Table 2.

Models containing island geography, current and historical climate as predictors of island biogeographical network roles. The standardized regression coefficients are reported for ordinary least squares regression and reported for both an averaged model based on weighted w_i and minimum adequate models (Diniz-Filho et al. 2008). We also report AIC_c and coefficients of determination (R²) from partial regression models separating the effect of geography, current and historical climate (Rangel et al. 2010). describes total variation explained by island area, elevation and isolation, whereas and reflects the additional (i.e., unique) variation explained by current climate and historical climate-change, respectively. The analysis was conducted separately for Wallacea (n = 119) and the West Indies (n = 57)

	Wallacea						West Indies
	Within module degree l			Among module connectivity r			Within module degree l			Among module connectivity r
	Σ wi	Averaged	MAM	Σ wi	Averaged	MAM	Σ wi	Averaged	MAM	Σ wi	Averaged	MAM
Area	1.00	+0.86	+0.91**	0.64	+0.19	+0.13NS	0.45	+0.23	+0.26NS	0.73	+0.27	+0.30*
Elevation	0.48	−0.12	−0.12NS	0.30	+0.03	–	0.39	+0.20	–	0.24	−0.04	–
Isolation mainland	1.00	−0.22	−0.22**	0.66	+0.15	+0.13NS	0.29	−0.13	–	0.92	+0.34	+0.33**
Isolation landmass	0.67	+0.11	+0.10NS	0.78	+0.18	+0.20*	0.59	+0.22	+0.26*	1.00	−0.76	−0.76**
MAT	0.33	+0.06	–	0.40	+0.13	–	0.98	−0.52	−0.44**	0.67	+0.26	+0.28*
MAP	0.48	+0.09	–	0.33	+0.10	–	0.28	−0.12	–	0.99	+0.58	+0.60**
Anomaly MAT	0.37	+0.06	–	0.29	+0.04	–	0.36	−0.15	–	0.24	+0.01	–
Anomaly MAP	0.46	+0.08	+0.10NS	1.00	−0.50	−0.47**	0.32	−0.14	–	0.25	−0.04	–
AICc			199.2			−180.3			138.4			−114.3
R2			0.72			0.32			0.39			0.67
			0.71			0.11			0.29			0.48
			0.00			0.00			0.10			0.09
			<0.01			0.21			0.00			0.00

^**P < 0.01,

^*P < 0.05, ^NSnonsignificant.