The chickpea (Cicer arietinum L.) is usually grown under rainfed, rather than irrigated conditions, where drought accompanied by heat stress is a major growth constraint. The aim of this study was to select chickpea genotypes having resistance to drought/heat stress and to identify the most appropriate selection criteria for this. A total of 377 chickpea accessions were sown 2 months later than normal for the Antalya region (Turkey) to increase their exposure to the drought and high-temperature conditions of a typical summer in this part of the world. Interspersed between every 10 test genotypes as benchmark genotypes, were plants of the two known genotypes ILC 3279 (drought-susceptible) and ILC 8617 (drought-susceptible), while ICC 4958 (known drought-resistant) and ICCV 96029 (known very early, double-podded) were also sown for confirmation. All plants were subsequently screened for drought and heat stress resistance. Soon after the two known susceptible genotypes had died, evaluations of the entire trial were made visually on a scale from '1' (free from drought/heat damage) to '9' (all plants died from drought/heat). Yield loss in many of the test genotypes and in the two known susceptible genotypes (ILC 3279 and ILC 8617) rose to 100 %. The desi chickpeas (smaller, dark seeds) were generally more drought- and heat-resistant than the kabuli chickpeas (larger, pale seeds). Two desi chickpeas, ACC 316 and ACC 317, were selected for drought and heat (> 40 degrees C) resistance under field conditions. Seed weight was the trait least affected by adverse environmental conditions and having the highest heritability, and it should be used in early breeding selections. When breeding drought- and heat-resistant chickpeas, path and multivariate analyses showed that days to the first flowering and maturity to escape terminal drought and heat stresses should be evaluated ahead of many other phenological traits, and harvest index, biological yield and pods per plant for increased yield should also be considered.