J. WeedSci. Tech. Vol. 42(3)289-293(1997) 雑 草 研 究 Short Report Weed Growth Suppression by Cogongrass (Imperata cylindrica) *

Tohru Tominaga** and Osamu Watanabe

Key words: allelopathy, Imperata cylindrica, agroecosystems6. In another study of ours, weed growth suppression. different weed species showed different reac- キ ー ワ-ド: ア レ ロパ シ ー, チ ガ ヤ, 雑 草 の 生 育 抑 tion to water extract from buckwheat culti- 制 vated soils. There seems to be a species specific relationship between donor and re- Imperata cylindrica (L.) Beauv. is a serious ceptor species. Allelopathy appears to have perennial grass weed in Southeast Asia ans is potential for selective weed control in the very difficult to control. On the other hand, future and this would lead to decreased herbi- this species is sometimes used as a mulching cide use in agroecosystems and be helpful for material3? and is empirically known to sup- sustainable agriculture. From this viewpoint, press weed growth. It makes dense stands clarification of the allelopathic effect of this and keeps the biomass of other species in the species not only on crops but also on weeds is stand lows. This suppression seems to be valuable. caused by the allelopathic effect of I. cylin- This study was conducted to determine the drica. effect of leaves of I. cylindrica mixed with Eussen and Soerjani2 reported that the soil on weed emergence and growth. water extract from leaves of this species Materials and Methods inhibited the root growth of corn, sorghum, rice and , and Eussen and Niemannl Mixture of leaves with soil isolated growth inhibiting substances from The leaves of 1. cylindrica were harvested the leaves and identified them as p- and o- on December 13, 1991 and stored at below coumaric acid, gentisic acid, vanillic acid, 5C until start of the experiment. On April 6, benzonic acid, p-hydroxybenzonic acid, vanil- 1992, the top 10 cm of soil from the vegetable lin and p-hydroxybenzaldehyde. garden of the Research Farm, Faculty of Allelopathy is believed to be important in Agriculture, Shinshu University was sampled and passed through an 8 mm mesh sieve. The * A part of this work was presented at the 32nd leaves were cut into 5 cm length pieces and Annual Meeting of the Weed Science Society of well mixed with the soil at the ratios of 1.6 Japan (1993). **Research Farm, Faculty of Agriculture, Kyoto and 3. 2 g per 100 g soil (d. w. /d. w.), which is Prefectural University the estimated ratio applied to fields in Kii- Kitainayazuma, Seika, Kyoto 619-02, Japan Oshima, Wakayama. The soil was put into 32 ***Faculty of Agriculture, Shinshu University cm width x 25 cm depth x 7 cm height ma, Nagano 399-45, Japan (Accepted November 15, 1996) plastic trays, which were kept in a green- 冨永 達**・渡辺 修軸***:チ ガヤの葉による雑草の house to avoid surface erosion by heavy rain- 生育抑制. **京都府立大学農学部附属農場 fall. Both sides of the greenhouse were left ***信州大学農学部 open throughout the study and the tempera- 290 J. Weed Sci. Tech. Vol. 42 (1997) ture was not controlled. Watering was done pressed as Erigeron sp. in this study. The as needed. Trays holding only soil from the predominant species were E. canadensis L., vegetable garden were prepared as a control Erigeron sp. and oleracea L. ; this and kept in the same way as those with the result agreed well with the weed species treated soil. composition of our vegetable garden. Effect on weed emergence There were no significant differences in the Weed seedlings emerging in each tray were total number of emerged seedlings among identified by species, counted and removed treatments (Fig. 1), but the number of P. once a week from April 13, 1992 to April 8, oleracea in April and May was 27.0% fewer in 1993. Three replications were made of each the 1. 6% plot and 19.3% fewer in the 3. 2 treatment and results were compiled. plot compared with those of control, respec- Effect on weed growth tively (Fig. 2-a). I. cylindrica leaves seemed The emerged seedlings were allowed to to suppress P. oleracea emergence. grow and the height of the five tallest There were two conspicuous emergence of each weed in each tray were measured on peaks in spring and autumn (Fig. 1). These June 8, 1992. Then, all weeds were harvested, peaks represented the emergence peak of dried at 80 C for 48 hours and weighed. summer and winter annual weeds, respective- Three replications were made of each treat- ly. Temporary decreases in early May and ment and results were compiled. early October may have been caused by the temporary drop in temperature during these Results and Discussion periods. Effect on weed emergence The emergence pattern differed among A total of 1,383 seedlings of 16 species in treatments. In the 1. 6% plot, the obvious control, 1,367 seedlings of 17 species in the 1. peak in autumn appeared in early September, 6% treatment plot and 1, 160 seedlings of 14 but in control the peak appeared in mid- species in the 3. 2% treatment plot emerged October. This was mainly due to the differ- during the study (Fig. 1). As it was very ence in the emergence pattern of E. canaden- difficult to distinguish the seedlings of sis and Erigeron sp. (Fig. 2-b, c). The seed- Erigemon phi ladelphicus L. from those of ling emergence peak of Erigeron sp. in the 1. Stenactis annuus (L.) Pers., they were ex- 6% plot appeared about one month earlier than that in control, and the number of emer-

