The differences of productivity among the three crops were analyzed in terms of radiation absorption and its efficiency for dry matter accumulation, using the data obtained in the dry matter production-population density experiments reported previously
12, 13, 14). In this experiment, photosynthetically-active radiation (PAR) intercepted by plant canopy (ΔPAR) during the experimental period (t
2-t
1) was calculated as ΔPAR=Σ^^(<SUB)2>__(i=t<SUB)1>0.444·S
i(1-exp
Ks.LAIi) where 0.444 is the proportion of full spectrum radiation in the range 400 to 700 nm; S
i is daily solar radiation; Ks is light-interception coefficient; LAI
i is daily value of leaf area index calculated assuming that LAI increased exponentially from t
1 to t
2. Light-interception coefficient was given by I/I
0=exp
-Ks.LAI where I
0 and I are the light intensities at the top and bottom of canopy, respectively, and the Ks value of each variety was estimated from the data measured three times for five population densities during the middle period of the growing season. The efficiency of dry matter accumulation per unit PAR intercepted (E
PAR, dry weight mg/kcal) during the experimental and/or full growing period was defined as E
PAR=ΔW/ΔPAR where ΔW is dry matter production and ΔPAR is the amount of PAR intercepted. E
PAR also could be described as E
PAR=NAR/(ΔPAR/(LAI)^^
-·Δt) [(LAI)^^
-=LAI
2-LAI
1/ln (LAI
2/LAI
1)] where NAR is net assimilation rate; LAI is mean leaf area index; Δt is number of days between t
1 and t
2; LAI
1 and LAI
2 are leaf area indices at times t
1 and t
2, respectively. The main results obtained are summarized as follows: 1. The maximum dry matter production for azuki bean and kidney bean was about 50-60%, on average, of that for soybean (Table 1). The maximum values for CGR, LAI and E
PAR during growing season were higher in soybean (Table 2), indicating highly positive correlations with the maximum dry matter production (r≥0.898**). 2. When plotted disregarding crops, varieties and densities, dry matter production hag significantly positive correlations with ΔPAR and E
PAR, while it was related positively only with LAD, and negatively with NAR. The regression of dry matter production on ΔPAR differed between two groups, one group's values of E
PAR were lower and the other higher than about 9 mg/kcal (Fig. 1). In addition, a simple correlation coefficient between ΔPAR and E
PAR was not significant, but a partial correlation between them was highly negative (Table 3). 3. The differences in total and pod+seed dry weight produced during grain filling period were in the same order as that exhibited in thc maximum dry matter production (Table 4). During grain filling both total and pod+seed dry matter production increased curvilinearly with increase of ΔPAR, but linearly with E
PAR. Takarashozu differed from this relation because it did not reach the LAI value required for 95% light interception except under the highest density (Fig. 2). 4. The very close regressions of CGR on E
PAR were found among the stages with different mean solar radiation, disregarding crops, varieties and densities, for which LAIs are more than the value required for 90% light interception (Fig. 3 and Table 6). A close correlation was also found between CGRmax. calculated by WASTON's method and E
PAR (density mean) (Fig. 5). 5. [the rest omitted]
View full abstract