Table 1. Linear-regression analysis of data points from different time slices using a sliding window with a width of four orders of magnitude, or with a shrinking window for the oldest calibrations.

Significant regression results (p ≤ 0.05) are shown in bold. A more detailed version of this table is available as Table S4.

		Calibration ages
		100−104 years	101−105 years	102−106 years	103−107 years	104−108 years	105−109 years	106−109 years	107−109 years	108−109 years
Coding markers	N	1	9	20	132	176	173	162	50	5
	R 2	0	0.05	0.24	0.31	0.23	0.09	0.04	0.11	0.04
	P-value	n/a	5.67 × 10−1	2.67×10 −2	3.43×10 −12	1.48×10 −11	4.79×10 −5	1.26×10 −2	1.76×10 −2	7.6 × 10−1
	Slope	n/a	−0.33	−0.34	−0.37	−0.29	−0.20	−0.15	−0.34	0.36
	Non-SSC a	n/a	–	***	***	***	***	***	***	–
Non-coding markers	N	8	25	27	43	48	31	29	12	1
	R 2	0.12	0.27	0.33	0.62	0.55	0.27	0.17	0.16	0
	P-value	4.09 × 10−1	7.41×10 −3	1.80×10 −3	4.47×10 −10	1.41×10 −9	2.58×10 −3	2.55×10 −2	2.00 × 10−1	n/a
	Slope	0.08	−0.34	−0.36	−0.48	−0.49	−0.58	−0.53	−0.89	n/a
	Non-SSC a	**	***	***	***	***	*	*	–	n/a

Notes.

^aResults of tests against spurious self-correlation (non-SSC). Higher p-value from the two tests is shown.

^*p < 0.05.

^**p < 0.005.

^***p < 0.0005.