论文部分内容阅读
Abstract [Objectives] This study was conducted to explore the control mechanism of agricultural nonpoint source pollution， and investigate the feasibility of promoting rice "three controls" nutrient management in Enping City.
[Methods] With highquality conventional rice as a material， such three treatments as three controls fertilization A（ZHY）and B（ZHY）and farmers conventional fertilization method FFP（ZXL）were set up， and the whole process of the lateseason plot experiment was recorded. The agronomic characteristics of rice population quality and yield components during rice growth and development under the "three controls" fertilization technology were analyzed.
[Results] Compared with the conventional fertilization method， the three controls A（ZHY）fertilization method improved rice yield by 27.13%， seed setting rate by 2.11% and 1 000grain weight by 3.30% when reducing N， P and K by 27.13%， 10.89% and 27.31%， respectively. In the case of three controls B（ZHY）omitting the last fertilization in the three controls fertilization method （4∶2∶3∶1）， which saved the formula fertilizer by 11.25%， no difference was caused in yield， but the seed setting rate and 1 000grain weight were still improved by 3.47% and 2.79%， respectively. Compared with the conventional fertilization method， the top first， second and third basal nodes of the three controls A（ZHY）fertilization method were shortened by 18.82%， 17.06% and 20.52%， respectively， which plays an important role in combating typhoon and resisting lodging.
[Conclusions] Compared with the conventional fertilization method， rice "threecontrols" nutrient management can improve yield and lodging resistance of rice， reduce fertilizer loss and agricultural nonpoint source pollution， and protect ecological environment.
Key words Rice; Three controls fertilization; Population quality; Yield components
On the basis of the "three controls" nutrient management （SSNM） of rice， experiment and demonstration have been carried out in Guangdong， Jiangxi and other provinces. This technology is characterized by fertilizer and medicine saving， high and stable yield， and obvious increase in yield. In the Enping rice area of Guangdong Province， overapplication of chemical fertilizers and pesticides， low fertilizer utilization rate and environmental pollution have become increasingly prominent. Farmers are accustomed to reapplication of nitrogen fertilizers in the early stage of rice growth， and the base fertilizer， regreening fertilizer and tillering fertilizer account for more than 80% of total nitrogen application. Because the total amount of nitrogen applied and the amount of nitrogen applied as the base and tillering fertilizers are too large， inefficient tillers occur in large quantities， resulting in low percentage of earbearing tiller， tooclose population and severe pests and diseases， and agrochemicals have to be applied in large quantities， threatening the food safety of rice and the sustainable development of agriculture. The largescale application of chemical fertilizers and pesticides has also increased the cost of growing rice， affecting the benefits of rice planting and increasing farmers income. The three controls fertilization technology of rice is aimed at the above problems. The main contents of the technology are as follows： controlling the total nitrogen application rate and the nitrogen application rate of the base fertilizer to increase the nitrogen use efficiency and reduce the environmental pollution， controlling the ineffective tillering and the highest number of seedlings to increase the percentage of earbearing tiller and quality of rice population and achieve high yield and stability， and controlling the occurrence of pests and diseases to reduce the amount of pesticides and improve the food safety of rice. In order to further study the feasibility of promoting the "three controls" technology in the area of Enping City， a lateseason plot experiment was carried out with highquality conventional rice as an experimental material， in which the population quality and yield components of rice were investigated by setting such 3 treatments as the three controls A（ZHY）（4∶2∶3∶1）， three controls B（ZHY）（omitting the last fertilization） and conventional fertilization in 3 replicates， and the yield increasing mechanism of the "three controls" nutrient management technology was analyzed. This study provides a theoretical basis for the promotion and application of the "three controls" nutrient management technology. Materials and Methods
General Situation of Test Area
Enping City of Guangdong Province has a typical southern subtropical climate. The average annual temperature is 23.1 ℃， and the highest and lowest temperatures are 36.4 and 4.9 ℃， respectively. The annual precipitation is 2 559.9 mm and the maximum daily precipitation is 185.5 mm. The experiment was conducted in the late autumn of 2015 at the Menkoudong rice 100 mu demonstration site Jingang Village， Enping Street， Enping City.The test site had been a paddy field over years before the test， which was planted with rice previously， and the soil texture was sandy loam with pH 4.76， organic matter content 29.22 g/kg， total nitrogen content 0.86 g/kg， alkalihydrolyzale nitrogen 115.5 mg/kg，effective phosphorus 28.9mg/kg and rapidly available potassium 29.14mg/kg. The tested rice variety was conventional highyield highquality rice variety Fan 19， and the target yield was 6 000 kg/hm2.
Test design
A total of 3 treatments were set， in 3 replicates， 9 plots in total， including the conventional fertilization method （referred to as FFP） and the "three controls" fertilization method （referred to as the three controls） treatments. The specific design scheme is shown in Table 1. The experiment adopted randomized block arrangement， and a protection line was set around each plot. Various treatments were separated by constructing 20 m×15 cm field foundations therebetween and covering with nylon membrane. The singlerow singleirrigation method was implemented in each plot. The area of each treatment plot was 140 m2， in which the planting specification was 20 m×15 cm and the density was 270 000-300 000holes/hm2. Quantitative manual planting was adopted through pulling lines to ensure consistent planting specification in each plot. The transplanting was performed at the stage of 3.5-4.0 leaves. Shallow water was kept in the field after planting， and the water was drained at 20-25 d to slightly sun the field. Other field management was carried out according to routine management. The specific fertilization methods are shown in Table 2 below.
According to the farmers fertilization method of XilianZhen（referred to as FFP（ZXL）， the base fertilizer and Kuaimei formula fertilizer were applied at 375 and 150 kg/hm2， respectively; 5 d after transplanting， the regreening fertilizer was applied at a rate of 500 kg/hm2; 15 d after transplanting， Kuaimei formula fertilizer was applied at 300 kg/hm2; 255 kg/hm2 of thickening fertilizer was applied 30 d after transplanting; and 75 kg/hm2 of potassium chloride fertilizer was applied 60 d after transplanting. Totally， the fertilizers were applied according to N： 202.61 kg/hm2， P2O5： 53.03 kg/hm2 and K2O： 176.44 kg/hm2. For the three controls， the "three controls" Fertilizer Application method of HuoyouZhen （4∶2∶3∶1） （referred to as the three controls A（ZHY）was set. In the late season， Kuaimei formula fertilizer was applied as the base fertilizer at 270 kg/km2; 15 d after transplanting， 135 kg/hm2 of Kuaimei formula fertilizer was applied; 352.5 kg/hm2 of the fertilizer was applied for the plants to grow thick 35 d after transplanting; and 60.75 kg/hm2 of the fertilizer was applied 60 d after transplanting. The threecontrol A（ZHY）had following total nutrient amounts： N： 162.00 kg/hm2， P2O5： 47.25 kg/hm2 and K2O： 128.25 kg/hm2.
