Table 4.

Multi-factor meta-regression of studies having data for each of the five regression moderators.

Moderator	ptest	ptest	ptest
	Carbon exchange rate	Stomatal conductance	Transpiration rate
Root colonization	<0.001	0.020	0.601
Shoot DW ES	0.001	<0.001	<0.001
Leaf P ES	0.741	0.001	0.016
NM leaf N:P	0.540	0.766	0.015
Leaf N:P ES	<0.001	0.155	<0.001
Gas exchange effect size	n	Intercept	I2 (%)	phetero	τ2	R2 analog (%)
Carbon exchange rate (CER)	224	0.462	0	0.824	0	-
Stomatal conductance (gs)	173	0.009	0	0.955	0	-
Transpiration rate (E)	124	1.609	6.5	0.281	0.046	100

CER, g_s, and E are AM/NM effect sizes; analysis was conducted on log-transformed values (lnR) from each study. ES, effect size, denotes that moderator is the study’s AM/NM effect size (response ratio) for shoot dry weight (DW), leaf P concentration, and ratio of leaf N to P concentration. NM leaf N:P is the ratio of leaf N concentration to leaf P concentration in the non-mycorrhizal control plants. p_test, “test of change,” p-value for moderator explaining significant further variation after the model has been adjusted for the other four moderators; n, number of studies; intercept, point at which predicted model line crosses y-axis; I², percent of variation due to real differences among study effects; p_hetero, probability that all observed (total) variation is due to sampling error (within-study variation); τ², between-studies variance. R² analog, the true variance explained as a proportion of the total true variance, =T²_explained/T²_total × 100 where T²_explained is the true variance explained by the regression and T²_total is the total true variance.