Abstract
Deer-proof fences have been used to protect and restore endangered perennial herbs in some areas of Japan where the forest floor vegetation has been diminished by browsing sika deer (Cervus nippon Temminck). However, little work has been done to evaluate the optimal timing for installing such deer-proof fences. Thus, I surveyed the occurrence, number of individuals, and number of mature individuals of 12 perennial species inside seven deer-proof fences over 5 years post-establishment, three of which were installed in 1997 (1997 fences), and the other four in 2003 (2003 fences). The occurrence of six species was equal between the 1997 fences and the 2003 fences, while that of the remaining six species was higher within the 1997 fences than the 2003 fences. Past records indicated that three of these latter six species were distributed equally between the areas of the 1997 and 2003 fences. These results suggest that the potential distribution of nine species found within the 1997 fences was equal to that of the 2003 fences. The number of individuals of four of these nine species was greater in the 1997 fences than in the 2003 fences, or showed differing relative rankings. These findings suggest that these four species have difficulty recovering, given the assumption that the fences were constructed in regions where the forest floor was subjected to the continual sika deer browsing. In contrast, the remaining five species did not show significantly different numbers of individuals between the 1997 and 2003 fences. Thus, these five species appear to be unaffected by 10 or 16 years, respectively, of sika deer browsing. These findings demonstrate that some species have difficulty recovering once the forest floor has been diminished by browsing sika deer. Thus, the early establishment of deer-proof fences is suggested to enable the best recovery of perennial herbs.
