Table S1: Summary of linear mixed models using restricted maximum likelihood (REML) from the initial field experiment and laboratory sequential-feeding bioassay.
Fixed effects
Random effects
Responses
Degrees of freedom (numerator, denominator)
F value*
Plant genotypic diversity treatment (O. biennis monocultures vs. polycultures)
Table S2: Average leaf area consumed (cm2) by Popillia japonicaon twenty Oenothera biennis genotypes grown in genotypic monocultures vs. polycultures.
Genotype
Monoculture mean*
Polyculture mean*
F
P
S18
4.7
2.8
5.3
0.023
S1
7.3
4.7
4.0
0.048
S6
8.4
8.8
0.1
0.81
S36
8.6
4.0
5.0
0.027
S21
8.9
8.1
0.2
0.68
S7
9.7
7.2
1.7
0.19
S15
10.1
5.5
7.6
0.006
S11
10.3
8.3
1.0
0.31
S20
10.9
7.3
2.3
0.13
S43
10.9
6.0
7.5
0.007
S9
11.3
9.1
0.9
0.34
A23
11.3
7.6
3.2
0.08
S17
11.7
13.4
0.4
0.54
A29
13.0
9.7
1.7
0.19
S38
14.5
8.0
7.2
0.008
A4
15.3
11.3
1.4
0.24
S32
15.7
15.2
0.0
0.89
S33
17.1
30.2
6.3
0.013
S5
19.2
8.0
11.4
0.001
S44
43.4
49.0
0.3
0.59
*Differences detected between least square means via post-hoc independent contrasts.
Figure S1: Relationship between the reduction in leaf area removed by herbivory in O. biennis polycultures (monoculture LS mean – polyculture LS mean) and the increase in fruit production in polycultures (polyculture LS mean – monoculture LS mean). Each point represents one O. biennis genotype (n = 20).
