Effect of allelopathic rice varieties combined with cultural management ...

13 downloads 2504 Views 136KB Size Report
Jan 2, 2008 - pattern and cultural management including planting density, flooding depth ... management; low-dose herbicide; weed suppression; paddy field.
Pest Management Science

Pest Manag Sci 64:276–282 (2008)

Effect of allelopathic rice varieties combined with cultural management options on paddy field weeds Chui-Hua Kong,1,2∗ Fei Hu,2 Peng Wang1 and Jing-Lun Wu3 1 Institute

of Applied Ecology, Chinese Academy of Sciences, Shenyang 110016, China China Agricultural University, Guangzhou 510642, China 3 Institute of Plant Protection, Jiangsu Academy of Agricultural Sciences, Nanjing 210014, China 2 South

Abstract BACKGROUND: A number of techniques, including cultural management, allelopathy and bioherbicide, have been considered as alternatives for synthetic herbicides, but successful weed control will require the careful integration of these multiple techniques. This study was conducted to assess the use of allelopathic rice varieties in combination with cultural management options on paddy weeds, in order to develop an allelopathy-based technique to reduce herbicide use in paddies. RESULTS: The weed-suppressive effects of the rice varieties tested varied highly with allelopathic trait, planting pattern and cultural management including planting density, flooding depth and duration and supply of nitrogen. Allelopathic rice varieties PI312777 and Huagan-1 demonstrated much stronger weed suppression than the non-allelopathic variety Huajianxian under the same planting pattern and cultural management. Their weed-suppressive effect was increased with cultural management options. In particular, if integrated cultural management options of allelopathic rice varieties included a low-dose (bensulfuron-methyl, 25 g AI ha−1 , a third of the recommended dose) herbicide application, the emergence and growth of most weeds found in paddy fields was completely controlled. No grain yield reduction for allelopathic varieties occurred under integrated cultural management options, whereas with the non-allelopathic variety a reduction of up to 45–60% was measurable even with the low-dose herbicide application. CONCLUSION: The allelopathic potential of rice varieties will likely have a great impact on paddy weed control if integrated with cultural management options and application of low doses of herbicides. Therefore, it is feasible to reduce herbicide input in paddies if allelopathic rice is grown under integrated cultural management practices.  2008 Society of Chemical Industry

Keywords: Oryza sativa L.; allelopathy; cultural management; low-dose herbicide; weed suppression; paddy field

1 INTRODUCTION Weeds pose an important biological constraint to rice productivity. Today, on a global scale, rice production is characterized by heavy use of synthetic herbicides. The fate of these herbicides is an object of great concern for agriculture, environment and human health. In this regard, there needs to be an alternative to such a heavy dependence on herbicides in paddy ecosystems.1 – 3 Several cultural practices, such as the transplanting of rice plants at a specific growth stage or maintenance of a certain depth of water coverage, are used to mitigate harm from weeds.4 However, these cultural practices do not provide adequate weed control in the absence of herbicides in paddies. Improvement in rice cultivars is an area that could be developed as a weed management strategy, and thus the development of weed-suppressing rice

cultivars is highly desirable. A few rice varieties with weed-suppressive effects have been identified from several rice germplasm collections.5 – 7 These rice varieties are very competitive against weeds in paddy fields, but their weed-suppressive effects have not been fully explained by considering just the physical characteristics of being strong in competition for light and nutrients.8,9 An increasing number of studies have shown that the participation of allelopathy, which results from the allelochemicals released from rice itself is believed to play an important role.10 – 12 Therefore, the weed-suppressive effect is a combination of competition for resources and production and the release of allelochemicals from these rice varieties. Research and development of allelopathic rice cultivars has been carried out throughout the world. Although significant results leading to the

∗ Correspondence to: Chui-Hua Kong, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang 110016, China E-mail: [email protected] (Received 26 July 2007; revised version received 13 September 2007; accepted 13 September 2007) Published online 2 January 2008; DOI: 10.1002/ps.1521

