Table Of ContentsNext Page

Evaluation and improvement of allelopathic rice germplasm at Stuttgart, Arkansas, USA.

David Gealy1, Brian Ottis2, Ronald Talbert2, Karen Moldenhauer3, and Wengui Yan1.

1 USDA-ARS Dale Bumpers National Rice Research Center, Stuttgart, AR 72160 email dgealy@spa.ars.usda.gov and wyan@spa.ars.usda.gov
2
University of Arkansas, Division of Agriculture, Department of Crop, Soil, and Environmental Science, Fayetteville, AR 72701 email bottis@uark.edu and rtalbert@uark.edu
3
University of Arkansas, Division of Agriculture, Rice Research and Extension Center, Stuttgart, AR 72160 email kmolden@uark.edu.

Abstract

Effective, affordable weed control is a challenge to sustainable rice production in the U.S. In the 1980s, evaluation of the allelopathic potential of rice germplasm in drill-seeded systems was initiated in Stuttgart, AR. These efforts led to the identification of several foreign lines with allelopathic activity against aquatic weeds, and some of these lines (e.g. PI 312777) also have suppressed barnyardgrass (Echinochloa crus-galli; BYG) more effectively and economically under reduced herbicide inputs than have commercial cultivars. These suppressive rice lines appear to produce greater root mass near the soil surface compared to non suppressive cultivars. Because plant type and grain quality of these lines have often been inadequate for the U.S. rice industry, a breeding program was initiated to combine the desirable characteristics of Katy long grain rice with several high yielding, suppressive lines. F5 or later generations from selections of the pedigree PI 338046/Katy//PI 312777 have been evaluated for several years. Some selections have produced acceptable yield and quality, but often yielded or suppressed BYG less than did parental lines or other standards. Several ‘competitive’ indica lines from Asia (e.g. 4593 from China) and commercial hybrids from the U.S. (e.g. XL8) have yielded as much or more than elite U.S. cultivars, and controlled BYG similar to the most suppressive rice lines. Thus, these germplasm lines may be useful in weed suppressive systems for U.S. rice.

Media summary

Breeding efforts in Arkansas have produced higher quality suppressive rice, often at the expense of reduced weed control. Indicas and hybrids are potential alternatives.

Key words

Barnyardgrass, Echinochloa crus-galli, suppressive rice, indica rice, hybrid rice

Introduction

Effective and affordable weed control is a continuing challenge to sustainable rice production in the U.S. Evaluation of the allelopathic potential of rice germplasm in drill-seeded systems has been conducted in Stuttgart, AR since the early 1980s (Dilday et al. 2001). These efforts led to the identification of several foreign lines with allelopathic activity against aquatic weeds, and some of these lines (e.g. PI 312777 and PI 338046) also have suppressed BYG in Arkansas (Dilday et al. 2001; Gealy et al. 2003; Gealy et al. 2005) and Asia (Olofsdotter et al. 2002). Some of the suppressive lines have reduced BYG more effectively and economically under reduced herbicide inputs than have commercial cultivars (Gealy et al. 2003). BYG suppression from an F3 progeny derived from a PI 312777/Lemont cross was generally intermediate to that obtained from either parent in a replacement series study (Gealy et al. 2005). PI 312777 and new hybrid rice cultivars have even provided elevated suppression of red rice growth and seed production in Arkansas (Estorninos et al 2005a and b; Ottis et al. 2005). Plant type and grain quality of these suppressive rice lines have often been inadequate for the U.S. rice industry. For instance, PI 312777 (prone to lodging) and Teqing have low milling yields (45% and 49% head rice, respectively) compared to the popular cultivar, Cypress (61% head rice) (Gealy et al. 2003). Thus, breeding efforts were initiated to combine the desirable characteristics of Katy long grain rice with several high yielding, suppressive lines. An initial promising line, RU9701151, was selected from the cross PI 338046/Katy and had acceptable milling yields (64% head rice) (Moldenhauer et al. 1999) with U.S. cooking quality, and good plant type. Other selections have produced reasonably high yields and quality, but have yielded less than commercial standards or have suppressed BYG less than the suppressive standards. U.S. rice growers are not likely to consider or accept new cultivars unless they produce yields and grain quality similar to or greater than those of existing cultivars and also exhibit qualitatively greater weed suppression. In preliminary studies with several ‘competitive’ indica lines from Asia ( e.g. line 4593 from China) and commercial hybrids from the U.S. (e.g. XL8 or imidazolinone-resistant XL8), rice yields were as high or higher than the highest yielding U.S. cultivars, and BYG suppression was as great as that from the most suppressive rice lines. Thus, these germplasm lines could be useful components of high yielding, weed suppressive systems for the U.S. This paper presents results of recent efforts at Stuttgart, AR to incorporate weed suppressive traits into high yielding rice cultivars with commercial quality grain and plant types.

