Magnetoencephalography (MEG) is a completely noninvasive method of functional imaging. MEG performs noninvasive functional imaging by recording the magnetic flux on the head surface associated with electrical currents in activated set of neurons, estimating the location of such sets, and projecting the location onto the MRI of the brain to identify and visualize the activated brain region. MEG has rapidly evolved in the last two decades due to the introduction of whole head systems and advances in computer technology. MEG is now the imaging modality of choice where a precise and high degree of localization is required. Ongoing studies show that it provides superior temporal and spatial resolution when compared to functional MRI. MEG is the only imaging technique that can reveal brain function over millisecond intervals. Magnetoencephalography was initially used to localize the primary sensory cortices, and depending on the nature of stimulus, this has been validated for visual, auditory or somatosensory areas. In order to localize brain networks involved during the engagement of cognitive tasks, both temporal and spatial resolution are critical. MEG is the only imaging technology capable of providing this information.We have successfully used magnetoencephalography to noninvasively localize brain areas involved with key language functions. These have been validated through the Wada procedure and with direct electrocortical stimulation. The utility of MEG in noninvasively localizing language function is reviewed. MEG also allows us to understand the differences in functional organization of the brain underlying the reading performances of dyslexic children.We have recently developed a protocol using magnetoencephalography to assess memory function (independent hippocampal activation). This is in the process of being validated by comparison with the Wada procedure.Antepartum knowledge of fetal health status is essential for early identification of fetal risk for injury or death. Most modern techniques monitor the fetal heart rate and use ultrasound to assess fetal movement. We propose that magnetoencephalography, with a new system design that provides good access to the fetal head, will allow assessment of fetal brain activity. Early results show that fetal auditory-evoked potentials can be recorded reliably.Magnetoencephalography has now been demonstrated to be capable of noninvasively mapping primary sensory functions, language function, and soon memory and other affective cognitive stimuli. Additionally, it appears as though this procedure can be extended to the fetus to allow assessment of fetal brain activity. As we continue to use magnetoencephalography to define the substrates for normal brain activity, we will be able to critically evaluate neurologic problems that interrupt or impair these functions. Hopefully, magnetoencephalography will allow early objective identification of these impairments and treatment protocols to be developed.