Table 2. The results of generalized linear mixed effects models (GLMM) testing the dependence of pollen germination and pollen tube length on different temperature levels and elevation.

	Pollen germination				Pollen tube length
Predictors (fixed effects)	Estimate	SE	z	P	Estimate	SE	t	P
Intercept	−0.018	0.098	−0.19	0.8524	−1.305	0.071	−18.38	<2 * 10− 16
Temperature 25	−0.053	0.065	−0.82	0.4118	−0.136	0.039	−3.49	0.0005
Temperature 30	−0.081	0.063	−1.30	0.1936	−0.425	0.038	−11.18	<2 * 10− 16
Temperature 40	−0.539	0.06	−8.92	<2 * 10− 16	−0.546	0.039	−14.17	<2 * 10− 16
Elevation highland	0.307	0.114	2.70	0.0069	−0.289	0.078	−3.71	0.0002
Temperature 25: Elevation highland	−0.086	0.090	−0.96	0.3390	0.026	0.051	0.50	0.6154
Temperature 30: Elevation highland	−0.369	0.088	−4.19	2.9 * 10− 5	0.182	0.052	3.51	0.0004
Temperature 40: Elevation highland	0.079	0.082	0.96	0.3388	0.093	0.050	1.88	0.0600

Note:

Parameter estimates, their standard errors (SE), z or t values, and P values are provided for fixed effects of the elevation (lowland x highland), temperature (four levels), and their interaction. Values of parameter estimates are provided at the scale of the linear predictors. Logit link function was used in the binomial GLMM testing the effect of elevation and temperature on pollen germination and log link function was used in the GLMM testing the effect of the predictors on pollen tube length. P values < 0.05 are highlighted in bold.