Methods

Demonstration plots.

In order to visualize weed suppressive activities of a number of suppressive and non suppressive rice entries, large-scale demonstration plots were established. Two known weed suppressive rice lines and seven other rice cultivars were drill-seeded 2 cm-deep on May 21, 2004 at a rate of 430 seeds/m2. The experimental design was a randomized complete block with three replications. Plots contained 9 rows, 10 m long with 18 cm spacing between rows. Supplemental BYG seed was broadcast over all rice plots after planting. Propanil at 1.1 kg/ha ( the normal use rate) was applied post-emergence to all plots when BYG reached the three- to four-leaf stage in order to establish a minimal level of weed control in all plots. Weed control was rated visually after BYG heading. Rough rice yields (adjusted to 12% moisture) were determined from several 1-m sections of the five middle rows of each plot and expressed as kg/ha. Results in similar demonstration plots have varied somewhat from year to year, but PI 312777 (and more recently, XL8 hybrid) has usually been among the most suppressive rice lines and Lemont and Kaybonnet have been among the least suppressive.

Evaluation of selections from crosses between rice and weed suppressive germplasm lines.

More than ten years ago, initial crosses (PI338046/Katy) were made in an attempt to improve commercial acceptability of weed suppressive rice lines. PI 338046 is moderately suppressive against ducksalad (Heteranthera limosa (Sw.) Willd.) and BYG, and also has relatively high milling yield, whereas Katy is a commercial line with commercial southern long grain quality (Dilday et al. 2001; Moldenhauer et al. 1999). These lines were subsequently crossed to produce PI338046/Katy//PI 312777 in an attempt to incorporate additional weed suppression traits from PI 312777 while avoiding its low milling quality. Selected progenies from these crosses were developed to the F5 or later generations before conducting replicated field tests on weed suppression. Our priority for selecting progenies has been: high yield > appropriate plant/grain characteristics (e.g. a lack of lodging and high milling yield) > weed suppression. Thus, only lines that produced yields similar to those of commercial cultivars have been retained for further weed suppression testing. Numerous selections were discontinued after their yields in initial tests were low. Twelve selections of PI 338046/Katy and PI 338046/Katy//PI 312777, the original parents, and numerous standard cultivars were drill-seeded on May 4, 2004 as described above into 3 m long plots at a rate of 430 seeds/m2 unless otherwise noted. Plot sizes in preliminary tests conducted in previous years were only 1.4 m long because of limited seed supplies. Experimental design was a split plot with four replications. Main plots were rice lines or cultivars. Subplots were BYG levels such that each weedy plot was paired with a weed-free plot of the same cultivar. The weed free plots were kept free of weeds by hand weeding and the use of selective herbicides. In addition to the rice plots, BYG plots without rice were included in each replication and used to determine relative BYG production levels. Although the primary species present in weedy plots was BYG, other grass species such as sprangletop (Leptochloa spp.) and broadleaf signalgrass (Urochloa platyphylla) were observed in some plots. Supplemental BYG seed was broadcast over all weedy rice plots after planting. Propanil at 1.1 kg/ha ( the normal use rate) was applied post-emergence to all weedy plots when BYG reached the three- to four-leaf stage. Urea at 110 kg N/ha was broadcast over all plots immediately before permanent flood. Approximately one month after permanent flood, rice culms in two 25- by 25-cm quadrats were harvested and counted to determine tiller densities. Weed control was rated visually after BYG heading. Rice heights were measured shortly before harvest. Rough rice yields (adjusted to 12% moisture) were determined from a 2-m section of the five middle rows of each plot and expressed as kg/ha and as a percentage of the weed free value as an indication of the ability to maintain optimum grain yields in the presence of weed interference.

Estimating barnyardgrass and rice root distribution using 13C analysis.

