Effects of Soil Copper Concentration on Growth, Development and Yield Formation of Rice (Oryza sativa)

Rice Science, 05, 12(2): 125-132 125 http://www.ricescience.org Effects of Soil Copper Concentration on Growth, Development and Yield Formation of Rice (Oryza sativa) XU Jia-kuan 1, 2, YANG Lian-xin 1, WANG Zi-qiang 1, DONG Gui-chun 1, HUANG Jian-ye 1, WANG Yu-long 1 ( 1 Key Laboratory of Crop Genetics and Physiology of Jiangsu Province, Yangzhou University, Yangzhou 225009, China; 2 Agricultural and Forestry Bureau of Changzhou, Changzhou 213001, China) Abstract: Pot experiments were conducted in 02 and 03 to investigate the effects of soil copper(cu) concentration on growth, development and yield formation of rice by using the japonica cultivar Wuxiangjing 14 and hybrid rice combination Shanyou 63. The plant height, leaf number, elongated internode number and heading date of rice plants were not affected at soil Cu levels below 0 mg/kg, but affected significantly at above 0 mg/kg. The inhibitory effects on rice growth and development were increased with the increment of soil Cu levels. The grain yields decreased significantly with raising soil Cu levels. The main reasons for the grain yield reductions under lower soil Cu levels (, 0 mg/kg) were mainly due to the decrease of number of spikelets per panicle, however, under higher soil Cu levels (more than 0 mg/kg), both panicle number and number of spikelets per panicle contributed to the yield loss. The decreases of panicle number by Cu stress were mainly attributed to slow recovery from transplanting, delayed tillering and reduced maximum tiller numbers. The reduction of number of spikelets per panicle under soil Cu stress resulted from the decreases of both shoot dry weight (SDW) at the heading date and the ratio of spikelets to SDW. Total biomass at maturity decreased significantly with the increase of soil Cu levels, while economic coefficient showed non-significant decrease except under soil Cu levels above 0 mg/kg. Key words: rice; soil Cu concentration; growth and development; yield Copper (Cu) level is one of the eight heavy metal manipulative indices in agricultural soil environment evaluation in China [1]. The second and third soil quality standard in Cu concentrations were, 0 mg/kg (ph 6.5 7.5), respectively. Owing to quick development of modern industry and the widespread use of pesticides containing Cu additives, the concentration of Cu in agricultural soils around the world increased rapidly, and at some sites average soil Cu levels have already exceeded 0 mg/kg [2-4]. Several studies have been reported about the effects of soil Cu stress on rice growth and development [5-13]. However, the mechanism of grain yield reduction by soil Cu stress remained unclear. Soil Cu levels designed in many previous experiments were low, generally less than 250 mg/kg, far below the third quality criterion of soil environment in China (0 mg/kg). To find out the reasons of grain yield loss under soil Cu stress, we conducted two pot experiments to investigate the effects of soil Cu Received: 17 January 05; Accepted: 5 April 05 Corresponding author: WANG Yu-long (ylwang@yzu.edu.cn) concentration between to 0 mg/kg on rice growth and development and grain yield formation and to provide a basis for establishment of rice cultivation system in Cu contaminated areas. MATERIALS AND METHODS Plant materials and Cu treatments Experiment 1 The pot experiment was conducted at Agricultural and Forestry Bureau of Changzhou (30 41' N, 119 o 50' E) in 02. The test rice cultivar was Wuxiangjing 14, an early-maturity late-season japonica cultivar. The soil for pot experiment was a silty loam with the alkali-hydrolysable N, P 2 O 5, K 2 O and total Cu contents.5, 10.2, 82.1 and 37.4 mg/kg, respectively. Ten-kilogram of soil was placed in each pot (26 cm in diameter and 28 cm in height). CuCl 2 3H 2 O was added to the soil to obtain a series of soil Cu levels of, 150, 300, 500 and 0 mg/kg, with replicates. The thoroughly mixed soil

