In spite of recent advantages in molecular genetics, electromyography (EMG) still has diagnostic value, especially in children. Neuromuscular disorders in children are not common compared to in adults but they are important. Many of them are disabling disorders in which there is no cure, only supportive treatment. Such neuromuscular disorders affect not only the patient, but the whole family, often because of increased care load. In addition to these chronic disorders, of which many have a genetic background, there are acute and subacute traumatic disorders and immunopathies. In children, every medical procedure may induce anxiety. It is therefore important that one or both parents are present at the investigation in order to support their child. It is advisable to apply a local anaesthetic prior to EMG. Pain relief may be needed to give before EMG (e.g. morphine). Controlled conscious sedation may sometimes be needed (e.g. midazolam, ketamine, nitrous oxide). In the majority of children, conventional needle EMG is used in order to study the motor unit properties. In a normal muscle at rest no abnormal spontaneous activity is identified. Thus, after the insertion activity has subsided, the muscle at rest is investigated for abnormal spontaneous activity which may occur in single muscle fibres or in a group of fibres. There may be abnormal spontaneous activity of different kinds in one and the same diseased muscle of neurogenic or myogenic origin. Ten different places in each muscle are examined, and the amount of abnormal spontaneous activity in these ten needle positions is measured. At slight voluntary activation motor unit potentilas (MUPs) are recruited. Preferably, about 20 MUPs are collected from each muscle in order to allow a thorough assessment. However, in children with severe disorders, and who activate only sporadically, it may not be possible to obtain more than a few MUPs. The MUPs are collected by quantitative method, sometimes mixed with a semiquantitative method, depending on the child’s co-operation. For each MUP there are different parameters to consider, e.g. amplitude, duration, shape and stability. Strong voluntary activation is used to study the interference pattern. It is difficult to perform this adequately in children, but, when feasible, it gives much information and complements the MUP analysis. Single-fbre EMG is a very useful and sensitive tool in the diagnosis of neuromuscular transmission disorders. It can be studied under voluntary activation or, particularly in children, with electrical stimulation. The oral communication will further elaborate on EMG and give practical tips.