Electromyographic (EMG) recording was performed synchronously from the levator palpebrae superioris (LP) and the orbicularis oculi (OO) muscles in 28 patients referred to us for treatment of blepharospasm with botulinum A toxin. At the time of this study, 19 patients were under the treatment with botulinum, four started treatment shortly after the EMG recording and five patients had not yet been treated. Based on the EMG patterns, we were able to classify five major groups of abnormalities. Group 1 (blepharospasm): consisted of 10 patients with dystonic discharges limited to OO, normal LP tonic activity, intact reciprocal inhibition between LP and OO and dense bursts of action potentials with high amplitude preceding the return of LP tonic activity, i.e. ‘postinhibition potentiation’ of LP, brought about by a brief contraction of OO. Group 2 (combined dystonic activities of LP and OO): seven patients belonged to this group. The EMG recording revealed alternating tremulous discharges in both LP and OO muscles, and short intervals of co-contractions due to moderately disturbed reciprocal inhibition. Group 3 (combination of blepharospasm, LP motor impersistence): the EMG patterns, observed in three patients, were characterized by a gradual cessation of LP activity, followed by a brief contraction of OO, which facilitated the return of LP activity, resulting in opening of the eyes. The EMG recordings, thus, revealed the crucial, beneficial role of postinhibition potentiation as a compensatory mechanism in this type of eyelid movement disorder. The EMG patterns were also characterized by short or prolonged periods of dystonic discharges limited to the OO muscles. Group 4 (combination of blepharospasm, involuntary LP inhibition): this group consisted of four patients. In addition to episodes of dystonic activities of OO, the EMG also showed some periods of involuntary inhibition of LP without any concomitant activities of OO. Two patients also exhibited a failure of inhibition of OO muscle activity, following the voluntary contraction of this muscle. The postinhibition potentiation was often not observed. Group 5 (involuntary LP inhibition): consisted of four patients with EMG patterns of involuntary inhibition of LP activity, without any dystonic discharges in OO. The postinhibition potentiation was not observed in this group. The response to the treatment with botulinum toxin was good in the first group and gradually worsened towards the fifth group. Application of botulinum into multiple sites of OO, especially its pretarsal portion, resulted in better response to the treatment in the second and fourth groups. We conclude that synchronous EMG recording of LP and OO is an indispensable investigation method to establish the origin of varying forms of eyelid movement disorders in patients with clinical symptoms of blepharospasm, and to analyse the cause of unsatisfactory response to the treatment with botulinum.