126 Rice Science, Vol. 12, No. 2, 05 was submerged in water for one month prior to the transplanting of seedlings. Seeds were sown in uncontaminated paddy field on 22 May, and seedlings were hand-transplanted into the pots (3 plants per pot) on 22 June. Fertilizers were applied as a basal dressing on 21 June (2.0 g N, 1.2 g P 2 O 5, 2.0 g K 2 O per pot), and at panicle initiation (PI) on 31 July (1.0 g N per pot). The pots were flooded with a water layer of 4 cm during the whole growth period. Experiment 2 The pot experiment was conducted at Yangzhou University (32 30 N, 119 25 E) in 03. The test rice cultivars were Wuxiangjing 14 and Shanyou 63, a mid-maturity indica hybrid combination. The soil for the pot experiments was a loam, with alkali-hydrolysable N, P 2 O 5, K 2 O and total Cu contents 110.5, 16.4, 96.5 and 75.4 mg/kg, respectively. Ten kilogram of soil was placed in each pot (25 cm in diameter and 30 cm in height). CuCl 2 3H 2 O as Cu source was added to the soil to establish a series of soil Cu levels of, 0, 0, 0, 0 and 0 mg/kg, with replicates. The thoroughly mixed soil was submerged in water for one month before rice seedlings were transplanted in. Seeds were sown in uncontaminated paddy field on 15 May, and seedlings were hand-transplanted into the pots (3 plants per pot) on 8 June. Fertilizers were applied as a basal dressing on 7 June (1.0 g N, 1.0 g P 2 O 5, 1.0 g K 2 O per pot), and at PI on 5 August (0.5 g N per pot). Water management was the same as Experiment 1. Samples preparation and analysis methods Leaf number on the main stem, tiller number and plant height were investigated every 8 d after transplanting. Heading date for each treatment was recorded when 50% of plants headed. Four pots for each treatment (two replicates) were sampled at maturity to determine the dry matter of root, leaf blade, shoot (including leaf sheath) and panicle, as well as the elongated internode number on the main stem. The samples harvested at maturity were also measured for yield components, i.e. panicle number, number of spikelets per panicle, 0-grain weight, and filled grain percentage. Fertile spikelets were selected by water floating and counted by manual. The 0-grain weight of fertile spikelets was determined after drying at in oven for 72 h. RESULTS Effect of Cu on growth and development of rice Effect of Cu on plant height Compared to the control, the plant height at maturity was decreased by 4.1, 6.6,.6, 27.4, 37.2 and 48.4% at soil Cu levels of, 0, 0, 0, 0 and 0 mg/kg, respectively (Table 1). The plant height was decreased significantly with the increase of soil Cu levels (r=-0.993**). Multiple comparisons showed that only the Cu concentrations higher than 0 mg/kg resulted in a reduction in plant height in comparison with the control. As showed in Fig. 1, highly significant differences existed in plant height at different growth stages between the control and the treatments of higher than 0 mg/kg. Such differences became larger with plant development, and then became smaller after the maximum differences appeared at 30, 42, 42 and 62 days after transplanting (DAT) for 0, 0, 0 and 0 mg/kg treatment, respectively. The differences were also enlarged with increasing soil Cu levels. The above results indicated that the inhibitory influence of soil Cu treatments on plant height increased with plant development, but decreased as soil Cu concentrations. And the time for peak Table 1. Effect of Cu on growth and development of rice cultivar Wuxiangjing 14 in 03. Character Soil Cu treatment (mg/kg) 75.4 (CK) 0 0 0 0 0 Plant height at maturity (cm) 101.8 97.6 95.1.8 73.9 63.9 52.5 Leaf number on the main stem 17.0 16.8 16.6 16.0 15.0 13.8 13.0 Elongated internode number on the main stem 6.0 6.0 6.0 5.5 5.0 4.0 4.0 Heading date (month-day) 08-23 08-23 08-24 09-03 09-07 09-11 09-16

XU Jia-kuan, et al. Effects of Soil Copper Concentrations on Growth and Development and Yield Formation of Rice 127 appearance was delayed with the increase of soil Cu levels. The result also suggested that plant height reduction became greater as soil Cu level increased. Effect of Cu on leaf number on the main stem Compared with the control, the leaf number on the main stem were reduced by 1.2, 2.4, 5.9, 11.8, 18.8 and 23.5% at soil Cu levels of, 0, 0, 0, 0 and 0 mg/kg, respectively (Fig. 1). The leaf number on the main stem decreased significantly with the increase levels of soil Cu (r=-0.975 ** ). The multiple comparisons showed that there were significant decreases (P < 0.01 or P < 0.05) at soil Cu levels of 0-0 mg/kg. The results indicated that the leaf number on the main stem was significantly influenced by soil Cu stress, and reduced with the increased levels of soil Cu. As indicated in Fig. 1, there was no significant reduction in leaf number on the main stem for the treatments( and 0 mg/kg Cu) in comparison with the control. While the treatments(higher than 0 mg/kg Cu) showed a highly significant reduction. For