It is generally believed that there is a scarcity of muscle spindles in the diaphragm and that there is no tonic activity at end expiration. This conclusion is based mainly on animal studies and the difficulty in differentiating tonic electromyogram activity from noise. We have, however, found a number of muscle spindles in the newborn human diagphragm, concentrated in the region of the central tendon. We also tried to detect tonic activity by decreasing it (by rapid-eye movement (REM) sleep or anesthesia) or increasing it (with abdominal loading). During REM sleep in five infants and five adults, using subcostal electrodes were observed a marked fall in tonic activity (P less than 0.001) compared to non-REM or quiet sleep. We also observed a reduction in diaphragmatic tonic activity with halothane anesthesia (P less than 0.001). With esophageal electrodes in adult subjects, there was a rise in tonic diaphragmatic activity proportional to the amount of abdomina load (P less than 0.001). We conclude that there are muscle spindles in the human diaphragm and that there is tonic activity at end expiration.