Introduction
From the lowest record of 160 to 200 micromol/mol in the Last Glacial Maximum (about 18,000 years ago) (Delmas et al., 1980; Neftel et al., 1985; Keeling et al., 1995), atmospheric CO2 concentration has been increasing and has become about 360 micromol/mol recently. It is predicted that the increase will continue and the CO2 concentration will reach more than 450 micromol/mol within next century (IPCC, 1996). C3 plants are supposed to increase their biomass with this increment of CO2 concentration, whereas C4 plants are not. Here I show experimental results indicating that the increase of CO2 concentration from glacial age to the future had been and will have been altering competitive relationships between C3 and C4 plants by changing morphology and growth rates of C3 plants.

Results and Discussion
Relative growth rate (RGR), a parameter represents rate of dry weight increment, increased with increasing CO2 concentrations both C3 and C4 plants. Compared to the RGR in the control (M), the reduction of RGR in the C4 plants by Low CO2 treatment (L) was much lower than that in the C3 plant. The high CO2 treatment (H) did not enhance RGR significantly in the C3 nor the C4 plants. The growth response of the C3 plant to CO2 concentration ([CO2]) is not agree with the general belief that increasing [CO2] should enhance growth of C3 plants. The reason of this uncommon result can be clarified later in this paper. C4 plants are expected to have low ability to increase their RGR at higher [CO2], because photosynthesis of C4 plants is saturated with low [CO2] (lower than 200 micromol/mol intercellular concentration in leaves). The present result is consistent with this expectation and suggests that global [CO2] change has little effect on growth of C4 plants.
Net assimilation rate (NAR), a representative parameter of whole plant net photosynthetic rate per unit leaf area, of the C3 plant increased significantly with increasing [CO2], but that of the C4 plants was not affected by [CO2] . These results also support the general expectation that enhancement of photosynthetic activity by global [CO2] increase is much greater in C3 plants than C4 plants.
Thus, global [CO2] increase is expected to increase growth and photosynthetic activity of C3 plants. Our results imply that there was an morphological change of C3 plants with changing global [CO2], and this change should be extended towards the future high CO2 world. Leaf area ratio (LAR), a morphological parameter showing the leaf area per unit plant weight, in the C3 plant decreased with increasing [CO2], whereas that in the C4 plants showed no change. This reduction of LAR in the C3 plant in H treatment canceled out the increase of NAR, and thus RGR of the C3 plant was not enhanced by the H treatment. Many reports suggest that, in C3 plants, enhancement of RGR by high [CO2] is caused mainly by increase of NAR.
The current results shown here support the general hypothesis that the growth and morphology of C4 plants are not responsive for the [CO2] from today to the future, though the growth of the C4 plants was reduced by the L treatment, which is comparable to [CO2] in the Last Glacial Maximum. There is another general hypothesis that C3 plant growth is more responsive to [CO2] than C4 plant growth. Our results suggest that this hypothesis is not always true in the future high [CO2] world because morphology of some C3 plants may be altered by the high [CO2] and this morphological change may cancel the increase of photosynthetic activity. These findings are consistent with the previous results of low [CO2] expeiments (Polley et al., 1993; Dippery et al., 1995), though the number of the experiments is quite limited yet.
It is quite likely that competitive interactions of C3 and C4 plants have been changed through geologic time to the future (Dippery et al., 1995) . However, this change is not so simple as being thought before. In the low [CO2] world in the last glacial age, it can be supposed that some C3 plants had RGR comparable to some C4 plants by oompensating reduction of NAR with increasing amount of leaves. It can be also supposed that, in some C4 plant, reduction of RGR was caused by the reduction of NAR in the low [CO2]. In this case, the result of competition between C3 and C4 plants cannot be esimated srtaightly. As shading by leaves is one of the critical factors determining result of competition for light (Tremmel and Bazzaz, 1995), this higher amount of leaves in the C3 plant may be advantageous in the low [CO2] world. In the future high [CO2] world, the competitive interactions of C3 versus C3 and C3 versus C4 plants may be weakened because some C3 plant reduces their shading ability by reduction of their amount of leaves. Thus, I have to enphasis that the alternation of competitive interactions between C3 and C4 plants through geologic time to the future must depend on degree of morphological change in C3 plants as well as that of changes in photosynthetic activities in both C3 and C4 plants.