a specific soil Cu level, its effect on leaf number on the main stem increased with plant development, and then decreased after reaching a peak; As the Cu concentrations increased, the plants tended to have less leaves on the main stem. Effects of Cu on elongated internode number on the main stem Compared with the control, the same elongated internode number on main stem at maturity was observed at soil Cu level of or 0 mg/kg, but the number of elongated internodes decreased by 0.5, 1.0, 2.0, 2.0 unit internode at soil Cu levels of 0, 0, 0 and 0 mg/kg, respectively (Fig.1). The multiple comparisons showed that the elongated internode number on the main stem significantly decreased at 0 0 mg/kg Cu levels. The results indicated that elongated internode number on the main stem was not significantly decreased under lower soil Cu levels, but significantly reduced at Cu levels higher than 0 mg/kg. The elongated internode number declined gradually with increased supply of Cu. Effects of Cu on heading date The heading dates of the plants exposed to the soil Cu levels of 0, 0, 0, 0 and 0 mg/kg were delayed by 1, 11, 15, 19 and 24 d, respectively, whereas the heading date at soil Cu level of mg/kg was similar with the control (Fig. 1). The heading date at soil Cu level of or 0 mg/kg showed no difference with the control, but was significantly delayed at soil Cu levels of 0, 0, 0 and 0 mg/kg (P < 0.01 or P < 0.05), indicating that the heading dates were not influenced under lower soil Cu levels, but severely delayed at higher soil Cu levels (above 0 mg/kg). Effect of Cu on grain yield and its components Effect of Cu on rice grain yield Considering that consistent results were obtained from 02 and 03 experiments, we only analyzed Percentage of CK (%) 110 90 70 50 30 10 0 A Plant height CK mg/kg Cu 0 mg/kg Cu 0 mg/kg Cu 0 mg/kg Cu 0 mg/kg Cu 0 mg/kg Cu 0 110 90 70 50 B Leaf age 0 Days after transplanting (d) Days after transplanting (d) Fig. 1. Effect of Cu on dynamics of plant height and leaf number on the main stem of Wuxiangjing in 03.

128 Rice Science, Vol. 12, No. 2, 05 Table 2. Effect of Cu on grain yield and its components of rice cultivar Wuxiangjing 14 in 03. Soil Cu treatments (mg/kg) No. of Panicles per pot No. of spikelets per panicle No. of filled grain percentage (%) 0-grain weight (g) Grain yield per pot (g) 75.4(CK) 21.6 A 162.5 A 91.4 A 23.6 A 74.9 A 22.0 A 142.5 BC 89.0 A 24.1 A 67.3 B 0 21.5 A 137.6 CD 88.2 A 24.0 A 63.4 B 0 16.0 B 118.8 D 89.2 A 22.8 B 47.2 C 0 6.0 C 83.3 E 92.3 A 22.9 B 12.8 D 0 4.7 CD 81.0 E 94.6 A 22.7 B 8.0 D 0 2.0 D 79.5 E 85.5 A.6 B 2.8 F In each column, the data followed by the same letters are not different significantly at 0.05 leve1. the experiment with Wuxiangjing 14 in 03. Table 2 indicated that the yields were decreased by 10.1, 15.4, 37.0, 83.9, 89.3 and 96.2% at soil Cu levels of, 0, 0, 0, 0 and 0 mg/kg, respectively. The toxic effect on rice yield significantly increased with increasing levels of Cu application. The results of ANOVA indicated that the differences among different soil Cu treatments reached significant level for yields (F=2.89 ** ). The multiple comparisons showed that the yield at any soil Cu treatments was consistently and significantly lower in comparison with the control; No observed yield difference was detected between soil Cu treatments of and 0 mg/kg, while both of them were significantly higher than those under 0, 0, 0, 0 mg/kg Cu. The yield under soil Cu level of 0 mg/kg was significantly higher than those under soil Cu levels of 0, 0, 0 mg/kg. Treatment exposed to 0 mg/kg Cu showed a significant reduction of yield in comparison with the treatments (0 and 0 mg/kg), while no significant difference existed between the treatments of 0 and 0 mg/kg (Table 2). The above results indicated that grain yields were significantly influenced by Cu even at level of mg/kg. The yield decreased sharply with the increasing doses of Cu. Effect of Cu on yield components Effect of Cu on panicle number ANOVA test indicated that there were significant differences in panicle number among different soil Cu treatments in 03 (F=.02**). The multiple comparisons showed that no notable differences were observed for panicle number between the control and soil Cu treatments of or 0 mg/kg. However, plants under lower levels of Cu treatments( and 0 mg/kg) generally produced significantly more panicles than those under higher soil Cu levels (0 0 mg/kg). The panicle number under soil Cu level