 2008 Society of Chemical Industry. Pest Manag Sci 1526–498X/2008/$30.00

Combination of cultural conditions and allelopathic rice varieties for control of paddy weeds

development of a commercially acceptable allelopathic cultivar have not been obtained so far, a few studies have clearly shown that the use of allelopathic rice cultivars can definitely reduce the cost of paddy weed control, particularly by reducing the amount of herbicide used.3,13,14 Such an allelopathybased technique for paddy weed control is most easily transferable to the low-input management systems prevailing throughout most Asian rice-farming systems.15,16 Actually, a number of techniques have been evaluated as alternatives to synthetic herbicides in weed management systems, including cultural management, allelopathy and bioherbicide. However, significant developments in weed management for lowinput agriculture will require the careful integration of these multiple techniques.17,18 Rice allelopathy might prove useful in controlling paddy weeds and increasing grain yields, but paddy weed control by using an allelopathic rice variety alone may not be adequate. Therefore, several management options will need to be combined with the suppressive effect of the allelopathic rice variety to facilitate effective weed control at reduced herbicide rates.3,16 In this study, the weed-suppressive effects of two allelopathic rice varieties and a non-allelopathic rice variety were evaluated under several cultural management options with and without allelochemical or low-dose herbicide application in greenhouse studies and field experiments, in an attempt to develop an allelopathybased technique to reduce herbicide use in paddy weed control.

2 MATERIALS AND METHODS 2.1 Plant material, paddy soil and allelochemical Three rice (Oryza sativa L.) varieties were selected for these studies. Allelopathic rice variety PI312777 was obtained from the USDA-ARS rice germplasm collection.5 Huagan-1, a variety with a strong allelopathic trait, was obtained from a Chinese rice germplasm collection.7 The Huajianxian variety is a commercial cultivar in South China that does not have an allelopathic trait.19 For greenhouse studies, ripened weed seeds and paddy soils were collected from rice fields at the Experimental Farm of South China Agricultural University (Guangzhou, China). Three weed species, Echinochloa crus-galli (L.) Beauv, Cyperus difformis L. and Eclipta prostrata L., were selected. The seeds of each weed were dried under sunlight and then stored in sealed glass jars. Soils were air dried, mixed and then sieved (2 mm mesh) to remove plant tissues. Soil was pH 6.15, with an organic matter content of 17.7 g kg−1 and the following fertility status: total N 0.87 g kg−1 , available N 108.64 mg kg−1 ; total P 0.28 g kg−1 , available P 19.76 mg kg−1 ; total K 0.91 g kg−1 , available K 102.9 g kg−1 . For use in greenhouse experiments, an allelochemical participating in the defence of rice against paddy Pest Manag Sci 64:276–282 (2008) DOI: 10.1002/ps

weeds, 5,7,4 -trihydroxy-3 ,5 -dimethoxyflavone, was isolated from allelopathic rice PI312777 plants by methods developed in the authors’ laboratory.20 2.2 Greenhouse experiments Two experiments were carried out in the greenhouse with the seed, soil and allelochemical described above. The experiments were conducted with five replications and repeated twice under identical conditions. The first experiment was to evaluate the allelopathic interference of test rice varieties with three paddy weeds (E. crus-galli, C. difformis and E. prostrata) by comparing inhibition of weed growth in rice/weed coplantings with and without activated charcoal to neutralize allelochemical effects. Ninety pregerminated rice seeds (ca 1 mm root growth) of each variety were sown in three rows in boxes (60 cm length, 30 cm width, 10 cm height) containing 10 kg paddy soil. The row distance was 20 cm, and 30 rice seeds were sown in one row. Twenty pregerminated test weed seeds (1 mm) were sown in two rows between each rice row (ten seeds in one row). In half of all the boxes used, trenches (10 cm depth, 1 cm width) were formed in the centre between the rice and test weed rows and filled with activated charcoal (5 g kg−1 soil). This was not done in the other half of the boxes. Boxes sowed only with test weed species, with or without trenches, served as controls. Any plant species germinating in the boxes other than rice or test weed species were hand removed soon after they were detected. The shoots of test weed species were harvested after 6 weeks and dried for at least 48 h at 80 ◦ C, and their dry weights were each determined. The second experiment was to determine the weedsuppressive effect of test rice varieties by allelochemical addition. Boxes sowed with the different rice varieties described in Section 2.1 were divided into two groups. To one group was added the rice allelochemical 5,7,4 -trihydroxy-3 ,5 -dimethoxyflavone at a test concentration (50 µg g−1 soil) that inhibited significantly the growth of paddy weeds.20 To the other group was added water only. Boxes filled with 10 kg paddy soil only, with neither the allelochemical addition nor the rice variety tested, served as controls. Emergent weeds from paddy soil in boxes were harvested and carefully identified after 6 weeks. They were divided into three groups of E. crus-galli, Cyperus ssp. and other weeds, and the shoot dry weights were each determined. All boxes described above were placed in a greenhouse with 20–30 ◦ C night and daytime temperatures, a light intensity of approximately 350 mol m−2 s−1 and 65–90% relative humidity. Boxes were randomized and watered by tap once a day. 2.3 Field experiments A series of field experiments was conducted at the Zhuhai Experimental Station of South China Agricultural University during March–November 2005 and 2006. The field had previously been 277

