The use of cerebrospinal fluid (CSF) analysis for studying in vivo alterations in central neuronal activity requires a relatively sophisticated knowledge of CSF physiology and pathology. Ventriculospinal concentration gradients, circadian rhythms, physical activity, stress, medications, precursor intake, concomitant illness, obstructed CSF circulation, age, and sex alter the baseline neurochemical composition of CSF. Differential probenecid blockade of the egress of acidic monoamine metabolites and cyclic nucleotides from the CSF may complicate interpretations of their accumulations. Degradation of CSF constituents during collection, storage, and analysis may introduce errors in quantification. These sources of CSF variability can be minimized with proper methodology.