of 0 mg/kg was significantly higher than those at soil levels of 0, 0, 0 mg/kg. The panicle number at soil Cu level of 0 mg/kg showed no significant difference with that at soil Cu level of 0 mg/kg, but significantly higher than that at soil Cu level of 0 mg/kg. No difference was detected for panicle number between soil Cu treatment of 0 and 0 mg/kg (Table 2). The above results indicated that panicle number had no obvious change under lower levels of soil Cu stress, but was significantly reduced at higher Cu levels. The panicle number is determined by the time of tiller occurrence and productive tiller ratio. The results indicated that seedlings showed no obvious toxic symptoms after transplanting under soil Cu levels below mg/kg. However, the seedling leaves turned yellow, and the seedling recovery from transplanting was delayed obviously with increased levels of soil Cu. As showed in Fig. 2, with the increasing levels of soil Cu, tillering was delayed (or even no tiller emerged), the maximum tiller numbers were reduced, and the time when maximum tiller numbers were reached was postponed. However, the productive tiller rates were greatly elevated under Cu stress. The results indicated that the decreases of panicle number(except for higher soil Cu treatment) were caused by the inhibitory effects on tillering. Effect of Cu on number of spikelets per panicle ANOVA test indicated that there was a significant difference in number of spikelets per

XU Jia-kuan, et al. Effects of Soil Copper Concentrations on Growth and Development and Yield Formation of Rice 129 Number of stems and tillers 50 45 35 30 25 15 10 5 0 CK mg/kg Cu 0 mg/kg Cu 0 mg/kg Cu 0 mg/kg Cu 0 mg/kg Cu 0 mg/kg Cu 0 Days after transplanting (d) Fig. 2. Effect of Cu on tiller dynamics of rice in 03. panicle among different soil Cu treatments in 03 (F=39.17**). The multiple comparisons showed that number of spikelets per panicle subjected to 0 mg/kg soil Cu were all significantly lower than that of the control. No significant difference in number of spikelets per panicle was detected between soil Cu levels of and 0 mg/kg, whereas both of them were significantly higher than those at soil Cu levels between 0 to 0 mg/kg. The number of spikelets per panicle under soil Cu level of 0 mg/kg was significantly higher than those under soil Cu levels of 0, 0, 0 mg/kg, but no significant differences existed between soil Cu treatments of 0, 0 and 0 mg/kg (Table 2). The results indicated that soil Cu treatments had a large impact on number of spikelets per panicle, which decreased with the increase levels of soil Cu. Number of spikelets per panicle are the product of shoot dry weight (SDW) at heading and spikelet number-sdw ratio. As showed in Fig. 3-A, number of spikelets per panicle correlated significantly (P < 0.01) and positively with the SDW at heading, with the correlation coefficients of 0.952** in 02, 0.966** in 03, and 0.915** in the years 02 and 03, indicating that the number of spikelets per panicle increased with the increase of SDW at heading. Fig. 3-B indicated the number of spikelets per panicle also were significantly (P < 0.01) and positively correlated with the ratio of spikelet number-sdw, with the correlation coefficients of 0.914** in 02, 0.747** in 03, and 0.781** in both 02 and 03, indicating that spikelets per panicle increased with the increase of spikelet number-sdw ratio. Direct effects of SDW at heading (X 1 ) and spikelet number-sdw ratio (X 2 ) on number of spikelets per panicle (Y) were determined based on a stepwise regression. The results revealed that the two parameters both significantly affected the number of spikelets per panicle, according to data in 02, 03 or 02 03. Direct path coefficients of X 1 and X 2 to Y were 0.8 and 0.459 in 02, 0.797 and 0.307 in 03, and 0.702 and 0.449 in 02 03, respectively, indicating that the reduction of SDW at heading, caused by Cu stress, primarily determined the number of spikelets per panicle compared to spikelet number-sdw ratio. The decreases of SDW at heading, which resulted from delayed tillering and following higher No. of spikelets per panicle 1 1 1 y = 7.1624x - 73.814 r = 0.781**(n =13) 02 03 15 25 30 35 Spikelet surviving ability (spikelets/g) No. of spikelets per panicle 1 1 1 y = 33.87x - 33.935 r = 0.915**(n =13) 2.0 3.0 4.0 5.0 6.0 Dry matter weight per stem (g) Fig. 3. Relationship between dry matter weight per culm, spikelet surviving ability and number of spikelets per panicle of Wuxingjing 14.

