100
R O G E R A . P O W E L L
rial behavior and with territory theory (Powell et al. 1997). Nonetheless, black
bears in North Carolina have negative values of Spearman’s r, indicating that
avoidance does occur among these bears at some scale.
Home range overlap differed significantly between stone martens of the
same sex and martens of opposite sex using two of the three indices for overlap
(table 3.1). Intrasexual territoriality has been deduced from observations of
home range overlap (Powell 1993b, 1994), but objectively quantified overlap
that incorporates probability of use has been lacking (but see Genovesi and Boitani 1997). Erlinge (1977, 1979) suggested that males of Mustela species
avoided females’ territories, indicating that large overlap in space may not indicate large overlap of important space. The data for stone martens indicate that
overlap between the sexes is significant. That Spearman’s r is negative both for
members of the same sex and for members of opposite sex indicates that members of opposite sex also exhibit some avoidance but not as much as do members of the same sex. Values of Spearman’s r vary as the size of the cells in the
grid varies and the mean difference in r between the sexes is maximal at an intermediate cell size. When a cell size larger than the average home range area is
used, home range overlap is similar within and between sexes. For cell size
smaller than home range area, overlap among members of the same sex becomes
smaller than overlap between the sexes. When cell size is much smaller than
mean home range area, overlap within and between the sexes again becomes
similar, indicating that members of the opposite sex with overlapping home
ranges avoid each other on a fine scale. Taber et al. (1994) found similar avoidance by groups of Chacoan (Catagonus wagneri) and collared peccaries (Tayassu
tajacu).
I p is little affected by cell size as long as cell size is small compared to home
range size because probabilities of being in cells are summed over the area of
overlap. Lloyd’s index m * and Spearman’s r are both sensitive to cell size and
therefore have the potential to give different results with different cell sizes. Cell
size must match the grain or scale of the hypotheses being tested. Therefore, I p
shows that stone martens of the same and opposite sex show different patterns
of home range overlap, whereas Spearman’s r is able to show that spatial avoidance may occur on a scale much finer than that of individual home ranges.
Lessons
Within Burt’s (1943) definition of home range, tremendous latitude exists for
different approaches to understanding what a home range is. No consensus
R O G E R A . P O W E L L
rial behavior and with territory theory (Powell et al. 1997). Nonetheless, black
bears in North Carolina have negative values of Spearman’s r, indicating that
avoidance does occur among these bears at some scale.
Home range overlap differed significantly between stone martens of the
same sex and martens of opposite sex using two of the three indices for overlap
(table 3.1). Intrasexual territoriality has been deduced from observations of
home range overlap (Powell 1993b, 1994), but objectively quantified overlap
that incorporates probability of use has been lacking (but see Genovesi and Boitani 1997). Erlinge (1977, 1979) suggested that males of Mustela species
avoided females’ territories, indicating that large overlap in space may not indicate large overlap of important space. The data for stone martens indicate that
overlap between the sexes is significant. That Spearman’s r is negative both for
members of the same sex and for members of opposite sex indicates that members of opposite sex also exhibit some avoidance but not as much as do members of the same sex. Values of Spearman’s r vary as the size of the cells in the
grid varies and the mean difference in r between the sexes is maximal at an intermediate cell size. When a cell size larger than the average home range area is
used, home range overlap is similar within and between sexes. For cell size
smaller than home range area, overlap among members of the same sex becomes
smaller than overlap between the sexes. When cell size is much smaller than
mean home range area, overlap within and between the sexes again becomes
similar, indicating that members of the opposite sex with overlapping home
ranges avoid each other on a fine scale. Taber et al. (1994) found similar avoidance by groups of Chacoan (Catagonus wagneri) and collared peccaries (Tayassu
tajacu).
I p is little affected by cell size as long as cell size is small compared to home
range size because probabilities of being in cells are summed over the area of
overlap. Lloyd’s index m * and Spearman’s r are both sensitive to cell size and
therefore have the potential to give different results with different cell sizes. Cell
size must match the grain or scale of the hypotheses being tested. Therefore, I p
shows that stone martens of the same and opposite sex show different patterns
of home range overlap, whereas Spearman’s r is able to show that spatial avoidance may occur on a scale much finer than that of individual home ranges.
Lessons
Within Burt’s (1943) definition of home range, tremendous latitude exists for
different approaches to understanding what a home range is. No consensus