The "three controls" Fertilizer Application method of HuoyouZhen（4∶2∶3∶0） （referred to as the three controls B（ZHY）was also set. In the late season， Kuaimei formula fertilizer was applied as the base fertilizer at 270 kg/km2; 15 d after transplanting， 135 kg/hm2 of Kuaimei formula fertilizer was applied; 202.5 kg/hm2 of the fertilizer was applied for the plants to grow thick 35 d after transplanting; and no fertilizer was applied 60 d after transplanting. The three controls B（ZHY）had following total nutrient amounts： N： 145.80 kg/hm2， P2O5： 42.53 kg/hm2 and K2O： 115.43 kg/hm2. The randomized block design was adopted in 3 replicates， and each plot had an area of 140 m2.
Seeding was performed on July 15， and transplanting was performed on July 27. All of the seedlings were cultivated in trays. The plants were transplanted according to the specification of 20 cm×20 cm. Wetting irrigation was implemented from the critical tillering stage until the tip of the top second leaf was observed， while a shallow water layer was kept in the field at other time. Water was drained 7 d before harvesting. Other field management was the same as general highyield fields， and pests and diseases were strictly controlled.
Investigation and determination items
Population quality and yield components
Ten holes were fixed in each plot with 3 replicates. The basal stem node length and diameter were measured， and the full grains and empty grains were calculated. The full grains were naturally airdried and weighed for 1 000grain weight.
Yield and its components
During the mature stage， the yield of each plot was calculated and converted into the yield at the water content of 14%.
Statistical analysis
Data processing and statistical analysis were performed using SPSS 17.0 and Excel2003. Results and Analysis
Effects of "three controls" nutrient management on rice yield and yield components
It can be seen from Table 3 that the yields of three controls A（ZHY）and three controls B（ZHY）were （5 872.58±177.40）and （5 325.42±198.54）kg/hm2， respectively， and the test field had higher soil fundamental fertility. Compared with FFP（ZHR）， the three controls treatments had obvious yield increasing effects. The yields of three controls A（ZHY）and three controls B（ZHY）were 849.33and 302.17 kg/hm2 higher than that of FFP（ZHR）， respectively， showing the increasing rates of 16.91% and 6.02%， respectively， which reached a significant level.
Such two yield components as the seed setting rate and 1 000grain weight of three controls A（ZHY）and three controls B（ZHY）were higher than the FFP treatment. Compared with FFP（ZHR）， the three controls A（ZHY）fertilization method increased the rice yield by 16.9%， seed setting rate by 2.11% and 1 000grain weight by 3.30%when reducing N， P and K by 27.13%， 10.89%， and 27.31%， respectively. Compared with the three controls A（ZHY）fertilization method， the three controls B（ZHY）fertilization method can save 11.25% of the formula fertilizer. Compared with FFP（ZHR）， the three controls B（ZHY）fertilization method increased the yield by 6.01%in the case of reducing N， P and K by 28.04%， 19.80% and 34.57%， respectively. With a significantly reduction of the last fertilization on the basis of the threecontrol fertilization method （4∶2∶3∶1）， three controls B（ZHY）did not caused a significant difference in yield， but saved 11.25% of the formula fertilizer. It can be seen that the three controls fertilization method had obvious material production advantages， and the 1 000grainweight was higher than the conventional fertilization method. It can be seen that large storage capacity and high seed setting rate and 1 000grain weight are important basis for the higher yield of the three controls technology than the FFP method.
Effects of "three controls" nutrient management on basal stem node length and diameter of rice
It can be seen from Table 3 that three controls A（ZHY）and three controls B（ZHY）treatments showed the basal stem node diameters in sharp contrast with FFP（ZHR）. Three controls A（ZHY）showed the top first， top second and top third basal stem node diameters 4.00%， 8.86%， and 0.56% thicker than FFP（ZXL）， respectively. Three controls B（ZHY）exhibited the top first， top second and top third basal stem node diameters 16.36%， 5.68% and 12.55% thinner than FFP（ZXL）， respectively. The differences were extremely significant. It can be seen from Table 4 that the basal stem lengths of three controls A（ZHY）and three controls B（ZHY）were in sharp contrast with FFP（ZXL）. Three controls A（ZHY）had the top first/plant height， top second/plant height， top third/plant height and length of the three basal stem nodes/plant height of 1.88%±0.212 2%， 4.56%±0.429 3%， 8.53%±1.350% and 14.96%±1.128 1%， respectively; three controls B（ZHY）exhibited the top first/plant height， top second/plantheight， top third/plant height and length of the three basal stem nodes/plant height of 1.77%±0.451%， 5.63%±0.417 8%， 8.87%±0.161 6% and 16.27%±0.357 9%， respectively; and FFP（ZHR）showed the top first/plant height， top second/plant height， top third/plant height and length of the three basal stem nodes/plant height of 2.30%±0.190 3%， 6.65%±0.400 7%， 11.55%±1.561 8% and 20.49%±1.377 5%， respectively. The lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes in the three controls A（ZHY）treatment were shorter than those in FFP（ZXL）by 22.34%， 45.83%， 35.48% and 36.96%， respectively. The lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes in the three controls B（ZHY）treatment were shorter than those in FFP（ZXL）by 29.94%， 18.12%， 30.21% and 25.94%， respectively. The differences were extremely significant.