130 Rice Science, Vol. 12, No. 2, 05 Total biomass (g/pot) Total (g pot -1 ) 0 1 1 1 0 CK 0 0 0 0 0 Soil Soil Cu Cu levels levels (mg/kg) (mg kg -1 ) Fig. 4. Effect of soil Cu concentration on total biomass production and economic coefficients in 03. ratio of small productive tiller (Fig. 2), together with decreased spikelet number-sdw ratio, were responsible for reduced spikelets per panicle under soil Cu stress. Effect of Cu on filled grain percentage and 0-grain weight Soil Cu stress had no obvious effect on filled grain percentage of both japonica rice cultivar Wuxiangjing 14 and indica rice cultivar Shanyou 63 (Table 2). 0-grain weight of Wuxiangjing 14 showed no difference under low soil Cu stress, but decreased under high Cu levels (Table 2). For Shanyou 63, 0-grain weight decreased with the increment of soil Cu levels (data not shown). Comparison of yield components under soil Cu stress indicated that number of panicles was affected the most severely, followed by number of spikelets per panicle, while both filled grain percentage and 0-grain weight changed little under Cu stress. Further analysis showed that the yield loss under lower levels of soil Cu ( or 0 mg/kg) was mainly attributed to the decreased spikelet number per panicle, while both panicle number and spikelet number per panicle were the main components contributed to yield reductions under higher levels of soil Cu (more than 0 mg/kg), with panicles playing a more important role than spikelet number per panicle (Table 2 and 3). Effect of Cu on total biomass and economic coefficient Our investigation showed that the total biomass under different soil Cu treatments from to 0 mg/kg were all significantly lower (P < 0.01 or P < 0.05) as compared to the control (Fig. 4). No difference existed in the total biomass between soil Cu treatments of and 0 mg/kg, but both of them were significantly higher than those under the other soil Cu treatments (above 0 mg/kg). The total biomass at soil Cu level of 0 mg/kg was significantly higher than that under soil Cu levels of 0, 0, 0 mg/kg. No significant difference was detected for total biomass between soil Cu treatment of 0 and 0 mg/kg, but both of them were significantly higher than those at soil Cu level of 0 mg/kg. The results indicated that the total biomass was significantly influenced at soil Cu level of mg/kg, and decreased with the increasing levels of Cu application. The results highlighted that economic coefficient showed some reductions with the increased levels of soil Cu. However, compared to the control, economic coefficient showed no change under soil Cu stress between to 0 mg/kg, but showed significant decreases under soil Cu levels of 0 and 0 mg/kg (data not shown). The results suggested that there were no significant impacts on economic coefficients within a wide range of soil Cu concentration except under extreme high soil Cu levels (>0 mg/kg). DISCUSSION Effect of Cu on rice growth and development Several researchers have obtained the different results about the response of plant height to soil Cu stress in previous reports. Kang et al [7] reported that Table 3. Path analysis of SDW at heading, spikelet number-sdw ratio to spikelet number per panicle. Path coefficient 02 03 02 03 1 Y 2 Y 1 Y 2 Y 1 Y 2 Y SDW (X 1, 1) 0.8 0.344 0.797 0.169 0.702 0.213 Spikelet number-sdw ratio (X 2, 2) 0.455 0.459 0.439 0.307 0.333 0.449

XU Jia-kuan, et al. Effects of Soil Copper Concentrations on Growth and Development and Yield Formation of Rice 131 the plant height of rice subjected to soil Cu levels of 90.33 241.5 mg/kg was significantly higher than the control (37.33 mg/kg); Chen et al [14] found no significant effect of soil Cu levels of 500 and 0 mg/kg on rice plant height. Su et al [8] and Hu et al [5] observed that rice plant height decreased sharply at soil Cu levels between 14.5 to 250 mg/kg and between 17.0 to 151.6 mg/kg, respectively. Our present research exhibited that the plant height at maturity was decreased by 4.2 and 6.6% at soil