantings, with the +1-wk planting showing the best nick, as in California in that year. The highest levels of outcrossing observed within the pollen source in the 0 wk and +1-wk plantings were similar, 3.6 to 8.1% in week 0 and 3.8 to 6.8% in +1 wk. The +2-wk planting flowered somewhat after peak pollen shed, and outcrossing was less at 1.7 to 2.9%. Outcrossing within the pollen source in Washington was thus much lower than that seen in California, where up to 70% outcrossing was noted.
|
Winds for those days coinciding with silking of the 0-wk receptors (Fig. 4, 2–7 August) were strong and predominately toward the east and north. Wind direction during this period is consistent with the greater outcrossing seen in the northeasterly plots than in the southeasterly ones, with totals of 15 and 2 xenia kernels, respectively (Table 5). Prevailing wind direction during peak pollen shed was variable, with wind toward the south or southwest (8, 9, and 12 August), toward the north and east (9, 10, 11, 12, and 13 August), or from any direction (13 August).
In the Washington 2002 trial, the 0-wk planting showed very good nick with the source (Table 6 and Fig. 5). Peak silking and 50% pollen shed both occurred around 14 August. However, outcrossing to receptors within the pollen source at this planting was only 12 to 16%, and thus much lower than that seen in California at similar nick. Three weeks gave 263 to 353 gdu separation from the source for receptors planted before or after it, respectively. These intervals almost eliminated flowering synchrony, and reduced outcrossing accordingly. This was especially true of the +3-wk planting, which had only begun silking at the time that pollen shed in the source plot was terminated.
|
Wind in Washington, 2002 (Fig. 6) showed less-regular patterns of direction than in California (Fig. 2), and was also stronger, with more intervals > 10 km h–1. Winds toward the north and northeast on any of 3 d (13, 15, and 17 Aug. 2002) could have accounted for the direction of gene flow observed.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
330 gdu of crop development. The Distance main effect was highly significant for all experiments (P < 0.0001, Table 7). The Time main effect was significant for three out of the four experiments at the 0.05 significance level, the exception being Washington 2001. The Time x Distance interaction was also highly significant for both experiments in California (P < 0.0001), but was not significant at the 0.05 level for either experiment in Washington. In those experiments where the Time x Distance interaction is significant, the relationship of time and distance appears to be complex, and further research is needed to elicit the fundamental relationship between these factors when they are employed together to limit maize pollination. Until a predictive model emerges, isolation standards invoking both time and distance will be most realistic when derived from empirical data collected on both factors acting together in specific environments.
|
80 gdu wk–1 is required for temporal isolation to be effective, and that Time is best measured by the accumulation of heat units. A total of 350 gdu did not totally eliminate gene flow at close distances, but did so at farther distances. There are abundant data relating the development and flowering of maize hybrids to heat units, and these could be used as a starting point to construct a more precise calibration of temporal separation.
Levels of Gene Flow
Outcrossing outside the pollen source itself was <1% in our studies, even where flowering synchrony was good and there was up to 70% outcrossing in receptor plots within the pollen source. Two hundred meters was sufficient to reduce outcrossing to <0.1% (95% upper confidence limit) at both locations in all 3 yr (Table 8).
|
A sterile hybrid border was placed around the source plot in these studies, and this border may have affected levels of gene flow. However, Jones and Brooks (1952) observed ambiguous results when evaluating gene flow across a 10-m-tall border of trees. They reported that the border reduced outcrossing at close distances, but appeared to have no effect at longer distances. Burris (2001), in a survey of outcrossing into maize seed production fields in the United States, found little effect of borders ranging from 6 to 24 rows on outcrossing at either the margins or the centers of the production fields. These results are consistent with the concept that long-distance transport depends on turbulent currents lofting particles to some height, presumably above the height of any border, before they become part of the escape fraction (Gregory, 1973). Pollen grains thus entrained in turbulence at some altitude would fall from the atmosphere in a random fashion, and would not be affected by bordering of either the source plot or the receptor field. The male-sterile border was a part of the practical model we evaluated, and our results are insufficient to establish whether such bordering is indeed effective in reducing PMGF in maize.
The environments that we studied are typically more arid than those of the major maize-growing regions of the United States, and thus the levels of gene flow we observed may be less than those expected in more humid regions such as the U.S. Corn Belt. Pollen is affected by atmospheric water potential, and 1 to 2 h under drying conditions can result in substantial loss of viability (Luna et al., 2001; Aylor 2004). However, the impact of humidity on gene flow, as opposed to pollen viability, has not been established. Bateman (1947), Klein et al. (2003), and Ma et al. (2004) worked in England, France, and Canada, where temperature and humidity may have been more favorable to pollen survival. Their results are similar to ours, although the relative contribution of source plot size and environment cannot be established between their studies and ours. If pollen longevity is restricted by atmospheric humidity to only 1 h, that time is still sufficient for a pollen grain to be blown a considerable distance. The influence of humidity on PMGF may perhaps be understood on a population basis, where the proportion of viable pollen grains in a population available for dispersal is reduced more quickly under arid than under more humid conditions. In this case, the occurrence of episodic wind events—their strength, direction, and timing with regard to crop flowering and pollen release—could be at least as important as humidity in determining PMGF. Studies directly evaluating humidity on gene flow under similar wind conditions remain to be done, but such experimentation is obviously difficult. The influence of the environment on gene flow is a complex problem played out at the interface of physics and biology, and any inference of isolation standards drawn from these data should be only cautiously applied to other environments and systems. Modeling approaches, such as suggested by Aylor et al. (2003), may be useful in generalizing specific results to more diverse environments.