CH Kong et al.

planted with rice for several seasons where the weeds E. crus-galli, C. difformis, Cyperus iria L., Lindernia procumbens Philcox, Alternanthera sessilis R.Br., E. prostrata, Leptochloa chinensis (L.) Nees and Monochoria vaginalis (Burm.f.) Presl ex Kunth occurred in a total of 111 ± 23 plants m−2 . Among these weeds, E. crus-galli (35.6% of all plants) and Cyperus spp. (17.8% of all plants) were dominant. The field was divided into numerous plots (2 × 3 m) that were in a completely randomized design with three replicates for the following three experiments. The first experiment was to evaluate the weedsuppressive effect of test rice varieties under three different planting patterns. The three rice varieties, PI312777, Huagan-1 and Huajianxian, were planted in randomly selected plots by means of direct seeding, seedling throwing and transplanting. For direct seeding, seeds were directly sown in plots. For seedling throwing, seedlings at the two-leaf stage were broadcast into plots. For transplanting, seedlings at the three-leaf stage were hand transplanted into plots. All planting densities were 3.0 × 105 plants ha−1 . No chemical or hand weeding was conducted during the trials. The weeds were harvested by cutting plants at soil level at the rice heading stage. The total dry weights of the various weeds in each plot were recorded. The second experiment was to evaluate the weed-suppressive effect of test rice varieties under various cultural practices. This was evaluated only in plots where the planting pattern was transplanting. As in the experiment described above, seedlings of the three rice varieties at the three-leaf stage were each transplanted into the experimental plots but had variations in cultural treatments. Cultural treatment variables included: (a) planting density – rice seedlings were transplanted into the plots at densities of 1.5 × 105 , 3.0 × 105 or 4.5 × 105 plants ha−1 ; (b) flooding depth and duration – rice growth was conducted with a different flooded depth (3 or 12 cm) and duration (5 or 15 days) after the seedlings were transplanted into the plots at a density of 4.0 × 105 plants ha−1 ; (c) nitrogen supply – three levels of N (105, 210 and 315 kg ha−1 ) were applied at 15 days after the seedlings were transplanted into the plots at a density of 4.0 × 105 plants ha−1 . No weed management was employed during the trials. The weeds were harvested by cutting plants at soil level at the rice heading stage, and each species was carefully identified and divided into three groups of E. crus-galli, Cyperus ssp. and other weeds. Their dry weights in each respective plot were measured. The third experiment was to evaluate the weedsuppressive effect of the selected rice varieties under integrated cultural management options with or without adding a herbicide component at a lower rate than recommended for chemical control alone. This experiment was carried out in the two seasons of spring (March–July) and autumn (July–November) 278

of 2006. Seedlings of three rice varieties at the threeleaf stage were each transplanted into the plots at a density of 4.0 × 105 plants ha−1 . Rice growth was carried out with a 5 cm flooded depth and 10 days flooded duration. A level of 315 kg ha−1 N supply was applied at 15 days after transplanting. One set of plots had no chemical or hand weeding. A second set was treated with bensulfuron-methyl 700 g kg−1 WP (Beron 100 WP) applied at a dose of 25 g AI ha−1 6 days after transplanting. This rate was one-third of the recommended dose for weed control in paddy ecosystems in China. After 110 days, both rice and weeds were harvested. Weed species in each plot were carefully identified and divided into two groups of grass and broadleaf weeds. The shoot dry weights of both grass and broadleaf weeds were measured. Rice yields were also recorded. In all field experiments described above, plots not planted with rice served as controls. The plots had received no herbicide treatment previously. Besides various treatments designed in the three experiments, all test rice varieties received fertilizer treatment (N, P2 O5 and K2 O at rates of 7.5, 9.0 and 8.5 g m−2 respectively) 1 day before the plots were saturated with water. Other management in paddy fields was carried out in accordance with local practices. 2.4 Data and statistical analysis The weed-suppressive effect was assessed by a comparison of the weed dry weights in the treated and control boxes or plots. Inhibition percentage was calculated by (1 − treatment/control) ×100 based on weed dry weight from the boxes or plots for each independent determination. Means ± SED from independent experiments for each determination are shown. The data were subjected to analysis of variance (ANOVA) in SAS with mean separations by Duncan’s multiple range test at the 0.05 probability level.

