应用群落生态学方法研究了转Bt基因抗虫保铃棉(Gossypium hirsutum)‘972’、‘109B’、‘690’与‘154’种植后对江苏沿海的大丰旱作棉区与沿江的如皋水旱轮作棉区棉田杂草的影响风险。同地同期测定结果显示：不同保铃棉品系棉田杂草物种丰富度(S)、个体数量(N)、群落多样性(H′)、均匀度(J)与优势集中性(C)值在对照品种（‘泗棉3号’与‘苏棉16’）棉田上下波动，优势种及其多度基本接近。两地五期的综合分析结果显示各保铃棉品系杂草群落多样性、均匀度都略高于对照棉，而优势集中性都略低于对照棉。不同抗性保铃棉品系棉田杂草群落特征参数组分对群落多样性的直接间接作用与对照品种棉田的趋势基本一致。同一生态类型棉区的保铃棉高抗、中抗品系与对照品种棉田杂草的种相似性系数与群落相似性系数都较高。上述结果表明，转基因抗虫保铃棉品系棉田的杂草群落稳定性并不比对照品种的差，因保铃棉品系本身产生新的恶性杂草风险的可能性并不明显。
In order to evaluate the ecological safety and planting risk of insect resistant transgenic cotton, the composition and diversity of the weed community in transgenic Bt cotton (four Bollgard strains: ’109B’,’154’,’690’ and ’972’) fields were studied by means of the quantitative analysis method in Dafeng of Jiangsu coastal district and Rugao of Jiangsu Changjiang River bank district. There were marked differences in water content of soil due to differences in crop rotation systems (i.e. rotation with rice following dry-cropping) between cotton fields. The experimental design was as follows: five treatments include four Bollgard strains and traditional variety and four replicates in each location. Each plot area is about 60 m2. The traditional variety check was Simian-3 in Dafeng and Sumian-16 in Rugao respectively. All weed species and their number were investigated with the sampled method of five spots per plot and the spots area investigated per plot was 2.5 m2 in Dafeng and 1.25 m2 in Rugao respectively. The average of weed species and number per square meter for each treatment was calculated. The parameters and components analysis of weed community for each treatment and coefficients of weed species/community similarity between each of four Bollgard strains and traditional variety check were calculated and analyzed also. The results showed: 1) no significant difference in weed species was found between each of four transgenic Bollgard cotton strains and the traditional variety checked in the same location. The dominant species and their abundance were very similar among them.2) the parameters of weed community structure included species richness(S)，total number of individuals(N), community diversity (H′), evenness(J), and concentration(C) in each of the transgenic Bt cotton (four Bollgard strains) fluctuated similarly to the ones in the traditional variety in the same date and same location. The results of comprehensive analysis for two locations showed that the value of community diversity(H′) and evenness(J) in transgenic Bt cotton was a little higher than the ones in traditional cotton check, but the value of concentration(C) in transgenic Bt cotton was a little lower than the ones in traditional variety check. 3) the path analysis of diversity index(H′) with other parameters showed that it all was dominant concentration(C) and evenness(J) that mainly decided the change of diversity index(H′) for weed community in five treatments. Diversity index(H′) would be reduced with increasing of dominant concentration(C) and decreasing of evenness(J) for all treatments.4) the coefficient of weed species similarity(V1) and the coefficient of weed community similarity(V2) between each of four transgenic Bt Bollgard cotton strains and traditional variety check was determined by the Sorensen’s formula and all very high. The value of V 1 is 0.718 0－0.844 5 in Dafeng and 0.489 4-0.603 7 in Rugao，and V2 is 0.451 1-0.598 1 in Dafeng and 0.329 3-0.484 7 in Rugao respectively.From the above results, it is clear that the stability of weed community of transgenic Bt cotton field (four Bollgard cotton strains) is not different significantly than traditional cotton check and its weed risk increased by planting transgenic Bt cotton is not obvious. There was a close correlation among the structure of weed community, crop rotation system and geographical factors. The main factor that affected the weed occurrence and distribution pattern of weed infestation and community in the cotton fields was the difference of soil water content, which resulted from the different cropping system.
全 文 ：