Genetic diversity of 36 rice entries from the United States Department of Agriculture(USDA)rice collection was assessed using 103 ILP(intron length polymorphism)and 54 SSR(simple sequence repeats)markers.A total of 236 and 332 alleles were detected by the ILP and SSR markers,respectively.On average,the SSR markers produced higher polymorphism information content value and number of alleles than the ILP markers.Whereas the Nei’s genetic distance measured using the SSR markers was much higher than that measured using the ILP markers.Mantel’s test indicated that there was a statistically significant correlation(r=0.827,P<0.001)between the two marker systems.UPGMA clustering based on the ILP and SSR markers resulted in consensus dendrograms.The cophenetic correlation coefficient(r=0.918,0.878 and 0.924,P<0.001 for the ILP,SSR and combined markers,respectively)showed a highly accurate dendrogram represented the genetic distance among these entries.The 36 entries were divided into four groups.Four African Oryza glaberrima accessions were clustered within a distinct group(Ⅰ),and the remaining entries were separated into three groups(Ⅱ,Ⅲ and IV).All the entries could be also clustered into two main groups:One was composed of Ⅲ and IV,considered as indica group, and the other was composed of Ⅰ(O.glaberrima)and Ⅱ(japonica-like).Model-based cluster analysis revealed that the japonica-like group maintained very pure ancestry while the indica group shared mixed ancestry,especially for group Ⅲ, which had seven admixtures sharing from 19.5%to 30.0%of ancestry with group Ⅳ(based on SSR markers).It is suggested that ILP and SSR markers could be very useful for the genetic study and breeding in rice. Genetic diversity of 36 rice entries from the United States Department of Agriculture (USDA) rice collection was assessed using 103 ILP (intron length polymorphism) and 54 SSR (simple sequence repeats) markers. A total of 236 and 332 alleles were detected by the ILP and SSR markers, respectively. Average average, the SSR markers produced higher polymorphism information content value and number of alleles than the ILP markers. Here the the Nei’s genetic distance measured using the SSR markers was much higher than that measured using the ILP markers. Mantel’s test indicated that there was a significant correlation (r = 0.827, P <0.001) between the two marker systems. UPGMA clustering based on the ILP and SSR markers resulted in consensus dendrograms.The cophenetic correlation coefficient (r = 0.918, 0.878 and 0.924, P <0.001 for the ILP, SSR and combined markers, respectively) showed a highly accurate dendrogram represented the genetic distance among these entries. 36 entries were divided into four groups s.Four African Oryza glaberrima accessions were clustered within a distinct group (Ⅰ), and the remaining entries were separated into three groups (Ⅱ, Ⅲ and IV). All the entries could be clustered into two main groups: One was composed of Ⅲ and IV, considered as indica group, and the other was composed of Ⅰ (O.glaberrima) and Ⅱ (japonica-like). Model-based cluster analysis revealed that the japonica-like group maintained very pure ancestry while the indica group shared mixed ancestry, especially for group III, which had seven admixtures sharing from 19.5% to 30.0% of ancestry with group IV (based on SSR markers) .It is suggested that ILP and SSR markers could be very useful for the genetic study and breeding in rice.