"Three controls" nutrient management on the regression equations of rice plant height （Y） and basal node agronomic traits （X）
It can be seen from Table 5 that the slopes of the regression equations of rice plant height （Y） and agronomic traits （X） in the threecontrol A（ZHY）and three controls B（ZHY）treatments were in sharp contrast with FFP（ZXL）. For three controls A（ZHY）， the lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes had the slopes and intercepts of （-0.041， 86.79）， （1.845， 79.33）， （-0.381， 89.56） and （-0.143， 88.60）， respectively. As to the third controls B（ZHY）， the lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes had the slopes and intercepts of （1.260， 94.14）， （2.632， 82.23） （0.068， 95.78） and （0.548， 87.79）， respectively. In the case of FFP（ZHR）， the lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes showed the slopes and intercepts of （-0.032， 87.90）， （5.624， 55.10）， （0.817， 79.70） and （0.899， 71.87）， respectively. The differences were extremely significant. The relative values of the slopes were smaller， but can also reflect the stable plant height and the growth of the base stem. The results showed that the slopes of the regression equations of rice plant height （Y） and agronomic traits （X） in the three controls A（ZHY）and the three controls B（ZHY）treatments were in sharp contrast with FFP（ZXL）. The top first lengths in three controls A（ZHY）and three controls B（ZHY）exhibited larger absolute values of slopes， which also reflected that the growth of the base stem was affected by the fertilizer， and the base stem node elongated and jointed in latter growth period. According to the relevant regression equations， some relevant contents of each treatment can be grasped and regulated from the mutual constraint and coordination relations between plant height， stem diameter and yield， but attention should be paid to factors such as human management， error and meteorology. Discussion
"Three controls" nutrient management increases rice yield
It can be seen from Table 3 that the yields of three controls A（ZHY）and three controls B（ZHY）were （5 872.58±177.40） and （5 325.42±198.54）kg/hm2， respectively， and the test field had higher soil fundamental fertility than FFP（ZHR）. Compared with FFP（ZHR）， the three controls treatments had obvious yield increasing effects. The yields of three controls A（ZHY）and three controls B（ZHY）were 849.33 and 302.17 kg/hm2 higher than that of FFP（ZHR）， respectively， showing the increasing rates of 16.91% and 6.02%， respectively， which reached a significant level. Such two yield components as the seed setting rates and 1 000grain weights of three controls A（ZHY）and three controls B（ZHY）were higher than the FFP treatment. Compared with FFP（ZHR）， the three controls A（ZHY）fertilization method increased rice yield by 16.9%， seed setting rate by 2.11% and 1 000grain weight by 3.30% when reducing N， P and K by 27.13%， 10.89%， and 27.31%， respectively. Compared with the three controls A（ZHY）fertilization method， the three controls B（ZHY）fertilization method can save 11.25%of the formula fertilizer. Compared with FFP（ZHR）， the three controls B（ZHY）fertilization method increased the yield by 6.01%in the case of reducing N， P and K by 28.04%， 19.80% and 34.57%， respectively. With a significantly reduction of the last fertilization on the basis of the threecontrol fertilization method （4∶2∶3∶1）， three controls B（ZHY）did not caused a significant difference in yield， but saved 11.25% of the formula fertilizer. It can be seen that the threecontrol fertilization method had obvious material production advantages， and the 1 000grain weight was higher than the conventional fertilization method. It can be seen that large storage capacity and high seed setting rate and 1 000grain weight are important basis for the higher yield of the three controls technology than the FFP method. The results of previous studies by other researchers research groups were consistent[4， 6-12]， and similar results were obtained for the field nitrogen fertilizer management techniques[13-16]. Studies have shown that this is an important material basis for the high yield of the "three controls" fertilization method. From the view of different growth stages， compared with the conventional fertilization method， the "three controls" fertilization method has no advantages in dry matter production in the early stage （before ear differentiation）， and its material production advantage mainly comes from the high chlorophyll content in the middle and late stages， but has nothing to do with the leaf area index. The high chlorophyll content in the middle and late stages is conducive to the increase of rice photosynthetic rate， which is conducive to the accumulation of dry matter and the promotion of large ear formation and grain filling. Previous studies have indicated that there is a significant positive correlation between rice yield and dry matter accumulation after heading[17-18]. In this study， based on the "three controls" fertilization method （4231）， the "three controls" fertilization method （42311） was tested. From the perspective of rice yield components， the "three controls" fertilization method （42311） improved the number of effective panicles， seed setting rate and 1 000grain weight to different degrees compared with the conventional fertilization method， which is the same as previous research results[4，6-12]. In the "three controls" fertilization method， the amount of nitrogen applied in the late stage is high， as well as the nitrogen uptake， and a higher chlorophyll content is maintained in the late stage， so that there are fewer tillers and more effective panicles. Increasing the percentage of earbearing tiller is conducive to improving the net photosynthetic productivity after heading[20]， which is conducive to the coordination of grain structure and the unification of panicle number and large panicles[17]and the realization of higher yield. "Three controls" nutrient management improves rice lodging resistance
The experimental research and application show that compared with the conventional fertilization method， the "three controls" nutrient management not only has high yield， but also strong lodging resistance and good stable yield[4， 6-8， 10-11]. The same result was obtained in this study. The 3 replicates of the conventional fertilization method all suffered from lodging， while no lodging occurred in the "three controls" fertilization method. In this study， regardless of whether or not lodging occurred， the "three controls" fertilization method had different degrees of yield increase in various replicates compared with the conventional fertilization method， showing good technical stability. It can be seen from Table 3 that three controls A（ZHY）and three controls B（ZHY）treatments were in sharp contrast with FFP（ZHR）in basal stem diameter. Three controls A（ZHY）showed the top first， top second and top third basal stem node diameters 4.00%， 8.86%， and 0.56% thicker than FFP（ZXL）， respectively. Three controls B（ZHY）exhibited the top first， top second and top third basal stem node diameters 16.36%， 5.68% and 12.55% thinner than FFP（ZXL）， respectively. The differences were extremely significant. The lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes in the three controls A（ZHY）treatment were shorter than those in FFP（ZXL）by 22.34%， 45.83%， 35.48% and 36.96%， respectively. The lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes in the three controls B（ZHY）treatment were shorter than those in FFP（ZXL）by 29.94%， 18.12%， 30.21% and 25.94%， respectively. These results provided a theoretical basis for the lodging resistance.
