Table 1. Tests of demographic changes based on ITS2 sequences.

	South PacificOcean	South West Pacific				South East Pacific Chile
		New Zealand	West NewZealand	East NewZealand$	Chatham
Sample size	201	93	30	51	12	108
F S (P value)	−5.139 (0.05)	−1.749 (0.40)	−3.704 (0.00)	−1.422 (0.19)	−2.373 (0.04)	−5.342 (0.00)
R2 (P value)	0.044 (0.11)	0.084 (0.42)	0.091 (0.11)	0.083 (0.26)	0.126 (0.09)	0.043 (0.16)
Spatial Expansion:
SSD (P value)‡	0.029 (0.51)	0.049 (0.25)	0.004 (0.40)	0.019 (0.00)	0.0003 (0.99)	0.0005 (0.14)
Rag (P value)‡	0.132 (0.72)	0.100 (0.59)	0.359 (0.62)	0.172 (0.04)	0.018 (0.99)	0.359 (0.51)
τ (90% CI)	8.771	7.104	4.833	nc#	1.344	0.260
	(0.303–13.897)	(0.809–12.855)	(0.000–7.295)		(0.728–3.974)	(0.093–0.610)

Departure from neutrality tested using Fu’s Fs [42] and Ramos-Onsins and Rozas’ R₂ statistic [43]. Significance determined using 10,000 simulated data sets implemented by DnaSP version 5.10.01 [44]. τ-values estimated from demographic expansion models using ARLEQUIN v 3.11 [33]. Goodness of fit tests for a model of population expansion calculated from the sum of squared deviation (SSD) and the Harpending’s raggedness index (Rag). Significance assessed by bootstrapping (10,000 replicates; ARLEQUIN v 3.11) [33].

^$All East New Zealand populations excluding Chatham;

^‡P values of (Expected SSD > Observed SSD) or P (Expected Rag > Observed Rag) superior to 0.05 means the null hypothesis of spatial expansion can’t be rejected;

^#nc: τ-value not calculated when spatial expansion is rejected.