1. Circadian locomotor rhythms were measured continuously for 16 months in 27 animals which were exposed during most of this time to the natural daily cycle of illumination (ND) at 47°58′ N. Lat., and held at 19 °C. However, at two solstices (June and December) and two equinoxes (March and September) the animals were transferred to continuous illumination (LL) for one month, where two groups of each species were exposed to two different intensities of LL, and then returned to ND. 2. Circadian period (τ) was virtually constant throughout the year.Dipodomys merriami showed the positive correlation between τ and light intensity predicted by Aschoff's rule, but mean τ inAmmospermophilus leucurus was essentially unchanged over a wide range of light intensities. The theoretically predicted seasonal relationship between τ and phase relationship of activity to Zeitgeber signals (ψ) was not found. 3. The duration of activity (α) inA. leucurus was unusual in showing little seasonal change in ND, despite marked seasonal changes in photoperiod. Perhaps due to this relative constancy of α in ND (αND), the change of α upon transition into LL was generally not significant. There were also no significant differences between any of the four seasonal values of α in LL (αLL). InD. merriami αND generally paralleled seasonal changes in length of night, and change of α upon transition into LL was significant at all seasons. Mean αLL fell within a narrow range over the year, during which there were almost no significant season-to-season differences. 4. Neither species showed a marked seasonal cycle of body weight. However, body weight ofA. leucurus consistently increased upon each transfer from ND to LL. 5. MaleA. leucurus, investigated by external examination, showed an annual gonadal cycle with a peak in spring, whereas mostD. merriami developed and then maintained testicular maturity throughout the study. 6. The role of τ in seasonal homeostasis of the circadian system may lie in its availability as a constant reference signal. It cannot be directly responsible in these two species for stimulation of adaptive seasonal functions. However, complementary changes in either α orψ may interact with τ to promote some component of adaptive seasonal cyclic functions. The apparent endogenous circannual gonadal cycle ofA. leucurus could be either generated or entrained via such interactions, but the opportunistically responsive gonadal function ofD. merriami probably has relatively little relation to such circadian functions.