Conclusions
Through the implementation of the agricultural nonpoint source pollution control project and the study on the control mechanism， on the basis of soil testing and formulated fertilization， the formula fertilizer recommended by the World Bank is fully used in the basic farmland in the project area. In rice planting， 5-7 kinds of fertilizers including urea， phosphate fertilizer， compound fertilizer， compound mixed fertilizer and potash fertilizer used by farmers previously are reduced to Kuaimei formula fertilizer （24719）， which meant that the varieties are reduced by more than 80%. At the same time， before the implementation of the project， the farmers generally applied the fertilizers twice， and the amount of fertilizer loss exceeded 20%. Before the project implementation， the average amount of fertilizers used by farmers was 1 050 kg/hm2 （including 300 kg/hm2 of Caihong compound fertilizer for base application， 150 kg/hm2 of urea for weeding， 450 kg/hm2 of Caihong compound fertilizer for tillering and 150 kg/hm2 of Lifeishi potassium fertilizer， equivalent to 537 kg/hm2 of pure fertilizer）. After the implementation of the project， 675 kg/hm2 of the formula fertilizer （24719） （equivalent to 337.5 kg/hm2 of pure fertilizer） is applied， reducing 199.5 kg/hm2 of pure fertilizer. Farmers in the project area adopt the three controls nutrient management （4∶2∶3∶1）of rice， during which fertilization is performed for 4 times （one base and three topdressing）. Furthermore， in the case of three controls B（ZHY）omitting the last fertilization in the three controls fertilization method （4∶2∶3∶1）， which meant that 11.25%of the formula fertilizer was reduced， no difference was caused in yield， but the seed setting rate and 1 000grain weight were still improved by 3.47% and 2.79%， respectively. It can be seen that the three controls fertilization method has obvious material production advantages. Compared with the conventional fertilization method， the basal node of the three controls A（ZHY）fertilization method was shortened as much as 20.52%， which plays an important role in combating typhoon and resisting lodging. In addition， with reference to relevant research and analysis， the "threecontrols" nutrient management can directly improve the absolute fertilizer utilization rate by 13.53%， improve the relative fertilizer utilization rate to 59.06%， and reduce fertilizer loss， thereby effectively suppressing water eutrophication， reducing agricultural nonpoint source pollution and protecting the ecological environment. References
[1] PENG SB， HUANG JL， ZHONG XH， et al. Research Strategy in Improving Fertilizernitrogen Use Efficiency of Irrigated Rice in China[J]. Scientia Agricultura Sinica， 2002， 35（9）： 1095-1103. （in Chinese）
[2] SU ZF， ZHANG J， WANG HB， et al. Study on relationship of tiller development of rice population with the effective ear percentage and rice formulation[J]. Jiangsu Journal of Agricultural Sciences， 1997， 18（1）： 36-40. （in Chinese）
[3] ZHONG XH， HUANG NR， ZHENG HB， et al. Specification for the "Threecontrols" nutrient management technology for irrigated rice[J]. Guangdong Agricultural Sciences， 2007（5）： 13-15， 43. （in Chinese）
[4] HUANG NR， ZHONG XH， ZHENG HB. Application effect of threecontrol fertilization technology for rice[J]. Guangdong Agricultural Sciences， 2007（5）： 16-18. （in Chinese）
[5] ZHONG XH， HUANG NR， ZHENG HB， et al. Some principles for the "Threecontrols" nutrient management technology for irrigated rice[J]. Guangdong Agricultural Sciences， 2007（5）： 19-22. （in Chinese）
[6] WU HR， DING WC， CAI TY， et al. Demonstration of rice "Three controls" nutrient management in Leizhou City[J]. Guangdong Agricultural Sciences， 2008（12）： 79-80， 86. （in Chinese）
[7] HUANG NR， LIANG XM， LI JM， et al. Application of the "Three controls" nutrient management technology of rice at Gaoyao of central Guangdong[J]. Guangdong Agricultural Sciences， 2009（3）： 17-19， 24. （in Chinese）
[8] DAI YX， HUANG NR， LI JY， et al. Demonstration and application of rice "Three controls" nutrient management in Xinhui District[J]. Guangdong Agricultural Sciences， 2009（3）： 20-21， 30. （in Chinese）
[9] HUANG YJ， ZHENG BN， XIE ZS. Application effect of rice "Three controls" nutrient management in Yangjiang City[J]. Guangdong Agricultural Sciences， 2009（3）： 22-24. （in Chinese）
[10] LI LZ， JIANG XZ. Demonstration and application of rice "Three controls" nutrient management in Maonan District[J]. Guangdong Agricultural Sciences， 2009（3）： 25-26， 35. （in Chinese）
[11] XIONG ZG， PAN XX， HU HM， et al. Application effect of rice "Three controls" nutrient management in Qujiang District[J]. Guangdong Agricultural Sciences， 2009（3）： 27-29. （in Chinese）
[12] LIANG GC， LI XT， DU M. Demonstration of rice "Three controls" nutrient management in Shantou City[J]. Guangdong Agricultural Sciences， 2009（3）： 29-30. （in Chinese）
[13] LIU LJ， SANG DZ， LIU CL， et al. Effects of realtime and sitespecific nitrogen managements on rice yield and nitrogen use efficiency[J]. Scientia Agricultura Sinica， 2003， 36（12）： 1456-1461. （in Chinese） [14] PENG XL， LIU YY， LUO SG， et al. Effects of the sitespecific nitrogen management on yield and dry matter accumulation of rice in cold areas of Northeastern China[J]. Scientia Agricultura Sinica， 2006， 39（11）： 2286-2293. （in Chinese）
[15] LIU LJ， XU W， WU CF， et al. Characteristics of growth， development and nutrient uptake in rice under sitespecific nitrogen management[J]. Scientia Agricultura Sinica， 2007， 21（2）： 167-173. （in Chinese）
[16] HE F， HUANG JL， CUI KH， et al. Effect of realtime and sitespecific nitrigen management on rice yield and quality[J]. Scientia Agricultura Sinica， 2007， 40（1）： 123-132. （in Chinese）
[17] LING QH， SU ZF， ZHANG HQ. Relationship between earbearing tiller percentage and population quality and its influential factors[J]. Acta Agronomica Sinica， 1995， 21（4）： 464-469. （in Chinese）
[18] SU ZF， WANG HB， DU YL， et al. Relationship between population quality and yield formation at middle growth stage in rice[J]. Scientia Agricultura Sinica， 1998， 31（5）： 19-25. （in Chinese）
[19] LING QH， ZHANG HC， CAI JZ， et al. Discussion on quality of high yielding rice population and its optimal control[J]. Scientia Agricultura Sinica， 1993， 26（6）： 1-11. （in Chinese）
Editor： Yingzhi GUANG Proofreader： Xinxiu ZHU
[Methods] With highquality conventional rice as a material， such three treatments as three controls fertilization A（ZHY）and B（ZHY）and farmers conventional fertilization method FFP（ZXL）were set up， and the whole process of the lateseason plot experiment was recorded. The agronomic characteristics of rice population quality and yield components during rice growth and development under the "three controls" fertilization technology were analyzed.