-●- Imperata 0% ged seedlings in September to November was -□- Imperata 1.6% 43.8% fewer in the 3. 2% plot than that in --▲-- Imperata 3.2% control. The reason for this acceleration of emergence is not clear. On the other hand, neither emergence patterns nor number of emerged seedlings of Amaranthus lividus L. var. ascendens (Lous.) Thell., Stellaria media (L.) Villars and (Retz.) Koel, was affected by the treatment. Eussen and Soerjani2) reported the effect of water extract of leaves of I cylindrica on Fig. 1. Emergence pattern of weed seedlings. crop germination. The total germination Tominaga and Watanabe : Weed Growth Suppression by Imperata 291

ratios of corn, sorghum, rice, cucumber and a) Portulaca oleracea tomato were not affected, but the germina- -●- Imperata 0% -□- Imperata 1.9% tion of tomato was delayed. The effect on --▲-- Imperata 3.2% germination is assumed to be different among receptor species. Effect on weed growth The predominant species at the harvest in June were A. lividus var. ascendens, S. media, P. oleracea and D. ciliaris. The effect of the treatment on height of weeds is shown in Fig. 3. Heights of A. lividus var. ascendens and S. media were 13.2 and 23. 3% shorter in the 1.6% plot, and 24.0 and 21.9% shorter in the 3. 2% plot as compared with those in control, respectively. There was no influence on the height of D. ciliaris., The shoot growth b) Erigeyon canadensis was slightly inhibited in the seedlings of corn, -●- Imperata 0% -□- Imperata 1.6% sorghum and rice, and stimulated in tomato --▲-- Imperata 3.2% by the water extract of the leaves of I. cylindrica2'. Different receptor species seem to respond in different ways. With the application of leaves of I. cylig- drica, the total biomass of weeds was sig- nificantly decreased to 38.8% in the 1. 6% plot and 27. 1% in the 3.2% plot as compared with control, respectively (Fig. 4). This decrease in total biomass was largely due to the lesser amount of biomass of A, lividus var. ascen- ders and S. media; these were severely reduced to 31. 3 and 24. 2% in the 1.6% plot, c) Erigeron sp. -●- Imperata 0% and 22. 5 and 21.3% in the 3. 2% plot, respec- -□- Imperata 1.6% tively, although there was no inhibiting effect --▲-- Imperata 3.2% on P. oleracea or D. ciliaris. The effect of I. cylindrica apparently differs among receptor species. The application of I. cylindrica leaves had no effect on the number of emerged weeds except for P. oleracea, but the total weed biomass was clearly suppressed. This result was generally consistent with what is empiri- cally known. Hirota et al. 4' reported that the Fig. 2. Emergence pattern of the seedlings of a) Portulaca oleraceda, b) Erigeron canadensis water extract of I. cylindrica leaves suppress- and c) Erigeron sp. ed lettuce seedling growth and identified the 292 J. Weed Sci. Tech. Vol. 42 (1997)

Amaranthus lividus Stellaria media Digitaria ciliaris

Fig. 3. Effect of Imperata cylindrica leaves on the plant height of three weeds. Vertical bars indicate L. S. D. at 5% level.

■ Amaranthus lividus ■ Portulaca oleracea References □ Stellaria media ■ Digitaria ciliaris 1) Eussen, J. H. H. and G. J. Niemann 1981. Growth ■ other species inhibiting substances from leaves of Imperata cylindrica (L.) Beauv. Z. Pflanzenphysiol. 102, 263-266. 2) Eussen, J. H. H. and M. Soerjani 1978. Al- lelopathic activity of alang-alang (Imperata cylindrica (L.) Beauv.), isolation of growth regulating substances from leaves. In "Studies on the Tropical Weed, Imperata cylindrica (L.) Beauv. var. major" ed. by J. H. H. Eussen, Drukkerrij Elinkwijk BV, Utrecht, pp. 1-10. 3) Harada, J., S. Zungsontiporn and K. Suzuki Fig. 4. Effect of Imperata cylindrica leaves on weed 1990. Some examples of the traditional utiliza- biomass. tion of plants for weed control. Weed Res., Vertical bars indicate+S. E. Japan 35 (Suppl.), 81-82 (in Japanese). 4) Hirota, M., K. Yoshida, K. Maekawa, T. inhibiting substances as p- hydroxybenzoic Tominaga and R. Irie 1995. Growth inhibiting acid and vanillic acid. Matsumoto et al.5), on substances from Imperata cylindrica. Biochem. the other hand, reported that the application Biotech. Agrochem. 69 (Suppl.), 235(in of powder of Anthoxanthum odoratum L. Japanese). severely suppressed the germination and root 5) Matsumoto, I., F. Sekiguchi, M. Nemoto and Y. growth of Trifolium repens L., but that the Fujii 1992. Control of white clover by allelopaty influence of I. cylindrica was less. The mecha- of some grasses. Weed Res., Japan 37 (Suppl.), 166-167 (in Japanese). nism of weed biomass suppression by I. cylin- 6) Putnam, A. R. 1978. Allelopathy in agroecosys- drica leaves is not clear and tests applying tems. Ann. Rev. Phytopathol. 16, 431-451. extracts of these leaves to various weed 7) Tominaga, T., H. Kobayashi and K. Ueki 1989. species are needed in the future. Tominaga and Watanabe : Weed Growth Suppression by Imperata 293

Seasonal change in the standing-crop of Imper- 8) Tominaga, T. and T. Uezu 1995. Weed suppres- ata cylindrica var. koenigii grassland in the Kui sion by buckwheat. Proc. 6th Intern. Symp. -Ohshima Island of Japan. Weed Res., Japan 34 Buckwheat, 693-697. (3), 204-209.