3 RESULTS Greenhouse experiments showed that the growth of the three paddy weeds E. crus-galli, C. difformis and E. prostrata was significantly inhibited by the two allelopathic rice varieties PI312777 and Huagan1, and the least inhibition occurred in the nonallelopathic rice variety Huajianxian. There was threefold increase in the weed-suppressive effect of allelopathic rice varieties as compared with the nonallelopathic variety, but no significant difference was observed between the two allelopathic rice varieties (Fig. 1A). In addition, the suppressive degree of the allelopathic rice varieties varied among the three paddy weeds tested. Inhibition was always much stronger in the grass weeds E. crus-galli and C. difformis than in the broadleaf weed E. prostrata. Furthermore, the weed-suppressive effect of both allelopathic rice varieties, PI312777 and Huagan-1, was significantly decreased when activated charcoal was applied to Pest Manag Sci 64:276–282 (2008) DOI: 10.1002/ps

Combination of cultural conditions and allelopathic rice varieties for control of paddy weeds 100

A

100

E. prostrata

80

% Inhibition

% Inhibition

80

E. crus-galli C. difformis

60 40 20

Seeding Throwing Transplanting

60 40 20

0 PI312777

Huagan-1

Huajianxian 0

100

PI312777

B

% Inhibition

80

Huagan-1

Huajianxian

Figure 2. Weed-suppressive effect of rice varieties tested under different cultural patterns. The total inhibition percentages (means ± SED) of various paddy weeds for independent experiments with three replicates for each determination are shown.

60 40 20 0 PI312777

Huagan-1

Huajianxian

Figure 1. Inhibition of rice varieties tested on different weed species in the greenhouse: A, soils without activated carbon; B, soils with activated carbon. Means ± SED from independent experiments with five replicates for each determination are shown.

the soils. The inhibition of E. crus-galli and C. difformis was decreased from around 70% to 30%. However, no reduction occurred in weed suppression by the non-allelopathic variety Huajianxian (Fig. 1B). This indicated that activated charcoal adsorbed the allelochemicals released from the two allelopathic varieties to mitigate weed suppression, and had no effect on the non-allelopathic variety without allelochemicals. These results implied that the weed suppression by the rice varieties tested was closely correlated with their allelopathic traits releasing phytotoxic allelochemicals into the soil. The presence of the phytotoxins released from allelopathic rice to paddy soils may be able to contribute to the suppression of paddy weeds. In order to confirm that the effect resulted from phytotoxins released by allelopathic rice varieties, an allelochemical, 5,7,4 trihydroxy-3 ,5 -dimethoxyflavone, involved in rice allelopathy was employed in greenhouse experiments. As a result, the weed-suppressive effect of all the rice varieties tested, either allelopathic or non-allelopathic, was significantly increased (Table 1). It is thought that, in competition with paddy weeds, the presence of

allelochemical enhances the weed-suppressive effects of given rice varieties. Field experiments showed similar results. Allelopathic varieties PI312777 and Huagan-1 demonstrated much stronger weed suppression than the non-allelopathic variety Huajingxian under the same cultural conditions. Moreover, the weed-suppressive effect varied greatly with rice planting patterns and cultural management options, including planting density, flooding depth and duration and N supply. Subsequently, the weed suppression of allelopathic rice varieties was much stronger under throwing or transplanting than under direct seeding, but no significant difference was observed between throwing and transplanting either allelopathic or non-allelopathic varieties tested (Fig. 2). This indicated that that paddy weed control was more effective under the transplanting or throwing than under the direct seeding pattern. Accordingly, the weed-suppressive effect of rice varieties was investigated using different cultural management options under the transplanting pattern. A high planting density could improve rice competitiveness and suppress the growth of paddy weeds effectively. Flooding could suppress the growth of paddy weeds, but flooding duration was much more effective than flooding depth (Table 2). The key period for rice competition with paddy weeds is at 12–19 days after transplanting. In this period, applied N could slightly stimulate E. crus-galli growth but effectively suppress the growth of Cyperus ssp. and other weeds, particularly broadleaf weeds (Table 2). However, there were

Table 1. Inhibition of paddy weeds by rice varieties combined with application of the allelochemical 5,7,4 -trihydroxy-3 ,5 -dimethoxyflavone in the greenhouse

Inhibition (%) (± SEM)ab Rice variety

E. crus-galli

Cyperus ssp.

