sions confidently. However, the Area (Prob.=
0.0492), Cperim ( Prob.=0.0441) and D
max
( Prob.=0.0262) show significant diVerences
between the samples cut along the jar, Grp 1, (the
normal case) and samples cut diagonally, Grp 3,
see Fig. 6(c), (d), (f ). D
max
also shows a signi®cant
diVerence between Grp 1 (the normal case) and
Grp 2 (where the samples have been cut across the
jar). In conclusion, the shape parameters are not
greatly aVected by somewhat oriented particles but
the size parameters are. Also the samples for the
filler fraction (