Scott D. Wilson, Jonathan D. Bakker, Janice M. Christian, Xingdong Li, Laura G. Ambrose, and John Waddington. 2004. Semiarid old-field restoration: is neighbor control needed? Ecological Applications 14:476–484.

Appendix C. Effects of neighbor control (C), seeding method (S), and mulch (M) on the total cover of nonnative species, the total cover of native species, and species richness in three seeding years. All measurements were made in August of the year after seeding.

		Nonnative cover					Native cover				Species richness
		df	SS	F	P		SS	F	P		SS	F	P
Plots seeded in 1994
Control		1	6.950	58.76	0.000		0.211	1.83	0.182		11.350	94.93	0.000
Seed		2	0.177	0.75	0.479		0.973	4.22	0.020		2.690	11.24	0.000
Mulch		1	0.403	3.41	0.071		0.037	0.32	0.573		0.092	0.77	0.384
C × S		2	0.342	1.45	0.246		1.210	5.27	0.009		0.369	1.54	0.225
C × M		1	0.140	1.18	0.283		0.005	0.04	0.841		0.031	0.26	0.611
S × M		2	0.030	0.13	0.880		0.310	1.35	0.270		0.602	2.52	0.091
C × S × M		2	0.083	0.35	0.706		0.170	0.74	0.483		0.011	0.04	0.957
Error		48	5.680				5.530				5.740
Total		59	13.800				8.450				20.880
Plots seeded in 1995
Control		1	1.550	19.19	0.000		0.173	1.47	0.230		0.523	7.36	0.008
Seed		2	0.399	2.47	0.094		1.340	5.69	0.006		2.630	18.50	0.000
C × S		2	0.482	2.98	0.059		0.297	1.26	0.292		0.006	0.05	0.956
Error		54	4.360				6.350				3.840
Total		59	6.790				8.160				7.000
Plots seeded in 1996
Mulch		1	0.348	3.02	0.099		0.155	2.44	0.136		0.187	1.46	0.243
Error		18	2.070				1.150				2.310
Total		19	2.420				1.300				2.500
Control		1	0.204	2.58	0.114		0.272	3.10	0.084		0.582	6.31	0.015
Seed		2	0.156	0.99	0.379		0.218	1.24	0.297		1.300	7.05	0.002
C × S		2	0.420	2.66	0.079		0.042	0.24	0.787		0.110	0.60	0.555
Error		54	4.270				4.740				4.980
Total		59	5.050				5.270				6.980

   Notes: All statistics are from ANOVAs within each seeding year. Significant F values are in bold. Neighbor control (applied or not) and seeding method (unseeded, broadcast, drilled) were main factors in all sites and years. Mulch (applied or not) was a main factor in the 1994 seeding year. In the 1996 seeding year, mulch effects in plots that received neighbor control and broadcast seeding were examined with one-way ANOVA. The native hay seeding method was omitted from the analyses. Cover data were arc-sin square root transformed and species richness data were log-transformed prior to analysis. All SS have been multiplied by 10 for presentation.