Table 2. Partial correlations from four multiple regression models in which the response of the log(cover + 1) of each introduced functional guild in 1999 was regressed on the log(cover + 1) of all four resident functional guilds in 1999.

	Introduced functional guild
Resident functional guild	Nonlegume forb	C3 grass	C4 grass	Legume
Nonlegume forb	-0.36***	-0.05	-0.12	-0.03
C3 grass	-0.11	-0.23**	0.01	-0.14
C4 grass	-0.47***	-0.32***	-0.48***	-0.24**
Legume	0.01	0.09	0.005	-0.12
Overall r2	0.30	0.14	0.24	0.11
Overall P	<0.0001	0.0003	<0.0001	0.002

Significance levels were determined by using sequential Bonferroni (22) corrections for four multiple regressions (i.e., smallest P value <0.0125, next smallest P < 0.0167, then P < 0.025, and the remainder P < 0.05). Uncorrected significance levels are presented for statistics that meet the sequential Bonferroni criteria: *, P < 0.05; **, P < 0.01; ***, P < 0.001.