Aim: To compare the effects of four stocking densities (5, 10, 15 or 20 birds per m2) on the performance, carcass characteristics and selected welfare indicators of broiler chickens. Methods: A total of 756 day-old male broiler chicks were assigned as six replicates of 13, 25, 38 and 50 birds per pen, to stocking densities of 2,000, 1,000, 667 and 500 cm2 floor space per bird, respectively. The birds were housed in floor pens in an environmentally controlled room. Normal commercial diets were fed ad libitum and fresh water was freely available throughout the 5-week study period. Liveweights of birds and feed intake were recorded at weekly intervals. Data on carcass parameters and welfare indicators (moisture, gait scores, feather scores, breast and hip lesions, and foot pad and hock burns) were obtained on Day 35. Adrenal glands were removed at slaughter, weighed and examined histologically to determine the total proportion of cortical and medullary tissues. Results: During the study, the birds at the lowest density grew faster (p<0.001−0.01) and consumed more (p<0.001− 0.01) feed than those at the other three stocking densities. The liveweight gains and feed intake of birds reared at densities of 10, 15 and 20 birds per m2 were similar (p=0.11−0.83 and p=0.17−0.57, respectively). Stocking density had no influence on the feed/weight gain, mortality or carcass characteristics. No breast or hip lesions were observed in carcasses from any of the groups. The weight and structure of the adrenal glands were similar across all groups although the staining of medullary tissue of birds was most intense in the highest stocking density group. As stocking density increased, so did moisture and gait score, hock and foot pad burn scores, and the feather score was adversely affected. Conclusions: The data showed that the relationship between stocking density and productivity traits was not linear in nature, in accordance with other studies. Under the conditions of the current study, the performance, mortality and carcass characteristics of broilers grown at densities of 10, 15 and 20 birds per m2 were similar. There were some welfare implications at high population densities as indicated by adverse effects on litter quality, gait scores and the incidence of hock and foot pad burns.