Four suppressive lines and three commercial rice cultivars were drill-seeded on May 21, 2003 at a rate of 430 seeds/m2 as described above into 3 m long plots. There were three replications. Plots of each cultivar were divided into weedy and weed-free (maintained by herbicide application and hand-weeding) sections. Supplemental BYG seed was broadcast over all rice plots after planting. Propanil at 1.1 kg/ha ( the normal use rate) was applied post-emergence to all plots when BYG reached the three- to four-leaf stage. Urea at 110 kg N/ha was broadcast over all plots immediately before permanent flood. Four soil core subsamples (10 cm-diam. x 15 cm-deep) were obtained midway between rice rows in the weedy sections of each cultivar plot after BYG and rice had reached maximum above ground biomass accumulation and the permanent flood had been drained. Root tissues (a mixture of BYG and rice) were extracted from soil, dried, weighed, ground, and analyzed for levels of 13C isotope depletion, which is indicative of the relative levels of rice (C3 plant) and BYG (C4 plant) biomass present (Gealy et al. 2005). The percentage of rice and BYG roots present in each of these cores was then extrapolated from a 13C depletion curve developed by analyzing known proportions of pure root tissue obtained from each species.

Effect of herbicide inputs on weed suppression.

Plots were drill-seeded on May 22, 2004 at 323 seeds/m2 (except for the reduced recommended rates of 150 seeds/m2 for hybrid rice) as described above into 5 m long plots. There were four replications. The design was a split plot with eleven rice entries (subplots), three herbicide input levels (main plots: no grass herbicide; reduced input, thiobencarb (1/2X) 2.2 kg ai/ha DPRE (delayed preflood); and optimum herbicide inputs, clomazone 0.44 kg ai/ha DPRE, propanil 3.3 kg ai/ha PREFLD (preflood), and halolsulfuron 69 g ai/ha PREFLD), and four replications. DPRE herbicide treatments were soil-applied May 31 and PRFLD herbicide treatments were foliar-applied June 28. The entire experiment was sprayed one week postflood with triclopyr + halosulfuron to control broadleaf weeds and nutsedge (Cyperus esculentus). Weed control ratings were made July 5. Rice grain yields at harvest were determined as described earlier.

Results

Demonstration plots.

In demonstration plots in 2004, BYG was generally less prevalent in suppressive rice lines (especially XL8 and Stg96L-26-093), than in commercial lines (especially Lemont and Drew) (Figure 1). Photographs show that Ahrent and PI 312777 provided similar control at midseason (Figure 1). Control ratings at BYG maturity (averaged over all replications) for PI 312777, XL8, Stg96L-26-093, Lemont, Drew, and Ahrent were 57%, 77%, 57%, 28%, 38%, and 47%, respectively. Control ratings for other cultivars (not shown) from the same study were Bengal (62%) and Teqing (53%) (LSD 0.05 = 15%).

Evaluation of selections from crosses between rice and weed suppressive germplasm lines.

Of the 12 selections evaluated in 2004, 10 had weed control of 60% or greater, 10 had weed-free yields of 6500 kg/ha or greater, and 11 yielded at least 70% of their respective weed-free checks (Table 1). Although promising, these values were generally intermediate between those of Katy and the weed-suppressive parents, PI 312777 and PI 338046, usually did not attain the levels of the PI 312777, and were similar to several southern U.S. long grain cultivars such as Drew and Saber. Head rice yields of most of the selections were much higher than those of standard weed suppressive lines, and although they usually were less than those of southern long grain cultivars, most are in an ‘acceptable’ range (Table 1). Visual differences in weed suppression by the selections STG01L-130-116 and STG01L-30-084 are shown in Figure 2. Interestingly, Chinese indicas and hybrid rice cultivars controlled weeds at least as well as did the 12 selections, and had higher yield potentials. In a companion study (data not shown), the highest weed-free grain yields (>9000 kg/ha) were also obtained from indicas such as Teqing, TN1 (Taichung Native 1; PI 495830), the newly acquired Chinese indicas 4593, 4597, 4612 and hybrids such as CL XL8 and XP723, and the highest weed control in the test (>80%) was from TN1 and the hybrid, XL7. TN1 is in the pedigree of both PI 338046 and PI 312777, and has also been suppressive or allelopathic to several weed species in field and laboratory tests (Dilday et al. 1994; Dilday et al. 2001; Gealy and Dilday 2001; Olofsdotter 2001a, 2001b; Rimando and Duke 2003; Gealy and Moldenhauer 2005). Kong et al. (2004) recently reported that PI 312777 and Feng-Hua-Zhan, produced more allelochemicals (resorcinols, flavones, and hydroxamic acids) than did a non allelopathic variety, Hua-Gen-Xian, and that maximum levels of allelochemicals were produced at the six-leaf stage. These allelochemicals appeared to be produced by above ground plant parts and secreted through the roots, and were not actually produced by roots. Further, BYG plants induced rice plants to produce and release high levels of these allelochemicals.