[Results] Compared with the conventional fertilization method， the three controls A（ZHY）fertilization method improved rice yield by 27.13%， seed setting rate by 2.11% and 1 000grain weight by 3.30% when reducing N， P and K by 27.13%， 10.89% and 27.31%， respectively. In the case of three controls B（ZHY）omitting the last fertilization in the three controls fertilization method （4∶2∶3∶1）， which saved the formula fertilizer by 11.25%， no difference was caused in yield， but the seed setting rate and 1 000grain weight were still improved by 3.47% and 2.79%， respectively. Compared with the conventional fertilization method， the top first， second and third basal nodes of the three controls A（ZHY）fertilization method were shortened by 18.82%， 17.06% and 20.52%， respectively， which plays an important role in combating typhoon and resisting lodging.
[Conclusions] Compared with the conventional fertilization method， rice "threecontrols" nutrient management can improve yield and lodging resistance of rice， reduce fertilizer loss and agricultural nonpoint source pollution， and protect ecological environment.
Key words Rice; Three controls fertilization; Population quality; Yield components
On the basis of the "three controls" nutrient management （SSNM） of rice， experiment and demonstration have been carried out in Guangdong， Jiangxi and other provinces. This technology is characterized by fertilizer and medicine saving， high and stable yield， and obvious increase in yield. In the Enping rice area of Guangdong Province， overapplication of chemical fertilizers and pesticides， low fertilizer utilization rate and environmental pollution have become increasingly prominent. Farmers are accustomed to reapplication of nitrogen fertilizers in the early stage of rice growth， and the base fertilizer， regreening fertilizer and tillering fertilizer account for more than 80% of total nitrogen application. Because the total amount of nitrogen applied and the amount of nitrogen applied as the base and tillering fertilizers are too large， inefficient tillers occur in large quantities， resulting in low percentage of earbearing tiller， tooclose population and severe pests and diseases， and agrochemicals have to be applied in large quantities， threatening the food safety of rice and the sustainable development of agriculture. The largescale application of chemical fertilizers and pesticides has also increased the cost of growing rice， affecting the benefits of rice planting and increasing farmers income. The three controls fertilization technology of rice is aimed at the above problems. The main contents of the technology are as follows： controlling the total nitrogen application rate and the nitrogen application rate of the base fertilizer to increase the nitrogen use efficiency and reduce the environmental pollution， controlling the ineffective tillering and the highest number of seedlings to increase the percentage of earbearing tiller and quality of rice population and achieve high yield and stability， and controlling the occurrence of pests and diseases to reduce the amount of pesticides and improve the food safety of rice. In order to further study the feasibility of promoting the "three controls" technology in the area of Enping City， a lateseason plot experiment was carried out with highquality conventional rice as an experimental material， in which the population quality and yield components of rice were investigated by setting such 3 treatments as the three controls A（ZHY）（4∶2∶3∶1）， three controls B（ZHY）（omitting the last fertilization） and conventional fertilization in 3 replicates， and the yield increasing mechanism of the "three controls" nutrient management technology was analyzed. This study provides a theoretical basis for the promotion and application of the "three controls" nutrient management technology. Materials and Methods
General Situation of Test Area
Enping City of Guangdong Province has a typical southern subtropical climate. The average annual temperature is 23.1 ℃， and the highest and lowest temperatures are 36.4 and 4.9 ℃， respectively. The annual precipitation is 2 559.9 mm and the maximum daily precipitation is 185.5 mm. The experiment was conducted in the late autumn of 2015 at the Menkoudong rice 100 mu demonstration site Jingang Village， Enping Street， Enping City.The test site had been a paddy field over years before the test， which was planted with rice previously， and the soil texture was sandy loam with pH 4.76， organic matter content 29.22 g/kg， total nitrogen content 0.86 g/kg， alkalihydrolyzale nitrogen 115.5 mg/kg，effective phosphorus 28.9mg/kg and rapidly available potassium 29.14mg/kg. The tested rice variety was conventional highyield highquality rice variety Fan 19， and the target yield was 6 000 kg/hm2.
Test design
A total of 3 treatments were set， in 3 replicates， 9 plots in total， including the conventional fertilization method （referred to as FFP） and the "three controls" fertilization method （referred to as the three controls） treatments. The specific design scheme is shown in Table 1. The experiment adopted randomized block arrangement， and a protection line was set around each plot. Various treatments were separated by constructing 20 m×15 cm field foundations therebetween and covering with nylon membrane. The singlerow singleirrigation method was implemented in each plot. The area of each treatment plot was 140 m2， in which the planting specification was 20 m×15 cm and the density was 270 000-300 000holes/hm2. Quantitative manual planting was adopted through pulling lines to ensure consistent planting specification in each plot. The transplanting was performed at the stage of 3.5-4.0 leaves. Shallow water was kept in the field after planting， and the water was drained at 20-25 d to slightly sun the field. Other field management was carried out according to routine management. The specific fertilization methods are shown in Table 2 below.
According to the farmers fertilization method of XilianZhen（referred to as FFP（ZXL）， the base fertilizer and Kuaimei formula fertilizer were applied at 375 and 150 kg/hm2， respectively; 5 d after transplanting， the regreening fertilizer was applied at a rate of 500 kg/hm2; 15 d after transplanting， Kuaimei formula fertilizer was applied at 300 kg/hm2; 255 kg/hm2 of thickening fertilizer was applied 30 d after transplanting; and 75 kg/hm2 of potassium chloride fertilizer was applied 60 d after transplanting. Totally， the fertilizers were applied according to N： 202.61 kg/hm2， P2O5： 53.03 kg/hm2 and K2O： 176.44 kg/hm2. For the three controls， the "three controls" Fertilizer Application method of HuoyouZhen （4∶2∶3∶1） （referred to as the three controls A（ZHY）was set. In the late season， Kuaimei formula fertilizer was applied as the base fertilizer at 270 kg/km2; 15 d after transplanting， 135 kg/hm2 of Kuaimei formula fertilizer was applied; 352.5 kg/hm2 of the fertilizer was applied for the plants to grow thick 35 d after transplanting; and 60.75 kg/hm2 of the fertilizer was applied 60 d after transplanting. The threecontrol A（ZHY）had following total nutrient amounts： N： 162.00 kg/hm2， P2O5： 47.25 kg/hm2 and K2O： 128.25 kg/hm2.