Other weeds

PI312777 Huagan-1 Huajianxian

88.6 (±9.1) a [57.9 (±5.1) b] 92.7 (±7.1) a [60.5 (±7.4) b] 43.3 (±4.5) c [24.3 (±3.8) d]

87.3 (±9.9) a [54.6 (±8.2) b] 89.9 (±9.7) a [59.1 (±5.7) b] 45.3 (±4.4) c [20.1 (±4.1) d]

67.8 (±6.8) a [35.6 (±5.6) b] 70.3 (±6.5) a [42.2 (±5.2) c] 35.1 (±4.2) b [12.3 (±1.9) d]

a b

Data in brackets are inhibition with rice varieties tested with no application of allelochemical (only water was applied during the experiments). Data in a column followed by the same letter are not significantly different, P = 0.05.

Pest Manag Sci 64:276–282 (2008) DOI: 10.1002/ps

279

CH Kong et al. Table 2. Inhibition of paddy weeds by rice varieties under different cultural management options in field

Inhibition (%) (± SEM)a Rice variety PI312777

Treatment −1

Density (plants ha

)

Flooding depth (cm) Flooding duration (days) N supply (kg ha−1 )

Huagan-1

Density (plants ha−1 )

Flooding depth (cm) Flooding duration (days) N supply (kg ha−1 )

Huajianxian

Density (plants ha−1 )

Flooding depth (cm) Flooding duration (days) N supply (kg ha−1 )

a

Treatment level

E. crus-galli

Cyperus ssp.

Other weeds

1.5 × 105

3.0 × 105 4.5 × 105 3 12 5 15 105 210 315

44.8 (±4.6) a 65.1 (±7.3) b 76.6 (±9.5) c 46.6 (±3.6) a 47.5 (±4.9) a 69.1 (±5.1) b 78.4 (±6.4) c −1.7 (±0.1) d −5.2 (±0.7) e −5.9 (±1.1) e

45.6 (±5.1) a 53.5 (±4.8) b 64.4 (±6.3) c 35.7 (±3.1) d 55.3 (±4.5) b 38.1 (±3.8) d 54.8 (±5.5) b 32.7 (±2.1) d 41.1 (±3.3) a 77.8 (±8.7) e

44.1 (±5.6) a 67.4 (±7.3) b 75.5 (±6.9) c 38.7 (±2.8) d 44.5 (±3.4) a 10.3 (±1.1) e 28.9 (±1.9) f 47.4 (±3.9) a 59.6 (±4.8) ab 79.5 (±6.6) c

1.5 × 105 3.0 × 105 4.5 × 105 3 12 5 15 105 210 315

55.2 (±4.6) a 67.4 (±5.8) b 86.3 (±7.3) c 53.5 (±4.4) a 56.7 (±5.1) a 73.3 (±6.6) d 84.1 (±8.9) c 3.2 (±1.1) e −5.7 (±0.7) f −6.8 (±1.5) f

52.5 (±5.6) a 65.4 (±6.3) b 74.4 (±6.8) c 51.6 (±4.9) a 54.4 (±5.3) a 52.1 (±6.1) a 64.8 (±6.9) b 37.8 (±2.2) d 48.1 (±3.6) a 79.7 (±7.1) e

52.8 (±4.6) a 54.4 (±5.1) a 62.5 (±5.8) b 52.4 (±6.6) a 57.9 (±5.7) ab 16.3 (±1.8) c 39.8 (±2.6) d 50.1 (±4.8) a 61.9 (±7.4) b 70.9 (±7.8) e

1.5 × 105 3.0 × 105 4.5 × 105 3 12 5 15 105 210 315

25.5 (±2.3) a 35.5 (±2.8) b 42.2 (±5.3) bc 24.7 (±1.8) a 27.9 (±2.2) a 36.2 (±4.5) b 45.4 (±3.8) c −5.7 (±0.8) d −10.3 (±1.2) e −12.4 (±1.4) e

25.1 (±3.2) a 26.1 (±2.4) a 33.1 (±2.8) b 21.5 (±1.9) c 24.7 (±3.6) a 22.8 (±2.4) c 35.9 (±2.8) b 22.7 (±1.6) c 41.4 (±3.2) d 72.1 (±6.4) e

35.1 (±3.4) a 36.5 (±3.6) a 41.3 (±3.2) ab 19.5 (±1.6) c 22.1 (±1.8) c 13.1 (±0.8) d 25.4 (±1.2) c 43.7 (±4.8) b 54.9 (±5.6) e 66.9 (±7.2) f

