Reports of the frequency of pulsatile LH release in orchidectomized rats are surprisingly variable. Estimates of the period between LH pulses vary from 18.5 to 250 min in nine reports from six laboratories, published in 1981-1983. In these studies, blood samples were drawn at intervals ranging from 2.5 to 60 min. We examined the relationship between estimated LH pulse frequency and blood-sampling interval. Six castrated male rats were cannulated, and blood samples were drawn at 2.5-min intervals through 4 h. Plasma LH levels were determined by radioimmunoassay. Using the data obtained at 2.5-min intervals, we simulated blood-sampling intervals of 5, 7.5, 10, 12.5, and 15 min by sequentially deleting data points. The original and simulated data sets were analyzed by both the PULSAR and Cycle Detector computer programs. The results show that as the sampling interval increased from 2.5 to 15 min, the apparent period of pulsatile LH release rose steadily from about 20 min to about 100 min. We found a high, positive correlation between blood-sampling interval and the apparent period of LH pulses. Estimates of LH pulse frequency in castrated male rats vary directly with the frequency at which blood samples are taken. Sampling intervals greater than 5 min in orchidectomized rats yield an LH pulse period that is most likely exaggerated.