Figure 1. Photos depicting midseason (July) BYG suppression in 10-m plots of several ‘suppressive’ (top row) and commercial (bottom row) rice lines in 2004. Top row (L-R): PI 312777, XL8, Stg96L-26-093 (PI 338046/Katy). Bottom row (L-R): Lemont, Drew, Ahrent.

Table 1. Weed control and agronomic data for selections from crosses with weed suppressive rice germplasm lines compared to various standards and parental lines in 2004.

Designation

Pedigree

Weed control rating (%)

Weed-free grain yield (kg/ha (H/T)1)

Weed-free grain yield (%)

Selections (F5 or later) from crosses with weed-suppressive lines

STG01L-30-084

PI338046/Katy//PI312777

64

6600 (61/70)*

79

STG01L-30-085

PI338046/Katy//PI312777

70

6040 (61/70)*

84

STG01L-30-089

PI338046/Katy//PI312777

56

6710 (na)

70

STG01L-30-097

PI338046/Katy//PI312777

49

3860 (na)

38

STG01L-30-115

PI338046/Katy//PI312777

70

6580 (63/70)*

71

STG01L-30-116

PI338046/Katy//PI312777

71

6860 (53/70)*

86

STG01L-30-117

PI338046/Katy//PI312777

71

6620 (58/70)*

85

STG01L-30-118

PI338046/Katy//PI312777

75

6600 (53/70)*

80

STG02P-39-124 (STG96L-26-092 was female in cross)

PI338046/Katy//PI312777

70

7480 (na)

76

RU9701151

PI338046/Katy

64

7600 (66/73)**

71

STG96L-26-092

PI338046/Katy

66

6920 (na)

75

STG96L-26-093

PI338046/Katy

66

7380 (na)

71

Weed suppressive and commercial standards

PI 312777

T65*2/TN1

80

8750 (45/70)+

88

PI 338046

IR8*2//B598A4-18-1 *2/TN1

74

7280 (60/??)+

86

Katy (Southern long grain)

 

59

6630 (66/73)**

74

Drew (Southern long grain)

Newbonnet/Katy

65

7150 (66/74)**

81

Lemont (Southern long grain)

Lebonnet/CI9881/PI331581

51

6240 (65/74)**

56

Saber (Southern long grain)

Gulfmont/ RU8703169/ Teqing

80

6460 (66/74)+

92

L205 (CA long grain)

 

58

5050 (na)

70

M202 (CA medium grain)

 

59

4540 (na)

59

CL-XL8 (150 seeds/m2)

Proprietary hybrid

64

9420 (63/72)+

61

CL-XL8 (430 seeds/m2)

Proprietary hybrid

89

9040 (63/72)+

91

Teqing

Chinese indica

68

9110 (49/70)+

83

4612

Chinese indica

80

9620 (41/69)+

76

LSD (0.05)

 

15

1245

19

1 ‘H/T’ indicates % head rice yield / % total rice yield (% total rice is % grain yield remaining after dehulling; % head rice is % grain yield remaining after dehulling and milling). ‘*’ indicates 2002 data from Stuttgart, ‘**’ indicates 1997 data averaged from four locations in Arkansas, ‘+’ indicates data from various other sources, ‘na’ indicates data not available.

Figure 2. Weedy plots of STG01L-130-116 (left) and STG01L-30-084 (right) in 2004. Weed control ratings, weed-free grain yields, and weedy grain yields are shown in Table 1. Additional baseline agronomic values for STG01L-130-116 and STG01L-30-084, respectively, in weed free plots were as follows: days to heading, 88 and 88 days; maximum tiller production, 1380 and 1120 tillers/m2; and maximum height, 103 and 108 cm. By comparison, these agronomic values were 90, 91, and 93 days, 1730, 1420, and 1680 tillers/m2, and 106, 107, and 93 cm, respectively for the parental lines PI 312777, Katy, and PI 338046 (not shown).

