Table 3.

Summary of the statistical analyses P–values indicating significant (P < 0.05) and non-significant (P > 0.05) results in elements comparing element concentrations in fur and blood between sites and the total fur and blood concentrations. CGF control site; Corea Game Farm; VDM Venetia diamond mine

	Fur	Blood	Total
Element	CGF~VDM	CGF~VDM	Fur~Blood
B	P < 0.05a	P > 0.05b	P < 0.05c2
Al	P > 0.05c1	P > 0.05*a	P < 0.05d
K	P < 0.05b	P > 0.05b	P > 0.05b
Ca	P > 0.05c1	P > 0.05*c1	P < 0.05d
V	P > 0.05*b	P > 0.05a	P < 0.05d
Cr	P > 0.05b	P > 0.05*c1	P < 0.05d
Mn	P > 0.05b	P < 0.05*c1	P < 0.05d
Fe	P > 0.05c1	P > 0.05b	P < 0.05c1
Co	P > 0.05*b	P > 0.05*c1	P < 0.05*c2
Ni	P > 0.05c1	P > 0.05*a	P < 0.05d
Cu	P > 0.05a	P > 0.05b	P < 0.05c1
Zn	P > 0.05c1	P < 0.05c1	P > 0.05b
As	P > 0.05b	P > 0.05*b	P < 0.05c2
Se	P > 0.05b	P > 0.05*b	P > 0.05b
Rb	P < 0.05c1	P > 0.05c1	P > 0.05d
Sr	P > 0.05b	P > 0.05*c1	P < 0.05d
Mo	P > 0.05c1	P > 0.05b	P > 0.05*b
Cd	P < 0.05*c1	P > 0.05c1	P < 0.05c2
Sn	P > 0.05*b	P > 0.05*a	P > 0.05*c1
Sb	P > 0.05*b	P > 0.05*b	P < 0.05*c2
Hg	P > 0.05c2	P < 0.05*c1	P < 0.05c2
Pb	P > 0.05c1	P > 0.05c1	P < 0.05*c2

Since the data distributions were not all normally distributed; where both data sets were not normally distributed, Mann-WhitneyU-test was performed (^a) where both data sets were not normally distributed, ANOVAs were used (^b) where both data sets were normally distributed, t-tests assuming equal (^c1) and unequal variances (^c2) were run where one data set was normally distributed and the other was not and a paired Wilcoxon test were run (^d) when comparing the fur and blood concentrations of elements that were not normally distributed

*Cannot compute exact p-values with ties (data with the same values)