Table 1. Similarity matrix of trace-element sequences for six cohorts of N. miryae.

	IUI a (28, 14)	IUI b (28, 4)	IUI c (29, 3)	PB a (29, 5)	PB b (29, 7)	NB (29, 7)	Expected
IUI a (08\12\03)	0.80 (0.16)						0.41–0.50
IUI b (16\03\04)	0.37 (0.06)	0.94 (0.06)					0.30–0.66
IUI c (15\03\04)	0.37 (0.05)	0.36 (0.13)	0.89 (0.09)				0.29–0.66
PB a (16\03\04)	0.35 (0.09)	0.44 (0.06)	0.37 (0.10)	1.00 (0.00)			0.36–0.60
PB b (22\03\04)	0.40 (0.06)	0.34 (0.06)	0.25 (0.09)	0.35 (0.06)	0.97 (0.05)		0.39–0.55
NB (26\03\04)	0.37 (0.07)	0.29 (0.08)	0.49 (0.11)	0.33 (0.07)	0.28 (0.06)	0.90 (0.09)	0.38–0.57

Each cell contains the mean similarity (± standard deviation) of sequences that roughly correspond to the first half the PLD. Row headings provide each cohort's name and back-calculated hatch date. Column headings provide, in parentheses, the cohort's PLD and sample size. The 5^th & 95^th percentiles of the expected (by chance) mean for cohort similarity are presented in the far right column (see Methods for details). Highlighted are similarity values that fall above the expected 95^th percentile (right figure of the far right column). The IUI is situated 6 km south of NB, with PB 1 km further south (lower case letters distinguish between multiple cohorts collected in any one of these sites).