Table 3.
Eigenvectors of conserved genetic variation (h), divergence in means (d), and divergence in G (e)
| d1 | d2 | h1 | h2 | e11 | e28 | |
|---|---|---|---|---|---|---|
| % variance | 73.5 | 18.5 | 46.4–60.8 | 13.0–29.6 | 40.7 | 12.7 |
| Jaw length | 0.017 | 0.035 | 0.187 | 0.185 | –0.020 | –0.349 |
| Head width | 0.365 | 0.369 | 0.139 | 0.391 | 0.006 | –0.182 |
| Pectoral width | 0.140 | 0.424 | 0.250 | 0.756 | –0.104 | –0.512 |
| Pelvic width | 0.216 | 0.511 | 0.197 | 0.320 | –0.019 | –0.315 |
| Humerus | 0.323 | 0.209 | 0.500 | –0.180 | 0.459 | –0.110 |
| Ulna | 0.532 | 0.039 | 0.491 | –0.186 | 0.411 | –0.216 |
| Femur | 0.373 | –0.375 | 0.401 | –0.172 | 0.531 | –0.512 |
| Tibia | 0.522 | –0.485 | 0.438 | –0.207 | 0.571 | –0.403 |
For h, eigenvectors derive from Krzanowski's common subspace analysis (Table S4) and percent variance is given as a range when h is projected through species‐specific G matrices. For e, eigenvectors derive from covariance tensor analysis (Table S5) and percent variance is the amount of divergence in G explained. For e vectors, subscripts refer to the subspace (E 1–E 6) and the vector number within the subspace (1–8).