Table 4 G-test of the null hypothesis that all three markers have the same cline centre

From: Evolutionary implications of discordant clines across the Baltic Mytilus hybrid zone (Mytilus edulis and Mytilus trossulus)

 

lnL c values assuming particular cline centre positions (km along transect)

lnL u

 

575

830

1100

1546

1736

1950

 

EFbis

−2.51

−0.65

−2.82

−6.99

−8.03

−8.86

−0.63

M7 lysin

−7.97

−6.31

−3.35

−1.37

−2.40

−5.48

−1.86

ME-mitotype

−17.86

−15.66

−11.07

−1.99

−0.43

−4.02

−0.43

Sum of lnL

−28.34

−22.59

−17.24

−10.35a

−10.86

−18.36

−2.92

Δ

25.08

19.67

14.32

7.43

7.94

15.44

 

χ2=2 × Δ

50.16

39.34

28.64

14.86

15.88

30.88

 
  1. Δ=(∑lnLc−∑lnLu).
  2. Values represent log-likelihood support for a cline shape estimate in which the cline centre is constraint to one of six null values (lnLc). These six values represent a range of cline centre positions derived from exact estimates and two log-likelihood support limits. lnLu is the unconstraint log-likelihood support.
  3. aUsing the maximum overall lnLc as null value (that is −10.35), the null hypothesis that clines share a common cline centre is rejected (χ2=14.86 with d.f.=2; the critical Chi-square value for d.f.=2 at P=0.001 is 13.82).
Back to article page