(A) Representative actogram showing wheel-running activity of control and Mfn2^–/–mice. Mice were maintained on a 12-hour light/12-hour dark cycle in wheel cages for 16 days prior to release to DD. (B) Free-running endogenous circadian period distributions of control and Mfn2^–/– mice. (C) Representative activity profile graph of wheel-running activity. The activity profile was averaged over 12 days of LD (days 5–16). (D) Phase of activity onset relative to the dark phase and (E) distribution of activity counts in the light or dark periods and total activity in control and Mfn2^–/– mice averaged for days 5–16 in the LD condition. (F) Core body temperature profiles over the entire LD cycle in the control (black) and Mfn2^–/– mice (red). (G and H) Sleep-wake pattern measured during transitions from the light to the dark phase (G) and the dark to the light phase (H). Unfiltered EEG and EMG signals and distribution of automatically detected vigilant states in 1 control mouse and 1 Mfn2^–/– mouse during the last hour of the light or dark phase, and the first 2 hours after shifting in the opposite phase. Panels on the right show extracted signals (framed in blue) in 1-minute time intervals. Red bars over the traces mark wake episodes. Inset in the red rectangle shows the EEG signal with sleep characteristic slow waves. Pie charts show the percentage distribution of wake (black), NREM (white), and REM (gray) episodes in control and Mfn2^–/– mice (n = 5 for each). ZT23–00: control versus Mfn2^–/– mice, P = 0.019 for wake; P = 0.014 for NREM. Unpaired 2-tailed t-test for B, D, and F (right panel); 1-way ANOVA with Tukey’s test for E; Mann Whitney test for F (left panel); multiple unpaired t-tests with Welch corrections with False Discovery Rate multiple comparisons and 2-stage step-up (Benjamini, Krieger, and Yekutieli) method for G and H. **P < 0.01, and ****P < 0.0001. (See also Supplemental Table 1 for details on the statistical information for each graph.)