Cu levels of and 0 mg/kg, respectively, but decreased significantly by.6, 27.5, 37.2 and 48.4% at soil Cu levels of 0, 0, 0 and 0 mg/kg, respectively (Table 1). The rice plant height was reduced greatly with the increasing levels of soil Cu (r=-0.993**). Further analysis showed that no significant differences existed for plant height at different growth stages between the control and soil Cu treatment of or 0 mg/kg, but large differences were observed between the control and soil Cu levels above 0 mg/kg. Maximum differences for plant height between the control and soil Cu levels of 0, 0, 0, 0 mg/kg appeared at 30, 42, 42 and 62 days after transplanting, respectively. The differences between the control and different soil Cu treatments increased with the rice development and then decreased after maximum differences were reached. The differences were increased with the increase of soil Cu levels. There was no research concerning the effects of soil Cu concentrations on elongated internode number, leaf number on the main stem and heading date of rice plant. This experiment indicated that, compared to the control, the above mentioned growth parameters had no changes or small changes under soil Cu levels of and 0 mg/kg, significantly changes under Cu levels of above 0 mg/kg. The toxic effect of Cu on the plant growth was increased with the increasing levels of Cu application. Influences of Cu stress on the plant height and leaf number became larger with the advance of growth stages, then became smaller. The influences increased with the raising levels of soil Cu (Fig.1). Such phenomena suggested that rice plants showed a gradual adaptation to soil Cu stress, and exhibited compensation ability at middle and late growth stages. Effect of Cu on grain yield and its causes The present investigation showed that the grain yield decreased significantly with the increase of soil Cu levels. The result was highly consistent with Yang et al [15], Chen et al [14], Su et al 8], Kang et al [7] and Hu et al [5]. The limited data on the reasons of rice yield reductions by Cu stress were derived from rice plants, which indicated that the decrease of panicle number [5, 6], unfilled grain number [15] or reduction of [7, 8, 14] both panicle number and 0-grain weight contributed to the grain yield decrease. The present study indicated that grain yield reductions under lower soil Cu levels (, 0 mg/kg) mainly resulted from the decrease of spikelet number per panicle, but under higher soil Cu levels (above 0 mg/kg), the grain yield reductions were mainly caused by the decreases of both panicle number and number of spikelets per panicle, with panicle number playing a more important role than spikelet number per panicle (Table 1). Hu et al [5] presented that the decreases of panicle number by soil Cu stress were caused by the reduction of tiller number in rice. The present research showed that the reduction of panicle number were mainly attributed to severe inhibition of tiller occurrence under higher soil Cu stress. Excess Cu concentration led to slow recovery from transplanting, delayed tillering and reduced maximum tiller numbers. Though the productive tiller rate increased under soil Cu stress, as the decrease in maximum tiller number was greater than this increase, panicle number decreased with elevated soil Cu concentration (Fig. 1). However, further research is necessary to clarify inhibition mechanisms on tiller occurrence under soil Cu stress. Few studies had paid attention to the reason of spikelets reduction per panicle under soil Cu stress. This research indicated that the number of spikelets per panicle correlated significantly (P<0.01) and positively with the SDW at heading and spikelet number-sdw ratio (Fig. 3). The multiple regression analysis showed that SDW at heading and spikelet number-sdw ratio all significantly affected the number of spikelets per panicle. The number of spikelets per panicle increased with the increase of SDW at heading and spikelet number-sdw ratio.

