The use of diathermy to achieve haemostasis after tonsillectomy remains controversial. We have reviewed the English language literature, and found no convincing evidence that diathermy is any more likely to cause post-operative haemorrhage than the use of ligatures. The results of a prospective, randomized study of 1036 consecutive tonsillectomies are presented. No significant difference was found in post-operative haemorrhage rates when either diathermy or ligatures were used. Diathermy was found to reduce operating time compared to ligatures. The possibilities for day-case tonsillectomy are discussed.