OBJECTIVEThe genetic diversity and genetic relationship of Jatropha curcas resources in Sichuan and Yunnan were studied in order to provide a theoretical basis for breeding fine varieties and protecting germplasm resources.

METHODTen J. curcas populations were studied by 12 cpSSR primers in this paper. On the base of amplified bands, genetic diversity parameters were analyzed by POPGENE version 1.32. Furthermore, UPGMA tree of 10 J. curcas populations established from pairwise population distance by NTSYSpc version 2.10.

RESULTTwenty-two polymorphic bands were detected, and the percentage of polymorphic loci (P) was 76.28%. Among of the 10 J. curcas populations, the average percentage of polymorphic loci of YNSB was higher than that of the other populations, and it reached 95.45%; On the other hand, that of YNLS was the lowest in all populations, and it was 45.45%. Nei's gene diversity index(H(e)), Shannon information index(I), Effective Num of alleles(A(e)) were respectively 0.4020, 0.576 7, 1.713 6. The total gene diversity (H(T)), the gene differentiation coefficient (G(st)), the gene flow (N(m)) and the gene diversity within populations (H(s)) were 0.443 3, 0.080 2, 3.058 5, 0.405 1, 0.035 7, respectively. The highest gene diversity ratio was showed within populations and the lowest among populations. The results by AMOVA analysis showed that 91.02% of genetic variation existed within populations while 8.98% of genetic variation existed among populations. On the base of the results, the conclusion was extracted that variation existed mainly within populations, and the variation within populations was bigger than that among populations. The result was consistent with that of the gene differentiation coefficient. The order of the genetic diversity was YNLS population < XSBN population < SCHPZ population < SCHD population < SCJH population < YNPR population < SCLB population < YNSB population < YNFY population < SCHL population. The range of Nei's genetic identity and genetic distance of 10 respectively populations were respectively 0.812 7-0.979 8, 0.020 4-0.207 3. All these showed the similarity was higher and there was a close relationship among the 10 respectively populations; Results based on the cluster analysis showed that 10 respectively populations were divided into 2 groups: one was SCJH population and CHPZ population, the other was SCHL population, SCHD population, SCLB population, YNSB population, YNFY population, YNPR population, XSBN population and YNLS population.

CONCLUSIONSignificant genetic diversity was observed among respectively resources in Sichuan and Yunnan. On the other hand, genetic relationship was close between populations.