A field experiment was conducted from October 1999 to September 2001 in order to establish the annual rotation of forage production with Italian ryegrass (Lolium multiflorum) and barnyard grass (Echinochloa crus-galli), and to monitor the vegetation change after starting the annual rotation in an abandoned paddy field. Ryegrass and barnyardgrass were sown simultaneously on October 15, 1999. Two ryegrass cultivars, early-maturing Tachiwase (hereafter referred to simply as EMR) and mid-late maturing Mammoth B (hereafter referred to simply as LMR), were sown at a rate of 300g/a in a main plot. The main plot was divided into three subplots, and baryardgrass was sown at a rate of 100, 200 and 3008/a in each subplot. EMR was harvested on May 1 and June 6 and LMR, on May 25 the following year. Barnyardgrass was harvested August 9 at the beginning of the heading stage and on October 3 after seed setting both in EMR and LMR plots. The experiment was conducted according to a split-split plot design with ryegrass cultivars as the main plot factor and sowing rate of barnyardgrass as the subplot factor with three replications. On October 20, 2000, the same ryegrass cultivars were sown on the same plots used in the experiment the previous year. Barnyardgrass was not sown since the species was expected to regenerate through natural reseeding the following year. Ryegrass yields significantly depended on the cultivars in both years, with greater yield of EMR than of LMR. The first crop yields of barnyardgrass were affected by the preceding ryegrass cultivars in both years, but the sowing rate exhibited no effect. No significant interactions were seen regarding ryegrass cultivars×sowing rate in the first crop yield of barnyardgrass. In the second crop of barnyardgrass, no significant differences were observed between the preceding ryegrass cultivars or among the sowing rates of barnyardgrass. There were significant differences in the seed production of barnyardgrass between the ryegrass cultivars, with greater production in EMR plots than in LMR plots in both years. However, a sufficient number of seeds were produced to establish the barnyardgrass sward the following year even in LMR plots. Gramineae weeds from seed banks were contained in the barnyardgrass yields in 2000, but they did not seem to have a detrimental effect on the crop quality since almost all weeds were useful as forage. The occurrence of weeds decreased the following year except for Cyperus serotinus. Damage by Cyperus weeds could be a restricting factor for this annual crop rotation since the species are specific to wet and semi-wet soils, including paddy fields, and are thought to be recalcitrant species in Japan.

View full abstract

The growth of Echinochloa crus-galli (L.) Beauv. var. formosensis Ohwi was investigated for one- and two-cut whole-crop silage rice cultivation using different rice cultivars in 2001 and 2002. Tetep and Taporuri canopies were always higher than that of E. crus-galli var. formosensis. Thus, the relative photosynthetic photon flux density in rice canopies, calculated as the relative value between photosynthetic photon flux density in the rice canopy and that of the paddy levee surface, quickly decreased during Tetep and Taporuri growth or regrowth periods.
There were positive relationships between E. crus-galli var. formosensis shoot dry weight and the relative photosynthetic photon flux density below the canopy at 20cm from the ground at 40 days after transplanting rice or the number of days after transplanting rice when relative photosynthetic photon flux densities were less than 20%. The dry weights of E. crus-galli var. formosensis planted with Tetep and Taporuri were less than those planted with other cultivars in one-cut rice cultivation. In addition, the weed losses of the cultivars themselves were smaller than for other cultivars. The dry weights of E. crus-galli var. formosensis and rice in two-cut rice cultivation were similar to those in one-cut rice cultivation, and those of E. crus-galli var. formosensis were less in two-cut rice cultivation than in one-cut rice cultivation.
E. crus-galli var. formosensis contaminated more than 40% of the Hinohikari whole-crop silage, but less than 10% contamination was found in Tetep or Taporuri whole-crop silage.

View full abstract