132 Rice Science, Vol. 12, No. 2, 05 Path analysis showed that, according to data of 02, 03 or 02 and 03 combined, direct path coefficients of SDW at heading to number of spikelets per panicle were much larger than that of spikelet number-sdw ratio to spikelet number per panicle (Table 3), indicating that the decrease of spikelet number per panicle due to higher soil Cu stress were caused by the reduction of both SDW at the heading date and spikelet number-sdw ratio, the decrease of SDW at heading resulting from delayed tiller occurrence and higher rate of small productive tiller (Fig. 2). Effect of Cu on total biomass and economic coefficient Rice grain yield is a function of total biomass and economic coefficient: in our study, the former was more sensitive to soil Cu stress than the latter. The total biomass decreased significantly with raising soil Cu levels, which is in accordance with the results of the previous studies [7, 8]. The biomass responses to soil Cu stress are probably associated with its inhibition on photosynthesis, because excessive Cu could restrain chlorophyll synthesis or result in chlorophyll decomposition [16]. There was little information about the effect of soil Cu stress on economic coefficient of rice plants. In present study no obvious changes in economic coefficient were observed under soil Cu levels below 0 mg/kg, only when reached 0 mg/kg or above, did the economic coefficient decrease significantly. The results suggested that there was no significant impact on economic coefficient within a wide range of soil Cu levels except under extreme high soil Cu levels (above 0 mg/kg). ACKNOWLEDGEMENTS This research was supported by Agricultural Three Item Project of Jiangsu Province (03) and Key Science and Technique Project of Changzhou in 03 (CE030). REFERENCES 1 The China Environmental Protection Bureau. GB15618-1995, Environment Quality Standard for Soil. Beijing: The Standard Press of China. 1995. (in Chinese) 2 Chang H Y, Sun B Y, Liu C S. Advances in the study of plants copper toxicity. J Shandong Agric Univ (Nat Sci), 00, 31(2), 227 230. (in Chinese with English abstract) 3 Zhong W K, Fan Y B, Wang M J. Pollution of heavy metals on crops and its countermeasures in China. Agro-Environ Prot, 01, (4) : 270 272. (in Chinese with English abstract) 4 He Z L. Soil-Chemical Balances of Pollution and Beneficial Elements. Beijing: China Environmental Science Press, 1998. 161 9. (in Chinese) 5 Hu Z Y, Shen H, Cao Z H. Distribution of Cu in soil-crop syculm polluted by Cu. Environ Sci, 00, 3 : 62 65. (in Chinese with English abstract) 6 Zhang J W, Liang W, Li D B, Gu J N, Yue Z T. Ecological effects of combined Cu, Pb and Zn soil pollution on rice production. Rural Eco-Environ, 1997, 13 (1) : 16. (in Chinese with English abstract) 7 Kang L J, Zhao M X, Zhao C A. Study on effect of Cu on rice and its migration accumulation behavior. Guangdong Trace Elements Sci, 1999, 6 (4): 43 44. (in Chinese with English abstract) 8 Su L S, Yuan H X. Effect of Cu and As on growth of rice. Trop & Subtrop Soil Sci, 1997, 6(4) : 194 197. (in Chinese with English abstract) 9 Xu X Y, Yang X E. Effects of copper and zinc on growth and superoxide dismutase activity of rice seedlings. J Shanxi Agric Univ, 1997, 17 (2) : 113 115. (in Chinese with English abstract) 10 Ge C L, Yang X Y, Liu X N, Sun J H, Luo S S, Wang Z G. Effects of heavy metal on the DNA methylation level in rice and wheat. J Plant Physiol Mol Biol, 02, 28(5) : 363 368. 11 Ge C L, Yang X Y, Sun J H, Wang Z G, Luo S S, Ma F, Gong Z. DNA damage caused by heavy metal stress in rice and wheat seedlings. J Plant Physiol Mol Biol, 02, 28 (6) : 419 424. 12 Fernando C L, Henrques F S. Copper toxicity in rice: Diagnositic criteria and effect on tissue Mn and Fe. Soil Sci, 1992, 154(2): 130 135. 13 Fernando C. Role of rice shoot vacuoles in copper toxicity regulation. Environ Exp Bot, 1998, 39 (3): 197 1. 14 Chen S Y, Xu H B, Xie M Y, Xu L F, Cao Y, Wu C Y, Yang Y Q. Study on movement of Cu and As in rice-soil syculm and its effects on rice. Rural Eco-Environ, 1995, 11(3): 15 18. (in Chinese with English abstract) 15 Yang G F, Li D B. Investigation and study of copper pollution in paddy soils near selected copper mines in south China. Rural Eco-Environ, 1990, (4): 55 58. (in Chinese with English abstract) 16 Ouzounidou G. Copper-induced changes on growth, methal content and photosynthetic function of Alyssum montanum plants. Environ Exp Bot, 1994, 34 (2) : 165 172.