In order to investigate the genetic diversity of 65 lines of pepper, collected from different regions of the country and also from China and the International Research Center for Vegetables (AVRDC), a randomized complete block design with 3 replications was done. Results of analysis of variance showed a significant difference among genotypes at 1% probability level for all traits. Based on the mean comparison of different traits using Tukey's method at 5% probability level, genotypes 3095, 0013 and 26-6 were better in terms of yield and characteristics associated with marketable products. To determine the genetic distance and grouping of genotypes, factor and cluster analysis performed. To understand the internal relations of traits and to determine the group of variables with the highest correlation, factor analysis performed using principal component analysis were through varimax rotation. In these analysis, four independent factors justified 74.7% of the variation. Cluster analysis was done using Ward method, and the lines were gathered into six different groups. So that, this grouping was associated with geographic distribution of lines and some of morphological traits that associated with marketability. Results of stepwise regression showed that three traits of stem diameter, stem length and number of fruits per plant had the greatest effect on fruit yield. To evaluate the direct and indirect effects of these traits on fruit yield, Path analysis performed. Results showed that the three mentioned traits had direct effect and indirect effect of these traits was very low. Given that in this experiment lines under study showed significant differences in terms of genetic variability for traits under evaluation. It was concluded in order to benefit from heterosis, selection of superior genotypes with greater genetic distance from each other can be useful in pepper hybrid variety production.