Abstract: The characterization of diversity in germplasm collection is important to plant breeders to utilize and to the genebank curators to manage the collection efficiently.The available large variability contained in germplasm accessions has not been adequately utilized in the peanut improvement programs which is mainly due to lack of information on important agronomic and economic traits,which require extensive evaluation.The development of core collection could facilitate easier access to peanut genetic resources,enhance their use in crop improvement programs,and simplify the genebank management.This paper describes the development of a core collection from 6390 accessions of peanut descriptor traits available from China genebank.The germplasm accessions were stratified by botanical types and then grouped by ecological origin within each of botanical varieties.Data on 15 morphological and seed chemical traits were used for clustering by SAS method.From each cluster,5 percent to 10 percent accessions were either randomly or compulsorily selected to constitute a core collection consisting of 576 accessions.The accessions widely grown in production or extensively used in breeding programs must be included in the core collection.The rest of the accessions included were randomly selected.Comparisons using mean,range,standard deviation,coefficient of variation and diversity index on different descriptors indicated that the genetic variation available for these traits in the entire collection has been preserved in the core collection.This core collection provides an effective mechanism for the proper exploitation of peanut germplasm resources for the genetic improvement of this crop.