The "three controls" Fertilizer Application method of HuoyouZhen（4∶2∶3∶0） （referred to as the three controls B（ZHY）was also set. In the late season， Kuaimei formula fertilizer was applied as the base fertilizer at 270 kg/km2; 15 d after transplanting， 135 kg/hm2 of Kuaimei formula fertilizer was applied; 202.5 kg/hm2 of the fertilizer was applied for the plants to grow thick 35 d after transplanting; and no fertilizer was applied 60 d after transplanting. The three controls B（ZHY）had following total nutrient amounts： N： 145.80 kg/hm2， P2O5： 42.53 kg/hm2 and K2O： 115.43 kg/hm2. The randomized block design was adopted in 3 replicates， and each plot had an area of 140 m2.
Seeding was performed on July 15， and transplanting was performed on July 27. All of the seedlings were cultivated in trays. The plants were transplanted according to the specification of 20 cm×20 cm. Wetting irrigation was implemented from the critical tillering stage until the tip of the top second leaf was observed， while a shallow water layer was kept in the field at other time. Water was drained 7 d before harvesting. Other field management was the same as general highyield fields， and pests and diseases were strictly controlled.
Investigation and determination items
Population quality and yield components
Ten holes were fixed in each plot with 3 replicates. The basal stem node length and diameter were measured， and the full grains and empty grains were calculated. The full grains were naturally airdried and weighed for 1 000grain weight.
Yield and its components
During the mature stage， the yield of each plot was calculated and converted into the yield at the water content of 14%.
Statistical analysis
Data processing and statistical analysis were performed using SPSS 17.0 and Excel2003. Results and Analysis
Effects of "three controls" nutrient management on rice yield and yield components
It can be seen from Table 3 that the yields of three controls A（ZHY）and three controls B（ZHY）were （5 872.58±177.40）and （5 325.42±198.54）kg/hm2， respectively， and the test field had higher soil fundamental fertility. Compared with FFP（ZHR）， the three controls treatments had obvious yield increasing effects. The yields of three controls A（ZHY）and three controls B（ZHY）were 849.33and 302.17 kg/hm2 higher than that of FFP（ZHR）， respectively， showing the increasing rates of 16.91% and 6.02%， respectively， which reached a significant level.
Such two yield components as the seed setting rate and 1 000grain weight of three controls A（ZHY）and three controls B（ZHY）were higher than the FFP treatment. Compared with FFP（ZHR）， the three controls A（ZHY）fertilization method increased the rice yield by 16.9%， seed setting rate by 2.11% and 1 000grain weight by 3.30%when reducing N， P and K by 27.13%， 10.89%， and 27.31%， respectively. Compared with the three controls A（ZHY）fertilization method， the three controls B（ZHY）fertilization method can save 11.25% of the formula fertilizer. Compared with FFP（ZHR）， the three controls B（ZHY）fertilization method increased the yield by 6.01%in the case of reducing N， P and K by 28.04%， 19.80% and 34.57%， respectively. With a significantly reduction of the last fertilization on the basis of the threecontrol fertilization method （4∶2∶3∶1）， three controls B（ZHY）did not caused a significant difference in yield， but saved 11.25% of the formula fertilizer. It can be seen that the three controls fertilization method had obvious material production advantages， and the 1 000grainweight was higher than the conventional fertilization method. It can be seen that large storage capacity and high seed setting rate and 1 000grain weight are important basis for the higher yield of the three controls technology than the FFP method.
Effects of "three controls" nutrient management on basal stem node length and diameter of rice
It can be seen from Table 3 that three controls A（ZHY）and three controls B（ZHY）treatments showed the basal stem node diameters in sharp contrast with FFP（ZHR）. Three controls A（ZHY）showed the top first， top second and top third basal stem node diameters 4.00%， 8.86%， and 0.56% thicker than FFP（ZXL）， respectively. Three controls B（ZHY）exhibited the top first， top second and top third basal stem node diameters 16.36%， 5.68% and 12.55% thinner than FFP（ZXL）， respectively. The differences were extremely significant. It can be seen from Table 4 that the basal stem lengths of three controls A（ZHY）and three controls B（ZHY）were in sharp contrast with FFP（ZXL）. Three controls A（ZHY）had the top first/plant height， top second/plant height， top third/plant height and length of the three basal stem nodes/plant height of 1.88%±0.212 2%， 4.56%±0.429 3%， 8.53%±1.350% and 14.96%±1.128 1%， respectively; three controls B（ZHY）exhibited the top first/plant height， top second/plantheight， top third/plant height and length of the three basal stem nodes/plant height of 1.77%±0.451%， 5.63%±0.417 8%， 8.87%±0.161 6% and 16.27%±0.357 9%， respectively; and FFP（ZHR）showed the top first/plant height， top second/plant height， top third/plant height and length of the three basal stem nodes/plant height of 2.30%±0.190 3%， 6.65%±0.400 7%， 11.55%±1.561 8% and 20.49%±1.377 5%， respectively. The lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes in the three controls A（ZHY）treatment were shorter than those in FFP（ZXL）by 22.34%， 45.83%， 35.48% and 36.96%， respectively. The lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes in the three controls B（ZHY）treatment were shorter than those in FFP（ZXL）by 29.94%， 18.12%， 30.21% and 25.94%， respectively. The differences were extremely significant.