Data in a column followed by the same letter are not significantly different, P = 0.05.

no similarities between allelopathic or non-allelopathic varieties tested. Based on the results described above, the magnitude of the weed-suppressive effects of the rice varieties tested varied significantly with cultural management options. Allelopathic varieties with a single cultural management option enhanced around 10–20% of weed suppression, while flooding duration was the most effective cultural option in enhancing weed suppression (Table 2). In particular, excellent control of grass weeds and strong suppression of broadleaf weeds occurred with allelopathic varieties integrated with cultural management options (Table 3). Although several broadleaf weeds, such as M. vaginalis and E. prostrata, were not completely controlled, their density was significantly reduced. Furthermore, allelopathic rice varieties integrated with cultural management options and a low dose of herbicide (bensulfuron-methyl, 25 g AI ha−1 ) completely controlled the emergence and growth of most paddy weeds found. In particular, herbicide application resulted in a significant reduction of broadleaf weeds (Table 3). It should be noted that the herbicide was applied at only a third of the recommended dose for weed control in paddy fields in China. 280

No grain yield reduction occurred in allelopathic rice varieties PI 321 777 and Huagan-1 under integrated cultural management options, whereas, with the nonallelopathic rice variety Huajianxian, a reduction of up to 45–60% compared with the two allelopathic varieties was measurable even with the low-dose herbicide application (Table 3).

4 DISCUSSION Rice allelopathy can potentially be used to improve weed management in paddy ecosystems. An increasing number of studies have clearly shown that allelopathic and/or weed-suppressive rice varieties can suppress certain weed species including barnyard grass (E. crus-galli), smallflower flatsedge (C. difformis), ducksalad [Heteranthera limosa (Sw.) Willd./Vahl], purple redstem (Ammania coccinea Rottb.), palisadegrass [Brachiaria brizantha (Hochst) Stapf.], signalgrass (B. decumbens Stapf.) and other paddy weeds.3,21,22 However, allelopathic rice cultivars alone cannot overcome weed problems in paddy ecosystems.1,3,13 Under field conditions, allelopathy does not occur independently of competition and other mechanisms of plant interference, and thus the weed-suppressive effects of rice Pest Manag Sci 64:276–282 (2008) DOI: 10.1002/ps

Combination of cultural conditions and allelopathic rice varieties for control of paddy weeds Table 3. Inhibition of paddy weeds by rice varieties tested under integrated cultural management options in the fielda

Inhibition (%) (± SEM) Grass weeds (spring/autumn)

Broadleaf weeds (spring/autumn)

Rice yield (ton ha−1 ) (spring/autumn)

PI312777

96.1 (±2.6)/89.1 (±5.6)

61.0 (±8.3)/55.2 (±7.3)

3.41 (±0.43)/2.78 (±0.68)

Huagan-1 Huajianxian

95.4 (±3.3)/93.1 (±4.3) 42.4 (±2.8)/37.3 (±2.8)

64.1 (±5.5)/61.7 (±3.5) 38.0 (±3.3)/30.8 (±5.3)

5.14 (±0.23)/4.23 (±0.51) 3.23 (±0.18)/2.41 (±0.23)

PI312777

99.5 (±0.2)/96.3 (±1.2)

87.4 (±5.2)/88.6 (±5.9)

3.44 (±0.22)/3.23 (±0.41)

Huagan-1 Huajianxian

100 (±0.0)/98.6 (±1.1) 55.4 (±3.1)/57.3 (±4.3)

90.3 (±3.8)/87.3 (±5.6) 70.9 (±5.7)/73.9 (±4.5)

5.74 (±0.33)/5.66 (±0.28) 4.05 (±0.15)/4.17 (±0.21)

Treatment

Rice variety

Integrated cultural management optionsb without herbicide

Integrated cultural management options with herbicidec

a

Experiments were carried over two seasons, spring and autumn, in 2006. Rice was transplanted into the plots at a density of 4.0 × 105 plants ha−1 with 5 cm flooded depth and 10 days flooded duration, and then a level of 315 kg ha−1 N supply was applied at 15 days after transplanting. c As b and bensulfuron-methyl applied at 25 g AI ha−1 , a third of the recommended dose. b

