Sixty-eight pepper (Capsicum chinense Jacq.) accessions were evaluated for yield (measured as total number of fruits per plant and total fruit weight per plant) and 15 other traits under open-field conditions in two trials, one conducted in the dry (Trial-1) and the other in wet season (Trial-2) of 2014. Each trial was set in a randomized complete block design with three replications (20 plants per replicate) at The University of the West Indies Field Station, Mount Hope, Trinidad and Tobago. There were significant differences (p < 0.01 to 0.001) among the accessions for all traits except plant height, plant canopy width in Trial-1, and early vigour in Trial-2. Total number of fruits per plants was weakly correlated (r = 0.37 in Trial-1, r = 0.45 in Trial-2; p < 0.01 to 0.001) with total fruit weight per plant, but was strongly associated with number of fruits per plant in the first and second 5 pickings (r = 0.78 to 0.97, p < 0.001) and moderately negatively correlated (r = -0.33 to -0.56, p < 0.01 to 0.001) with fruit width and average fruit weight. Total fruit weight per plant was positively associated with fruit weight per plant in the first and second 5 pickings (r = 0.66 to 0.89, p < 0.001), and negatively correlated (r = -0.41 to -0.69, p < 0.001) with days to 50% flowering and fruiting. A comparison of linear regression lines showed that the relative ranking of the accessions over seasons did not change for days to 50% flowering and fruiting, fruit weight per plant in the second 5 pickings, total fruit weight per plant and fruit width. The implications of these results are discussed.