Estimating barnyardgrass and rice root distribution using 13C analysis.

Suppressive lines generally had lower BYG root densities at both soil depths and greater rice root densities and a greater proportion of rice than BYG roots in the top 5 cm of soil compared to commercial cultivars (Table 2). At the 5 to 15 cm depth, commercial rice cultivars produced more root mass than did the suppressive lines. The net result of this differential distribution is that suppressive rice lines produced 64 to 80% of their roots in the top 5 cm of soil, whereas less than 50% of roots of commercial cultivars were produced in this soil section. Thus, the root architecture of these suppressive lines may favor their competitiveness (and/or allelopathic activity) in the upper soil from which most weeds germinate and emerge. PI 312777 and PI 338046 have produced more root mass than the commercial cultivars, Lemont and M-201 (Dilday et al., 2001). Release of phytotoxins from roots of these suppressive lines as noted by Kong et al. (2004) could enhance their total weed interference. However, the genetic control of allelopathic activity against BYG was apparently independent from that of rice root morphology in an analysis of a mapping population of rice (japonica x indica) using quantitative trait loci analysis (Bach-Jensen et al. 2001).

Table 2. Distribution of rice and BYG root biomass in soil cores sampled between rice rows in weedy plots in 2003 as estimated by 13C isotope depletion analysis.1

Cultivar/line

BYG root dry weight density

Root dry weight density

Root dry weight

BYG root dry weight density

Root dry weight density

Root dry weight

Roots in the top 5 cm

 

0 to 5 cm depth

5 to 15 cm depth

 
 

mg / 1000 cm3 soil

mg / 1000 cm3 soil

% of total rice + BYG root weight

mg / 1000 cm3 soil

mg / 1000 cm3 soil

% of total rice + BYG root weight

% of total in 15 cm

Suppressive lines

PI 312777

792

750

52

101

152

69

71

Teqing

617

591

59

20

75

77

80

XL8 hybrid

480

544

53

41

151

78

64

STG96L-26-093
(PI338046 / Katy)

961

850

45

46

209

80

67

Commercial standards

Kaybonnet

815

460

40

77

325

80

41

Francis

1404

298

20

74

230

77

39

Lemont

1645

581

25

109

305

70

49

1Data presented are means of three replications, but were not statistically analyzed.

Differential 13C isotope depletion could also be used to evaluate root interactions with several other grass and broadleaf weed species found in or near rice fields because the delta 13C depletion values for these species appear to be similar to those of BYG. In a small companion study conducted in 2003 (unpublished), delta 13C depletion values (mean +/- standard deviation) for C4 species such as BYG, bearded sprangletop, broadleaf signalgrass, palmer amaranth (Amaranthus palmeri), and purslane (Portulaca oleracea) were 12.4 +/-0.1, 14.2 +/-0.1, 12.7 +/-0.3, 13.3 +/-0.1, and 12.1 +/-0.2, respectively. By comparison, depletion values for the C3 species Cocodrie rice and the aquatic weed, red stem (Ammannia coccinea Rottb.) were 28.0 +/-0.1 and 28.8 +/-0.3, respectively. Thus, it is apparent that roots of any of the C4 species evaluated in this study would be readily distinguishable from the C3 rice or red stem plants in mixtures because of the sizeable differences in the 13C depletion values of these C3 and C4 plants. The small standard deviation values for 13C depletion show that the measurements are highly stable within plant species.

Before application of supplemental N fertilizer and permanent flood, all rice entries in weed free plots had removed 83 to 87% of the NO3 initially present in the top 8 cm of soil, except for Lemont, which removed only 68% (analyzed by University of Arkansas Soil Testing Laboratory, Marianna, AR; data not shown). At the 8 to 15 cm depth, PI 312777, XL8 and Kaybonnet had removed 73 to 80% of the initial NO3, whereas all other entries had removed only 45 to 56% of this NO3. Visual control ratings at BYG maturity were positively correlated (R=0.48; R=0.88 without Kaybonnet) with the percentage NO3 depletion at the 8 to 15 cm depth. Kaybonnet appeared to be an outlier and provided only 40% control of BYG (equal to the lowest of all rice lines) even though it depleted as much NO3 as did PI 312777, which provided 57% BYG control (not shown). This anomaly suggests that factors other than preflood nitrogen depletion (i.e. allelopathy) are also involved in the suppressive activity of PI 312777 and XL8.