"Three controls" nutrient management on the regression equations of rice plant height （Y） and basal node agronomic traits （X）
It can be seen from Table 5 that the slopes of the regression equations of rice plant height （Y） and agronomic traits （X） in the threecontrol A（ZHY）and three controls B（ZHY）treatments were in sharp contrast with FFP（ZXL）. For three controls A（ZHY）， the lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes had the slopes and intercepts of （-0.041， 86.79）， （1.845， 79.33）， （-0.381， 89.56） and （-0.143， 88.60）， respectively. As to the third controls B（ZHY）， the lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes had the slopes and intercepts of （1.260， 94.14）， （2.632， 82.23） （0.068， 95.78） and （0.548， 87.79）， respectively. In the case of FFP（ZHR）， the lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes showed the slopes and intercepts of （-0.032， 87.90）， （5.624， 55.10）， （0.817， 79.70） and （0.899， 71.87）， respectively. The differences were extremely significant. The relative values of the slopes were smaller， but can also reflect the stable plant height and the growth of the base stem. The results showed that the slopes of the regression equations of rice plant height （Y） and agronomic traits （X） in the three controls A（ZHY）and the three controls B（ZHY）treatments were in sharp contrast with FFP（ZXL）. The top first lengths in three controls A（ZHY）and three controls B（ZHY）exhibited larger absolute values of slopes， which also reflected that the growth of the base stem was affected by the fertilizer， and the base stem node elongated and jointed in latter growth period. According to the relevant regression equations， some relevant contents of each treatment can be grasped and regulated from the mutual constraint and coordination relations between plant height， stem diameter and yield， but attention should be paid to factors such as human management， error and meteorology. Discussion
"Three controls" nutrient management increases rice yield
It can be seen from Table 3 that the yields of three controls A（ZHY）and three controls B（ZHY）were （5 872.58±177.40） and （5 325.42±198.54）kg/hm2， respectively， and the test field had higher soil fundamental fertility than FFP（ZHR）. Compared with FFP（ZHR）， the three controls treatments had obvious yield increasing effects. The yields of three controls A（ZHY）and three controls B（ZHY）were 849.33 and 302.17 kg/hm2 higher than that of FFP（ZHR）， respectively， showing the increasing rates of 16.91% and 6.02%， respectively， which reached a significant level. Such two yield components as the seed setting rates and 1 000grain weights of three controls A（ZHY）and three controls B（ZHY）were higher than the FFP treatment. Compared with FFP（ZHR）， the three controls A（ZHY）fertilization method increased rice yield by 16.9%， seed setting rate by 2.11% and 1 000grain weight by 3.30% when reducing N， P and K by 27.13%， 10.89%， and 27.31%， respectively. Compared with the three controls A（ZHY）fertilization method， the three controls B（ZHY）fertilization method can save 11.25%of the formula fertilizer. Compared with FFP（ZHR）， the three controls B（ZHY）fertilization method increased the yield by 6.01%in the case of reducing N， P and K by 28.04%， 19.80% and 34.57%， respectively. With a significantly reduction of the last fertilization on the basis of the threecontrol fertilization method （4∶2∶3∶1）， three controls B（ZHY）did not caused a significant difference in yield， but saved 11.25% of the formula fertilizer. It can be seen that the threecontrol fertilization method had obvious material production advantages， and the 1 000grain weight was higher than the conventional fertilization method. It can be seen that large storage capacity and high seed setting rate and 1 000grain weight are important basis for the higher yield of the three controls technology than the FFP method. The results of previous studies by other researchers research groups were consistent[4， 6-12]， and similar results were obtained for the field nitrogen fertilizer management techniques[13-16]. Studies have shown that this is an important material basis for the high yield of the "three controls" fertilization method. From the view of different growth stages， compared with the conventional fertilization method， the "three controls" fertilization method has no advantages in dry matter production in the early stage （before ear differentiation）， and its material production advantage mainly comes from the high chlorophyll content in the middle and late stages， but has nothing to do with the leaf area index. The high chlorophyll content in the middle and late stages is conducive to the increase of rice photosynthetic rate， which is conducive to the accumulation of dry matter and the promotion of large ear formation and grain filling. Previous studies have indicated that there is a significant positive correlation between rice yield and dry matter accumulation after heading[17-18]. In this study， based on the "three controls" fertilization method （4231）， the "three controls" fertilization method （42311） was tested. From the perspective of rice yield components， the "three controls" fertilization method （42311） improved the number of effective panicles， seed setting rate and 1 000grain weight to different degrees compared with the conventional fertilization method， which is the same as previous research results[4，6-12]. In the "three controls" fertilization method， the amount of nitrogen applied in the late stage is high， as well as the nitrogen uptake， and a higher chlorophyll content is maintained in the late stage， so that there are fewer tillers and more effective panicles. Increasing the percentage of earbearing tiller is conducive to improving the net photosynthetic productivity after heading[20]， which is conducive to the coordination of grain structure and the unification of panicle number and large panicles[17]and the realization of higher yield. "Three controls" nutrient management improves rice lodging resistance
The experimental research and application show that compared with the conventional fertilization method， the "three controls" nutrient management not only has high yield， but also strong lodging resistance and good stable yield[4， 6-8， 10-11]. The same result was obtained in this study. The 3 replicates of the conventional fertilization method all suffered from lodging， while no lodging occurred in the "three controls" fertilization method. In this study， regardless of whether or not lodging occurred， the "three controls" fertilization method had different degrees of yield increase in various replicates compared with the conventional fertilization method， showing good technical stability. It can be seen from Table 3 that three controls A（ZHY）and three controls B（ZHY）treatments were in sharp contrast with FFP（ZHR）in basal stem diameter. Three controls A（ZHY）showed the top first， top second and top third basal stem node diameters 4.00%， 8.86%， and 0.56% thicker than FFP（ZXL）， respectively. Three controls B（ZHY）exhibited the top first， top second and top third basal stem node diameters 16.36%， 5.68% and 12.55% thinner than FFP（ZXL）， respectively. The differences were extremely significant. The lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes in the three controls A（ZHY）treatment were shorter than those in FFP（ZXL）by 22.34%， 45.83%， 35.48% and 36.96%， respectively. The lengths of the top first， top second and top third basal stem nodes and the total length of the top three basal stem nodes in the three controls B（ZHY）treatment were shorter than those in FFP（ZXL）by 29.94%， 18.12%， 30.21% and 25.94%， respectively. These results provided a theoretical basis for the lodging resistance.