varieties is at least a combined effect of allelopathy and competitive ability.3,6 This study indicated that two allelopathic rice varieties resulted in a total reduction of paddy weeds of 30–50%, which agrees with some previous studies claiming that allelopathy accounts for 34% of the overall competitive ability in rice, and 65% of the suppression of E. crus-galli by PI312777 may be attributable to allelopathy.13,23 Although the relevance of allelopathy and competition remains obscure in this study, it is a fact that the two allelopathic rice varieties tested can effectively reduce paddy weeds. In addition, weed-suppressive effects of rice varieties may be affected by cultural management and other factors in the field. The weed suppression of allelopathic rice varieties must be evaluated and optimized with cultural practices under variable environmental and cultural conditions.2,24 The data generated in this study are encouraging, as they show that allelopathic rice varieties are likely to have a great impact on paddy weed control when combined with integrated cultural management options. The general aim of allelopathic rice varieties is to lessen reliance on herbicides and reduce labour input in paddy ecosystems, and thus the effect of allelopathic rice varieties must be accompanied with the evaluation of economics of reduced herbicides and labour rates.2,3 Gealy et al.3 evaluated the economic consequence of applying less than the recommended rates of the herbicide propanil to several Asian rice cultivars. As a result, grain yields increased, and barnyardgrass biomass decreased with increasing propanil rates.3 The data generated in this study also showed that, by using allelopathic rice varieties together with the application of a low dose of herbicide (bensulfuronmethyl, 25 g AI ha−1 , a third of the recommended dose), the emergence and growth of most weeds found in paddy fields were completely controlled and no grain yield reduction occurred under integrated cultural management options. These studies clearly revealed that it is feasible to reduce herbicide input in paddies if allelopathic and/or weed-suppressive Pest Manag Sci 64:276–282 (2008) DOI: 10.1002/ps

rice varieties are grown under integrated cultural management practices. Paddy weed control using allelopathy could also be incorporated into other allelopathic plant mulches in the field.25 However, the weed-suppressive effects of these plant mulches are usually short lived, and weeds re-emerge and may cause considerable fieldwork labour input. On the other hand, rice and weeds always simultaneously grow in paddies, and thus the optimization of allelopathic rice varieties with various cultural techniques showing suppressive effects on the spontaneous growth of paddy weeds would substantially reduce herbicide use and labour input. Although no commercial rice cultivars carrying allelopathic properties have been developed so far,13,14 it is now more widely accepted that allelopathic rice cultivars with a natural weed control system may be able to provide a reduction in the use of other weed management options.2 This study expanded and merged the concepts of biological and chemical control of weeds by combining several cultural management options that integrated chemical-releasing allelopathic rice varieties into paddy management systems. Management options for paddy weed control are limited under diverse agroecological conditions.17,26 Allelopathic rice varieties combined with cultural management options therefore offer an interesting and promising alternative technique for chemical control of weeds in paddy ecosystems. Rice allelopathy as a naturally occurring defence mechanism of rice against paddy weeds would involve multiple factors, particularly in soil dynamics of allelochemicals, inducible processes and plant signalling.11,19,27 In addition, weed suppression by allelopathic rice varieties has shown considerable variations from year to year, which are attributed to agronomic practice and environmental conditions.2,24 An in-depth understanding of allelopathy in rice/paddy weed interactions can be developed into an allelopathy-based technique for sustainable weed management in paddy ecosystems. Further clarification of allelopathic interference of rice with 281

CH Kong et al.

paddy weeds, with the principal goal of making weed suppression more reliable, is warranted.

ACKNOWLEDGEMENTS The authors thank two anonymous reviewers for thoughtful criticisms and English corrections that substantially improved this manuscript. This work was financially supported by the National Natural Science Foundation of China (NSFC Grant No. 30430460) and the Eleventh Five-year Plan of Science and Tech Programme of China (2006BAD08A09).