Effect of herbicide inputs on weed suppression.

In untreated plots, control levels on July 5, 2004 for suppressive rice types such as PI 312777, 4593, and CL XL8, were less than 60%, but were generally greater than for the commercial standards (Table 3). Weed suppression from the reduced thiobencarb rate and from the optimum herbicide input treatment generally did not differ among rice lines and was 83 to 100% (Table 3). Thus, even lower levels of thiobencarb or of the three herbicides present in the optimum treatment may have provided good BYG control under these environmental conditions. All herbicide treatments were particularly effective against BYG in 2004, in part because the timing and quantities of rainfall appeared to be optimal for herbicide activity and rice productivity. In a similar experiment in 2003, thiobencarb appeared to be much less effective against BYG than in 2004, but there were still few differences among cultivars (Table 3). Earlier reports have shown that greater economic returns would be expected when using reduced rates (1.1 kg/ha or zero) of propanil on suppressive rice lines than when using normal rates of propanil on commercial cultivars (Gealy et al. 2003). Data in this paper (note relative comparisons of % weed suppression by PI 312777, Drew and Saber in Table 1, Table 3, and Figure 1) and results from many previous experiments indicate that weed suppression results can vary from year-to-year, apparently as a response to environment and cropping practices. Thus, contingencies for production factors such as herbicide rate, environmental variations, rainfall and irrigation timing, and possibly seeding density may be necessary to achieve reasonable levels of long term success in these systems.

Table 3. Effect of herbicide inputs on BYG control in rice plots at Stuttgart.


Rice cultivar/line

BYG control rating, July 5, 2004 (and July 9, 2003)1

Untreated check
(no grass herbicide applied)

Thiobencarb
2.2 kg ai/ha DPRE
(reduced input)

Clomazone 0.44 kg ai/ha DPRE;
Propanil 3.3 kg ai/ha PREFLD;
Halolsulfuron 69 g ai/ha PREFLD
(optimum herbicide input)

Suppressive types (%)

PI 312777

59 (13)

95 (66)

100 (100)

4593 (Chinese indica)

58 (68)

94 (65)

100 (100)

CL XL8 (150 seed/m2) hybrid rice

43 (35)

90 (30)

100 (100)

CL XL8 (323 seed/m2) hybrid rice

54 (tni)

94 (tni)

100 (tni)

XP710 (150 seed/m2) hybrid rice

16 (tni)

91 (tni)

95 (tni)

STG96L-26-093
(PI338046 / Katy)

20 (tni)

83 (tni)

100 (tni)

STG01L-30-117

19 (tni)

88 (tni)

99 (tni)

Commercial standards

Drew

18 (24)

83 (48)

99 (100)

Francis

35 (15)

93 (55)

100 (100)

Rexmont

43 (18)

90 (21)

99 (100)

Saber

26 (24)

93 (58)

95 (100)

LSD (0.05)

16 (24)

1 July 9, 2003 BYG control ratings (in parenthesis) from a 2003 study are included for comparison. The 2003 study was planted, treated, and rated using methods and dates essentially identical to those in 2004. In 2003 data, ‘tni’ indicates that the treatment was not included.

Conclusions

Development of viable weed suppressive rice cultivars require broad collaborations among breeders, molecular geneticists, weed scientists, agronomists, physiologists, ecologists, and others. In all likelihood, none of the rice lines currently under investigation in Arkansas or conceived in theory will be capable of complete weed control. However, weed suppressive lines may be useful in conventional production systems if they can be used in combination with low herbicide inputs, or in minimum input production systems (organic farms, commercial waterfowl habitat, etc.) where sub optimum grain productivity and quality may be more acceptable to customers. Whether by competitive mechanisms, allelopathic mechanisms, or both, indica lines and commercial hybrids recently introduced to the southern U.S. seem to be well suited to suppress weeds and produce high yields in drill-seeded systems of the southern U.S.