Wind Direction
Wind direction was not strongly correlated with outcrossing in Washington, and wind diagrams alone would have had little value in predicting the direction of gene flow, an effect also noted by Ma et al. (2004). Strong but brief wind events at certain critical times may have had the most impact on directionality of gene flow. Gene flow effective wind events may thus be essentially random and episodic—dependent on the alignment of pollen shed, wind speed, and exposure of receptive silks. Eddies and counter-flows are also prevalent in the atmosphere (Gregory, 1973; Aylor, 1990), and wind is typically measured in a single stratum. Thus, it is not surprising that we cannot account for its complexity or its effect on gene flow. To use wind data in a predictive fashion will depend on the integration of wind speed and direction, time of day, and flowering status of source and receptors.
The diagrams (Fig. 2, 4, and 6) show that wind was usually more persistent in California than in Washington—that it blew for longer times from fewer directions. Wind persistence may contribute to the higher levels of gene flow seen in California compared with Washington, since a larger number of viable pollen grains would have been driven in the predominant downwind direction, and not dispersed in several directions. A similar effect may help to explain the lower levels of outcrossing to receptors within the source plot seen in Washington compared with California. Fluctuating wind in the latter location would allow pollen grains to fall more directly onto the originating plant and its near neighbors, such that more receptive silks are fertilized by indigenous pollen grains.
Wind speed, persistence, and gustiness may also help account for the higher levels of gene flow observed by Jones and Brooks (1950)( 1952) in Oklahoma. High wind speed increases turbulence within the crop canopy and liberates more pollen from plant surfaces (Aylor and Parlange, 1975). Turbulence also favors long-distance transport by increasing the proportion of grains lofted into the escape fraction (Gregory, 1973; Aylor, 1990). However, Jones and Brooks do not present wind data, and no definite conclusions can be drawn.
Constraints on Gene Flow in Maize
Pollen-mediated gene flow in maize is constrained in several ways. The size and weight of maize pollen grains ensures that most pollen remains close to its source plant, and that which becomes airborne quickly settles from the atmosphere. The velocity of deposition of maize pollen is 0.2 to 0.3 m s–1 (Raynor et al., 1972; Di-Giovanni et al., 1995; Aylor, 2002). Assuming a constant wind speed of 10 km h–1 and constant settling rate of 1 m 3 s–1, pollen grains lifted to a height of 20 m would settle from the atmosphere under the effect of gravity within about 60 s, or about 166 m downwind. Of the few grains that become airborne at all, a small proportion may be picked up by turbulent currents and kept aloft, to be dispersed at longer distances.
In order for immigrant pollen to result in successful outcrossing in a field at some distance, several processes must align: The pollen grain must be released, picked up by turbulence, and carried at least several meters aloft, escaping immediate deposition by gravity. The grain must be deposited into a different maize field, settling past the native tassels, each producing millions of competing indigenous pollen grains. The immigrant pollen grain must alight on a genetically compatible strand of silk that has thus far escaped fertilization by abundant native pollen. Even when flowering at the same time, outcrossing between two maize crops at some distance from each other appears to be a rare event. The probability of outcrossing is further reduced as the flowering periods of the crops are separated by time.
The use of distance isolation is an intuitive and reasonably well-studied technique to ensure genetic purity of maize. Temporal separation, while also intuitive, is less well-defined. The study of both factors at once requires an additional level of complexity. This work is the first to study the concurrent effect of time and distance together, and a practical demonstration of the effectiveness of these barriers to gene flow when used for genetic isolation of maize.
|
ACKNOWLEDGMENTS |
|---|
Received for publication December 12, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Related articles in Crop Science:
This article has been cited by other articles:
|
|
 |
|
  M. Langhof, B. Hommel, A. Husken, J. Schiemann, P. Wehling, R. Wilhelm, and G. Ruhl Coexistence in Maize: Do Nonmaize Buffer Zones Reduce Gene Flow between Maize Fields? Crop Sci., January 16, 2008; 48(1): 305 - 316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. I. Gustafson, I. O. Brants, M. J. Horak, K. M. Remund, E. W. Rosenbaum, and J. K. Soteres Empirical Modeling of Genetically Modified Maize Grain Production Practices to Achieve European Union Labeling Thresholds Crop Sci., September 8, 2006; 46(5): 2133 - 2140. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Plant Registrations | Soil Science Society of America Journal | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
