為瞭解以次級?穴為巢的鳥?對巢箱在繁殖或棲息上的?用，吾人於太?閣國家公園之關原地區設置?種規格之巢箱（?口與底面積：舊巢箱3 cm和169 cm?；新巢箱4.5 cm和182.25 cm?），於2004?2月至2005?8月間，監測巢箱中鳥?築巢巢材與棲息所?下排遺的出現情形，分別代表巢位與棲所之?用，以檢測（1）有鳥排遺出現的巢箱占總?之比?（鳥排遺占用?）與氣候的關係；（2）鳥?對巢位與棲所之?用在新、舊巢箱間之差?；（3）巢位與棲所之出現，彼此間有無關?，以及（4）同一巢箱其?用型態間在各?間的變化。結果顯示，在平均溫?較低時，鳥排遺占用?較高(p=0.003)，另當?水日?較多時，?水?對鳥排遺占用?之影響亦較高(p=0.003)。在巢位選擇上，根據苔蘚巢材之占用?(p＜0.001)、已完成巢之比?(p＜0.001)皆為舊巢箱高於新巢箱以及實際繁殖巢之資?顯示，青背山雀(Parus monticolus)與煤山雀(P. ater)傾向選擇使用?口與底面積較小之舊巢箱繁殖；而普通?(Sitta europaea)所?用之樹皮巢材則無顯著差?，惟普通?於2005?中皆用新巢箱繁殖，似有傾向選擇使用?口與底面積較大之新巢箱。在棲所方面，鳥排遺占用?在新、舊巢箱間無顯著差?(p＞0.1)。以所有的巢位與棲所占用巢箱的比??看，並未有明顯分別使用新、舊巢箱之情形，然其在巢箱的使用上仍有所分隔，在人為收取舊巢材的2004與2005?繁殖季中，鳥?傾向選擇?同巢箱分別進?築巢與棲息（2004?：p=0.003；2005?：p＜0.001）；且於繁殖季中實際有繁殖的巢箱，其後於非繁殖季中出現鳥排遺之機?有低於其他巢箱之趨勢(p=0.08)。另外，於人為清除舊巢材後，在前一?具有苔蘚、樹皮或鳥排遺之巢箱，隔?仍有較高的機?被鳥?作為相同之?用，顯示巢箱周圍的微棲地環境有可能影響鳥?在巢位與棲所上的選擇。

From Feb 2004 to Aug 2005, we surveyed nest materials and feces, representing breeding sites and roosting sites of the birds respectively, between 2 different types of nest boxes (entrance and bottom: 3 cm and 169 cm2 of old box; 4.5 cm and 182.25 cm2 of new box) at Guan-Yuan in the Taroko National Park, and studied (1) the effect of weather on the presence of feces in nest boxes, (2) the difference between the new and old nest boxes used by the birds on breeding and roosting, (3) the association between breeding and roosting sites and (4) temporal changes of use type in the same nest boxes. The results showed that feces had higher appearance in nest boxes when average temperature was lower (p=0.003). So did it when the precipitation was higher in a longer period of rainy days (p=0.003). In terms of breeding site selection, based upon the moss occupancy rates (p＜0.001), complete nests ratio (p＜0.001) which are all higher in old boxes than in new ones and the data of breeding nests, green backed tit (Parus monticolus) and coal tit (P. ater) tended to choose old nest boxes to breed. As for Eurasian nuthatch (Sitta europaea), there were no significant differences between old and new boxes of bark flake occupancy rates and complete nest ratio. However, that all breeding nests were found only in new nest boxes in 2005 seemed to indicate that nuthatch tended to choose new nest boxes. In terms of roosting site selection, the percentage of each type of nest boxes with feces showed no significant difference. The utilization of nest boxes was either for breeding or roosting, but rarely for both purposes. When we cleaned the old nest materials before breeding season, the birds tended to use different nest boxes for breeding and roosting respectively in breeding season (2004: p=0.003; 2005: p＜0.001). And in non-breeding season, feces had lower appearance in nest boxes that had been used for actual breeding in previous breeding season than in other nest boxes (p=0.08). In addition, the nest boxes with nest materials or feces in previous year have more probability to be used again in the same way in following year. This indicated that the microhabitats around nest boxes could affect the selection of breeding and roosting sites.