Table 2.

Parameters of cosinor regressions.

		Plasma leptin			Plasma cortisol			Plasma insulin			Plasma glucose
		Mean	SEM	P	Mean	SEM	P	Mean	SEM	P	Mean	SEM	P
LP Sham (n = 23)	a	7.06	0.49	<0.001	5.03	0.71	<0.001	1.79	0.22	<0.001	1.02	0.05	<0.001
	b	2.18	0. 70	0.006	1.46	1.02	0.17	0.21	0.32	0.53	0.04	0.08	0.62
	c	21.34*	1.22	<0.001	13.43	2.60	<0.001	6.68	5.36	0.27	14.60	7.20	<0.001
LP PinX (n = 21)	a	6.95	0.67	<0.001	4.72	0.65	<0.001	1.78	0.10	<0.001	1.24	0.03	<0.001
	b	0.75	0.95	0.43	2.66	0.92	0.009	0.38	0.15	0.017	0.05	0.04	0.19
	c	6.75	4.80	0.17	14.08*	1.32	<0.001	4.84*	1.48	0.004	13.84	2.81	<0.001
SP Sham (n = 25)	a	8.48	0.61	<0.001	8.27	0.74	<0.001	1.32	0.09	<0.001	1.12	0.05	<0.001
	b	2.45	0.85	0.01	2.82	1.09	0.015	0.12	0.13	0.37	0.11	0.06	0.09
	c	13.22#	1.34	<0.001	14.26*	1.36	<0.001	18.92	4.22	0.11	18.80	2.36	<0.001
SP PinX (n = 28)	a	8.70	0.05	<0.001	6.01	0.88	<0.001	1.81	0.20	<0.001	1.00	0.035	<0.001
	b	0.11	0.06	0.91	3.30	1.21	0.011	0.61	0.28	0.037	0.10	0.05	0.041
	c	8.81	2.36	0.81	8.67#	1.45	<0.001	20.18#	1.79	<0.001	12.96	1.81	<0.001

The table shows the three parameters of cosinor regression, including a the mean level, b the amplitude, and c the acrophase of the rhythm (see Material and Methods for details). For the acrophase, the reference time is Zeitgeber 0 (i.e., lights on) for both photoperiods. LP, long photoperiod; SP, short photoperiod; Sham, sham-operated; PinX, pinealectomized. Italicized parameters are not statistically significant (P > 0.05). For a given hormone (column), mean acrophases lacking common superscripts (

^*or

^#) are significantly different (t-test or One-Way ANOVA for 2 or more groups, respectively). No superscript is included in case the overall regression was not significant due to non-significant acrophase (c) and/or amplitude (b), or when there was only one significant regression precluding statistical comparison with the other groups (as in the case of plasma glucose).