Acknowledgments

We thank Howard Black for his invaluable technical assistance, the University of Arkansas Soil Testing and Research Laboratory, Marianna, AR (http://www.uark.edu/depts/soiltest/) for conducting soil NO3 analyses, the University of Arkansas Stable Isotope Laboratory, Fayetteville, AR (http://biology.uark.edu/uasil/home.html) for conduction 13C isotope depletion analyses, and the Arkansas Rice Research and Promotion Board for partial funding of this research.

References

Bach-Jensen L, Courtois B, Shen L, Li Z, Olofsdotter M, and Mauleon R (2001). Locating genes controlling allelopathic effects against barnyardgrass in upland rice. Agronomy Journal 93, 21-26.

Dilday RH, Lin J, and Yan W (1994). Identification of allelopathy in the USDA-ARS rice germplasm collection. Australian Journal of Experimental Agriculture 34, 907-910.

Dilday RH, Mattice JD, Moldenhauer KA, and Yan W (2001). Allelopathic potential in rice germplasm against ducksalad, redstem and barnyardgrass. Journal of Crop Production 4, 287-301.

Estorninos LE Jr, Gealy DR, Talbert RE, and Gbur EE (2005a). Rice and red rice (Oryza sativa) interference. I. Response of red rice to sowing rates of tropical japonica and indica rice cultivars. Weed Science 53 (in press).

Estorninos LE Jr, Gealy DR, Talbert RE, and Gbur EE (2005b). Rice and red rice interference: II. Rice response to population densities of three red rice (Oryza sativa) ecotypes. Weed Science 53 (in press).

Gealy, DR and Dilday RH (2001). Progress in suppressing barnyardgrass (Echinochloa crus-galli) with foreign and domestic rice cultivars. In Bobby R. Wells Rice Research Studies - 2000, eds. RJ Norman and J-F Meullenet. Arkansas Agricultural Experiment Station, Series 485. Fayetteville, AR: University of Arkansas, pp. 62-68.

Gealy DR, Estorninos LE Jr, Gbur E, and Chavez R (2005). Interference interactions of two rice cultivars and their F3 hybrid with barnyardgrass (Echinochloa crus-galli) in a replacement series study. Weed Science 53 (in press).

Gealy DR and Moldenhauer KA (2005). Progress in developing weed suppressive rice cultivars for the southern U.S. in Handbook of Sustainable Weed Management. eds. H Singh, D Batish, and R Kohli. Binghamton, NY:Haworth Press. (in press; www.HaworthPress.com).

Gealy DR, Wailes EJ, Estorninos LE Jr, and Chavez RSC (2003). Rice cultivar differences in suppression of barnyardgrass (Echinochloa crus-galli) and economics of reduced propanil rates. Weed Science 51, 601-609.

Kong CH, Xu XH, and Zhou SC (2004). Weed suppressing effect and its mechanism of allelopathic rice variety. In Plant Protection Towards the 21st Century, Proceedings, 15th International Plant Protection Congress, Beijing, China, ed. Y. Y. Guo. Beijing:Foreign Language Press, p. 611.

Moldenhauer KAK, Gibbons JW, Lee FN, Norman RJ, Bernhardt J, Dilday RH, Rutger JN, Blocker MM, and Tolbert AC (1999). Breeding and evaluation for improved rice varieties-the Arkansas rice breeding and development program. In Bobby R. Wells Rice Research Studies 1998, eds. RJ Norman and TH Johnson. Arkansas Agricultural Experiment Station, Series 468. Fayetteville, AR: University of Arkansas, pp. 20-27.

Olofsdotter M (2001a). Getting closer to breeding for competitive ability and the role of allelopathy - an example from rice (Oryza sativa). Weed Technology 15, 798-806.

Olofsdotter M (2001b). Rice - A step toward use of allelopathy. Agronomy Journal 93, 3-8.

Olofsdotter M, Bach-Jensen L, and Courtois B (2002). Improving crop competitive ability using allelopathy-an example from rice. Plant Breeding 121, 1-9.

Ottis B, Smith K, Scott R, and Talbert R (2005). Rice (Oryza sativa L.) yield and quality as affected by cultivar and red rice (Oryza sativa L.) density. Weed Science 53 (in press).

Rimando AM, and Duke SO (2003). Studies on rice allelochemicals. In Rice: Origin, History, Technology, and Production-Crop Production Series #6149, eds. C. Wayne Smith and Robert H. Dilday. New York:John Wiley & Sons, Inc., pp. 221-244.

Top Of PageNext Page