REFERENCES 1 Kim KU and Shin DH, Rice Allelopathy. Kyungpook National University, Taegu, Korea (2000). 2 Belz RG, Allelopathy in crop/weed interactions – an update. Pest Manag Sci 63:308–326 (2007). 3 Gealy DR, Wailes EJ, Esterninos LEJ and Chavez RSC, Rice cultivar differences in suppression of barnyardgrass (Echinochloa crus-galli) and economics of reduced propanil rates. Weed Sci 51:601–609 (2003). 4 Moody K, Weed management in rice, in Handbook of Pest Management in Agriculture, 2nd edition. CRC Press, Boca Raton, FL (1991). 5 Dilday RH, Identification of allelopathy in USDA-ARS rice germplasm collection. Aust J Exp Agric 34:907–910 (1994). 6 Olofsdotter M, Navarez D, Rebulanan M and Streibig JC, Weed suppressing rice cultivars – does allelopathy play a role? Weed Res 39:441–454 (1999). 7 Kong CH, Xu XH, Hu F, Cheng XH, Lin B and Tan ZW, Using specific secondary metabolite as marker to evaluate allelopathic potential of rice variety and individual plant. Chinese Sci Bull 47:839–843 (2002). 8 Bastiaans L, Kropff MJ, Kempuchetty N, Rajan A and Migo TR, Can simulation models help design rice cultivars that are more competitive against weeds? Field Crop Res 51:101–111 (1997). 9 Duke SO, Belz RG, Bearson SR, Pan Z, Cook DD and Dayan FE, The potential for advances in crop allelopathy. Outlook Pest Manag 16:64–68 (2005). 10 Duke SO, Scheffler BE, Dayan FE, Weston LA and Ota E, Strategies for using transgenes to produce allelopathic crops. Weed Technol 15:826–834 (2001). 11 Macias FA, Molinillo JMG, Varela RM and Galindo JCG, Allelopathy – a natural alternative for weed control. Pest Manag Sci 63:327–348 (2007). 12 Singh HP, Batish DR and Kohli RK, Allelopathic interactions and allelochemicals: new possibilities for sustainable weed management. Crit Rev Plant Sci 22:239–311 (2003).

282

13 Olofsdotter M, Jensen LB and Courtois B, Improving crop competitive ability using allelopathy – an example from rice. Plant Breeding 121:1–9 (2002). 14 Duke SO, Weeding with allelochemicals and allelopathy – a commentary. Pest Manag Sci 63:307 (2007). 15 Shibayama H, Weeds and weed management in rice production in Japan. Weed Biol Manag 1:53–60 (2001). 16 Minorsky PV, Allelopathy and grain crop production. Plant Physiol 130:1745–1746 (2002). 17 Liebman M and Davis AS, Integration of soil, crop and weed management in low-external-input farming systems. Weed Res 40:27–47 (2000). 18 Charudattan R, Biological control of weeds by means of plant pathogens, significance for integrated weed management in modern agroecology. Biol Control 46:229–260 (2001). 19 Kong CH, Li HB, Hu F, Xu XH and Wang P, Allelochemicals released by rice roots and residues in soil. Plant Soil 288:46–57 (2006). 20 Kong CH, Laing WJ, Xu XH, Hu F, Wang P and Jiang Y, Release and activity of allelochemicals from allelopathic rice seedlings. J Agric Food Chem 52:2861–2865 (2004). 21 Gibson KD, Hill JE, Foin TC, Caton BP and Fischer AJ, Water-seeded rice cultivars differ in ability to interfere with watergrass. Agron J 93:181–190 (2001). 22 Fischer AJ, Ramirez HV, Gibson KD and Pinheiro BdaS, Competitiveness of semidwarf upland rice cultivars against palisadegrass (Brachiaria brizantha) and signalgrass (B. decumbens). Agron J 93:967–973 (2001). 23 Xiong J, Lin WX, Zhou JJ, Wu MH, Chen XX and He HQ, Studies on biointerference between barnyardgrass and rice accessions at different nitrogen regimes, in Proceedings of Fourth World Congress on Allelopathy, Charles Sturt University, Wagga Wagga, NSW, Australia, 21–26 August 2005, ed. by Harper JDI, An M, Wu H and Kent JH. International Allelopathy Society, Puerto Real, Spain, pp. 501–504 (2005). 24 Gealy DR, Dilday RH and Rutger JN, Interaction of flush irrigation timing and suppression of barnyardgrass with potentially allelopathic rice lines, in B.R. Wells Rice Research Studies 1997, ed. by Norman RJ and Johnston TH. Arkansas Agricultural Experiment Station, Division of Agriculture, University of Arkansas, Research Series 460, pp. 49–55 (1998). 25 Khanh TD, Hong NH, Xuan TD and Chung IM, Paddy weed control by medicinal and leguminous plants from Southeast Asia. Crop Prot 24:421–431 (2005). 26 Heraux FMG, Hallett SG and Weller SC, Combining Trichoderma virens-inoculated compost and a rye cover crop for weed control in transplanted vegetables. Biol Control 34:21–26 (2005). 27 Kong CH, Zhao H, Xu XH, Wang P and Gu Y, Activity and allelopathy of soil of flavone O–glycosides allelochemicals from rice. J Agric Food Chem 55:6007–6012 (2007).

Pest Manag Sci 64:276–282 (2008) DOI: 10.1002/ps