Selecting a Peruvian sweetpotato core collection on the basis of morphological, eco-geographical, and disease and pest reaction data

Sweetpotato [Ipomoea batatas (L.) Lam.] ranks among the seven most importan t food crops of the world. The International Potato Center (CIP) holds one of the largest sweetpotato (2n = 6x = 90) gene-banks with more than 5000 cu ltivated accessions from America, Africa, Asia and the Pacific. This collec tion is clonally maintained because it comprises farmer-selected cultivars that have been asexually propagated for many years. Because of this, numero us duplicate accessions of the same cultivar are to be expected. Considerin g that almost 30% of the sweetpotato accessions assembled in this collectio n were from Peru. the first step to select a sweetpotato core collection wa s to identify duplicates in this group. Duplicate identification, using det ailed comparisons of morphological characters and electrophoretic banding p atterns of total proteins and esterases, reduced the number of Peruvian acc essions in the collection from 1939 to 673. The number of duplicates of the same cultivar ranged from 1 to 99 accessions. A Peruvian sweetpotato core collection was selected to enhance the utilization of this germplasm. A tot al of 21 morphological descriptors were scored in all the different Peruvia n cultivars. The unweighted pair-group method using an arithmetic average ( UPGMA) determined the pairwise distance for members of distinct clusters ba sed on these morphological descriptors. A core subset was selected consider ing the square root of the number of accessions for each Peruvian departmen t and respective cluster, as defined by UPGMA. The original core collection consists of 85 accessions (12.6%) from all Peruvian departments, except th at of Madre de Dios,where sweetpotato was never collected, and from all agr oecological zones except Paramo, which has only 0.5% of the accessions of t he entire collection. The sampling for this core collection was appropriate as determined by comparisons of means and frequency distributions for all morphological descriptors. Furthermore, this sampling was validated by the partial assessment of this sweetpotato germplasm for resistance to diseases and pests, tolerance to salt, storage root dry matter content, and vegetat ive period.