TABLE 2.
Differences in the number of flower heads, the diameter of mature flower heads, the duration of bloom period and the pollen production between plants of different Tanacetum vulgare chemotypes.
| Response factor | Fixed effects | Parameter estimates ± SE for chemotypes |
Variance estimates ± SE of random effects | Statistical test and result | ||||
| β-thu. mono | art.ket. mono | α, β-thu. mix | art. acet. mix | myrox. Mix | ||||
| Flower heads (n) | Chemotype | 46.5 ± 9.4 N = 12 | 39.4 ± 10.9▼ N = 9 | 57.7 ± 8.1 N = 16 | 44.3 ± 8.1 N = 16 | 66 ± 10.3▲ N = 10 | MG: 0.0 ± 3.4 (g = 12) BO: 0.0 ± 0.0 (g = 21) | GLMM (Poisson, log link) χ2 = 82.72, P < 0.001 |
| Ø flower heads (mm) | Chemotype | 9.2 ± 0.4 N = 11 | 9.4 ± 0.5▲ N = 7 | 8.2 ± 0.5 N = 7 | 9.4 ± 0.4▲ N = 12 | 8 ± 0.4▼ N = 9 | MG: 0.6 ± 0.8 (g = 12) BO: 0.0 ± 0.1 (g = 18) | LMM (Gaussian, identity link) χ2 = 12.82, P = 0.04 |
| Bloom period (d) | Chemotype | 22.8 ± 4.7 N = 12 | 15.5 ± 5.8 N = 9 | 24.0 ± 4.2▼ N = 16 | 14.4 ± 4.2▼ N = 16 | 23.6 ± 5.1 N = 10 | MG: 0.2 ± 0.5 (g = 12) BO: 0.7 ± 0.8 (g = 21) | GLMM (Poisson, log link) χ2 = 63.5, P < 0.001 |
| Pollen (mg) | Chemotype | 18.9 ± 4.0 N = 12 | 14.8 ± 4.6 N = 9 | 20.1 ± 3.5 N = 16 | 18.3 ± 3.5 N = 16 | 27.4 ± 4.3 N = 10 | MG: 0.0 ± 0.6 (g = 12) BO: 0.0 ± 0.0 (g = 18) | LMM (Gaussian, identity link) χ2 = 3.92, P = 0.81 |
Number of flower heads and pollen production are summed over the entire blooming period. Displayed are parameter estimates ± standard error values for the (generalized) linear mixed models [(G)LMM] and levels of significance for differences between chemotypes. Variance estimates are shown for the random effects maternal genotype (MG) and bloom onset (BO; i.e., the day on which the first mature flower was observed) and the number of groups of the respective random effect (g) is given in brackets. Significant values (P < 0.05) are highlighted in bold and pointing up ▲ or down ▼ triangles are added behind the highest and lowest parameter estimate, respectively.