A greater understanding of radiobiology led to the development of two classes of radiation fractionation schedules for the treatment of head and neck cancers. The aim of accelerated fractionation is to reduce tumour proliferation, which is a major cause of relapse, by shortening the total duration of radiotherapy. By contrast, hyperfractionation exploits the differential sensitivity of tumour cells and normal tissues to radiation, to increase the therapeutic gain. The results of clinical trials of various types of altered fractionation schedules in head and neck carcinomas are examined in this review. Acceleration of radiation by 1 week without dose reduction and hyperfractionation are consistently better than standard fractionation for locoregional control of intermediate to advanced carcinomas without an increase in late toxic effects. However, improvement in survival of patients has not been consistent. Clinical investigations show that improvement in locoregional disease control and consistent gain in survival have been achieved with combinations of radiotherapy and concurrent chemotherapy in patients with mostly stage IV carcinomas. However, these benefits have been at the expense of increased late morbidity. Consequently, concurrent radiochemotherapy is now preferred for non-surgical treatment of patients with locally advanced carcinomas, whereas altered fractionation is generally selected for patients with intermediate-stage tumours or who are medically unfit to receive chemotherapy. Further data is needed before the combination of altered fractionation with chemotherapy can be recommended outside of a study setting.