Measures of single unit activity that quantify the irregularity of spike activity (log interval entropy; ENT) and pattern (mutual information between adjacent log intervals; MUT) as well as mean spike frequency (MSF) were applied to recordings from the supraoptic nucleus (SON) at all stages of the day/night cycle in urethane-anaesthetised rats. C-association methods were developed to assess the significance of the rhythms of activity revealed. Continuous cells in intact animals showed significant rhythms for MSF (P=0.001) and ENT (P=0.001) but not MUT. After osmotic stimulation a significant rhythm was seen in all three parameters as it was after pinealectomy (carried out under halothane anaesthesia > 4 weeks before recording) and after osmotic stimulation in pinealectomised animals. Phasic cells also showed statistically significant daily rhythms in MSF (P=0.001), ENT (P=0.016) and (MUT P=0.001). After osmotic stimulation significant rhythms (P=0.001) remained for all three parameters. Rhythms remained after pinealectomy for MSF and ENT (but not MUT) although all three parameters showed significant rhythms after osmotic stimulation. Cosinor analysis showed that, for ENT, the amplitude of the rhythm was significantly reduced in continuous (but not phasic) cells after osmotic stimulation (P=0.002, t =3.1021). A reduction was also seen after pinealectomy (P=0.001, t=3.7004). There was no difference between the mesor (the value midway between the highest and the lowest values of the cosine function best fitting the data.) values for these groups. Similar cosinor analysis showed that while there were no differences in the amplitude of the rhythms in phasic cells there were significant differences in the mesor values between intact and osmotically stimulated animals (P = 0.001, t=3.763) and after pinealectomy (p=0.001, t=5.3005). No such differences were seen in continuous cells. Cosinor analysis of the ENT values also revealed that the amplitude of the rhythms was significantly greater (P=0.002, t=3.837) in continuous than in phasic cells. A significant difference was also seen between the timing of the peak in continuous cells following pinealectomy (from ZT 2249 to ZT 02.30, P=0.034, t=2.1274). The Mann-Witney test showed that after pinealectomy the MSF of both phasic and continuous cells was significantly increased (P=0.001). C-association methods also revealed daily rhythms of MSF and ENT (P=0.001) in the suprachiasmatic nucleus (SCN) in vivo. Cosinor analysis of the entropy values in intact and pinealectomised animals showed that the amplitude of the rhythm was significantly greater (t=1.9901; P=0.047) in intact (mesor 0.3525 bits) than in pinealectomised animals (mesor 0.1786 bits). We have shown for the first time that there is a daily rhythm of electrical activity in the SON in vivo and that pinealectomy profoundly affected both SON and SCN. Particular attention should thus be given to the time of day when interpreting recording studies in the hypothalamus.