Conclusions
Through the implementation of the agricultural nonpoint source pollution control project and the study on the control mechanism， on the basis of soil testing and formulated fertilization， the formula fertilizer recommended by the World Bank is fully used in the basic farmland in the project area. In rice planting， 5-7 kinds of fertilizers including urea， phosphate fertilizer， compound fertilizer， compound mixed fertilizer and potash fertilizer used by farmers previously are reduced to Kuaimei formula fertilizer （24719）， which meant that the varieties are reduced by more than 80%. At the same time， before the implementation of the project， the farmers generally applied the fertilizers twice， and the amount of fertilizer loss exceeded 20%. Before the project implementation， the average amount of fertilizers used by farmers was 1 050 kg/hm2 （including 300 kg/hm2 of Caihong compound fertilizer for base application， 150 kg/hm2 of urea for weeding， 450 kg/hm2 of Caihong compound fertilizer for tillering and 150 kg/hm2 of Lifeishi potassium fertilizer， equivalent to 537 kg/hm2 of pure fertilizer）. After the implementation of the project， 675 kg/hm2 of the formula fertilizer （24719） （equivalent to 337.5 kg/hm2 of pure fertilizer） is applied， reducing 199.5 kg/hm2 of pure fertilizer. Farmers in the project area adopt the three controls nutrient management （4∶2∶3∶1）of rice， during which fertilization is performed for 4 times （one base and three topdressing）. Furthermore， in the case of three controls B（ZHY）omitting the last fertilization in the three controls fertilization method （4∶2∶3∶1）， which meant that 11.25%of the formula fertilizer was reduced， no difference was caused in yield， but the seed setting rate and 1 000grain weight were still improved by 3.47% and 2.79%， respectively. It can be seen that the three controls fertilization method has obvious material production advantages. Compared with the conventional fertilization method， the basal node of the three controls A（ZHY）fertilization method was shortened as much as 20.52%， which plays an important role in combating typhoon and resisting lodging. In addition， with reference to relevant research and analysis， the "threecontrols" nutrient management can directly improve the absolute fertilizer utilization rate by 13.53%， improve the relative fertilizer utilization rate to 59.06%， and reduce fertilizer loss， thereby effectively suppressing water eutrophication， reducing agricultural nonpoint source pollution and protecting the ecological environment. References
[1] PENG SB， HUANG JL， ZHONG XH， et al. Research Strategy in Improving Fertilizernitrogen Use Efficiency of Irrigated Rice in China[J]. Scientia Agricultura Sinica， 2002， 35（9）： 1095-1103. （in Chinese）
[2] SU ZF， ZHANG J， WANG HB， et al. Study on relationship of tiller development of rice population with the effective ear percentage and rice formulation[J]. Jiangsu Journal of Agricultural Sciences， 1997， 18（1）： 36-40. （in Chinese）
[3] ZHONG XH， HUANG NR， ZHENG HB， et al. Specification for the "Threecontrols" nutrient management technology for irrigated rice[J]. Guangdong Agricultural Sciences， 2007（5）： 13-15， 43. （in Chinese）
[4] HUANG NR， ZHONG XH， ZHENG HB. Application effect of threecontrol fertilization technology for rice[J]. Guangdong Agricultural Sciences， 2007（5）： 16-18. （in Chinese）
[5] ZHONG XH， HUANG NR， ZHENG HB， et al. Some principles for the "Threecontrols" nutrient management technology for irrigated rice[J]. Guangdong Agricultural Sciences， 2007（5）： 19-22. （in Chinese）
[6] WU HR， DING WC， CAI TY， et al. Demonstration of rice "Three controls" nutrient management in Leizhou City[J]. Guangdong Agricultural Sciences， 2008（12）： 79-80， 86. （in Chinese）
[7] HUANG NR， LIANG XM， LI JM， et al. Application of the "Three controls" nutrient management technology of rice at Gaoyao of central Guangdong[J]. Guangdong Agricultural Sciences， 2009（3）： 17-19， 24. （in Chinese）
[8] DAI YX， HUANG NR， LI JY， et al. Demonstration and application of rice "Three controls" nutrient management in Xinhui District[J]. Guangdong Agricultural Sciences， 2009（3）： 20-21， 30. （in Chinese）
[9] HUANG YJ， ZHENG BN， XIE ZS. Application effect of rice "Three controls" nutrient management in Yangjiang City[J]. Guangdong Agricultural Sciences， 2009（3）： 22-24. （in Chinese）
[10] LI LZ， JIANG XZ. Demonstration and application of rice "Three controls" nutrient management in Maonan District[J]. Guangdong Agricultural Sciences， 2009（3）： 25-26， 35. （in Chinese）
[11] XIONG ZG， PAN XX， HU HM， et al. Application effect of rice "Three controls" nutrient management in Qujiang District[J]. Guangdong Agricultural Sciences， 2009（3）： 27-29. （in Chinese）
[12] LIANG GC， LI XT， DU M. Demonstration of rice "Three controls" nutrient management in Shantou City[J]. Guangdong Agricultural Sciences， 2009（3）： 29-30. （in Chinese）
[13] LIU LJ， SANG DZ， LIU CL， et al. Effects of realtime and sitespecific nitrogen managements on rice yield and nitrogen use efficiency[J]. Scientia Agricultura Sinica， 2003， 36（12）： 1456-1461. （in Chinese） [14] PENG XL， LIU YY， LUO SG， et al. Effects of the sitespecific nitrogen management on yield and dry matter accumulation of rice in cold areas of Northeastern China[J]. Scientia Agricultura Sinica， 2006， 39（11）： 2286-2293. （in Chinese）
[15] LIU LJ， XU W， WU CF， et al. Characteristics of growth， development and nutrient uptake in rice under sitespecific nitrogen management[J]. Scientia Agricultura Sinica， 2007， 21（2）： 167-173. （in Chinese）
[16] HE F， HUANG JL， CUI KH， et al. Effect of realtime and sitespecific nitrigen management on rice yield and quality[J]. Scientia Agricultura Sinica， 2007， 40（1）： 123-132. （in Chinese）
[17] LING QH， SU ZF， ZHANG HQ. Relationship between earbearing tiller percentage and population quality and its influential factors[J]. Acta Agronomica Sinica， 1995， 21（4）： 464-469. （in Chinese）
[18] SU ZF， WANG HB， DU YL， et al. Relationship between population quality and yield formation at middle growth stage in rice[J]. Scientia Agricultura Sinica， 1998， 31（5）： 19-25. （in Chinese）
[19] LING QH， ZHANG HC， CAI JZ， et al. Discussion on quality of high yielding rice population and its optimal control[J]. Scientia Agricultura Sinica， 1993， 26（6）： 1-11. （in Chinese）
Editor： Yingzhi GUANG Proofreader： Xinxiu ZHU