Table 3.

Summary statistics of stepwise multiple regression analyses of likely factors influencing organ longevity in eight Nepenthes species in Brunei, northern Borneo

Model fit	Intercept	Slope	Variance explained (R2)	Prob	Change in R2	Significance of R2 change
Construction cost (CC)	−1·369	12·066	0·696	0·001	–	–
TMA*	3·384	1·419 (1·62, 1·82)	0·179	0·132	–
Tissue density (D)	3·316	2·000	0·332	0·031	–	–
Lignin (L)	−4·958	3·496	0·275	0·05	–	–
Nitrogen (N)	0·573	0·051 (− 2·39, –2·14)	0·001	0·961	–	–
CC + D	−1·094	11·651, 0·15	0·697	0·001	0·001	0·849
CC + L	−1·078	12·364, –0·21	0·697	0·001	0·001	0·886
CC + N	−1·781	13·690, –1·126	0·785	0·001	0·089	0·056
CC + D + L	−0·762	11·95, 0·159, –0·234	0·698	0·006	0·002	0·971
CC + D + L + N	−2·878	12·38, 0·098, 0·899, –1·271	0·795	0·004	0·099	0·293

All traits were log transformed prior to regression analysis in order to compare allometry exponents with the global data set of Wright et al. (2004). In the multiple regressions, each trait was inserted in order of magnitude of its explanatory power (R² value) when only the term is in the model. CC-mass had the highest R² value and was inserted first, followed by the term with the second highest R² value and so on. The 95 % confidence interval values for the global data are given in parentheses in the Slope column for organ longevity vs. nitrogen content and organ longevity vs. tissue mass per unit area.

* TMA, Tissue mass per unit area.