The effects of seven constant temperatures (10–40°C at 5°C intervals) and seven after-harvest periods (30–540 days after harvest) were evaluated on seed germination of nine Amaranthus species (A. albus, A. blitoides, A. cruentus, A. deflexus, A. graecizans, A. hybridus, A. lividus, A. retroflexus and A. viridis). Seeds of A. blitoides and A. viridis were also tested at alternating temperatures of 10/30°C (12/12 h thermoperiod) in continuous darkness and in an alternating 12/12 h dark/light photoperiod. With the exception of A. blitoides and A. viridis, germination increased as temperature increased from 20 to 35°C; the latter representing the optimum temperature (70–100% germination). At 10 and 15°C constant temperatures, no significant seed germination occurred in A. albus, A. deflexus, A. graecizans and A. lividus, while in A. cruentus, A. hybridus and A. retroflexus there was no germination at 10°C, but at 15°C more than 60% germination occurred. Germination was influenced strongly by after-ripening period in A. cruentus, A. hybridus and A. retroflexus, partially in A. deflexus, and barely in A. graecizans and A. lividus. Seeds of A. blitoides and A. viridis required alternating temperatures and light to achieve high germination percentage (>90%). Primary dormancy in Amaranthus plays a fundamental role in extending germination over a longer period, so that the probability of seedling survival is maximised. The present study adds to the understanding of the environmental control and germination ecology of Amaranthus species and provides data that can contribute to predicting weed emergence dynamics.