Table 1 Relationship of cranial shape and log-centroid size to the most important variables examined in this study (individual ANOVAs using residual randomization).

From: Evolution of the bovid cranium: morphological diversification under allometric constraint

 

Df

R2

F

Z

P

Type

Shape~

 Tribe * Distance from mean

14

0.57

6.52

11.13

0.001

ols

 Tribe

7

0.47

10.16

10.86

0.001

ols

 Size + Facial length

2

0.21

12.23

7.07

0.001

pgls

 Facial length

1

0.15

17.03

6.03

0.001

pgls

 Size * Diet

7

0.17

2.55

5.86

0.001

pgls

 Size

1

0.09

9.54

5.72

0.001

pgls

 Diet * Distance from mean

7

0.13

1.95

4.10

0.001

pgls

 Subfamily

1

0.05

4.91

3.47

0.001

ols

 Hypsodonty

1

0.04

2.27

2.29

0.01

pgls

 Diet

3

0.04

1.28

1.27

0.11

pgls

 Habitat

4

0.10

1.38

1.19

0.11

ols

 NPP

3

0.04

0.79

−0.47

0.68

ols

Size~

 Tribe

7

0.68

24.3

8.76

0.001

ols

 Facial length

1

0.42

67.09

4.79

0.001

pgls

 Subfamily

1

0.27

34.98

4.03

0.001

ols

 Diet

3

0.15

5.30

2.97

0.003

pgls

 Hypsodonty

1

0.16

10.56

2.55

0.002

pgls

 Habitat

4

0.05

0.72

−0.20

0.58

ols

 NPP

3

0.00

0.03

−2.47

0.996

ols

  1. After phylogeny (represented by subfamilial and tribal differences), facial elongation and size provide the strongest explanations of shape, while diet, habitat, and net primary productivity (NPP) are not significant predictors of shape. Size explains large proportions of the variation in facial length, diet, and hypsodonty. Differences among subfamilies and tribes are of borderline significance, and differences among habitats and NPP categories are not. All tests with interaction factors (*) have significant interaction effects (results given in the text). pgls (phylogenetic least squares) tests take phylogeny into account while ols (ordinary least squares) tests do not. Tests with tribe, hypsodonty, habitat, and NPP use reduced species subsets. Results sorted by Z score (effect size).
Back to article page