Appl. Entomol. Zool. 40 (4): 575–577 (2005) http://odokon.ac.affrc.go.jp/ Small protection plates against marten predation on nest boxes Noriyuki YAMAGUCHI,1,*,† Katsura M. KAWANO,1 Yasuhiro YAMAGUCHI2 and Takashi SAITO3 1 Department of Biology, Faculty of Science, Kyushu University; Fukuoka 812–8581, Japan 2 Wildlife Management Laboratory, National Agricultural Research Center; Tsukuba, Ibaraki 305–8666, Japan 3 Yamashina Institute for Ornithology; Abiko, Chiba 270–1145, Japan (Received 27 October 2004; Accepted 3 June 2005) Abstract The nest box is a powerful research tool for ecological and conservational studies of birds and several other cavity nesters. However, skilful predators such as martens sometimes invade nest boxes, thus disturbing researches. To pro- tect nest boxes against predation by martens, we attached a small plate inside the nest box below the entrance hole, and we report here on the advantage of this device. In 1999 and 2000, respectively, 73.0% and 64.8% nest boxes were used by the Great Tit and the Varied Tit. After fitting boxes with the plate in 2000, the percentage of predated nests by martens decreased from 22.4% to 5.9%, and the percentage of successful nests increased from 29.3% to 43.8%. Key words: Marten; nest box; protection against predation (dominant overstory trees are Machilus thunbergii, INTRODUCTION Quercus glauca, Styrax japonica, and Magnolia The nest box is a powerful and popular research obovata), ranging in elevation from 210 m to 380 m tool for ecological and conservational studies. above sea level. Dominant secondary cavity nesters Many researchers commonly use nest boxes, notic- are Varied Tits (Parus varius) and Great Tits ing careful regard to estimate the results that they (Parus major). Potential nest predators of the birds obtained. Nest boxes have some effects that sup- in the study area are snakes (mainly the Japanese press the predation rate (Nilsson, 1984a, b), in- rat snake [Elaphe climacophora]), Japanese crease clutch size (Nilsson, 1984b), and lower ec- martens (Martes melampus), Jungle Crows toparasite load (Møller, 1989, 1992). However, (Corvus macrorhynchos), and possibly Japanese problems can also occur in the use of nest boxes, Green Woodpeckers and feral cats (Felis catus) when skilful predators like martens, weasels, rac- (NY and KMK, pers. obs.). coons, feral cats and crows, intensively prey upon We erected 126 nest boxes in 263 ha and 182 birds in nest boxes (Perrins, 1965; Dunn, 1977; nest boxes in 291 ha in 1999 and 2000, respec- Sonerud, 1985; Robertson and Rendell, 1990). It is tively. The distances between the boxes were ap- required to prevent predation in some studies and proximately 50 m. Boxes were placed on trees at a control approaches. Here, we introduce a simple height of approximately 2.5 m from the ground. In method for protecting nests in nest boxes against 2000, we attached a small wooden plate inside all marten predation. nest boxes before erecting them (Fig. 1). We erected 126 of the 182 boxes on the same trees we selected in 1999. The remaining 56 boxes were MATERIALS AND METHODS erected on other trees in the expanded area. Study area and research designs. Our study We visited each nest box every 3 d and recorded was performed at the nature reserve at Mt. Abu- nesting species hatching date, fledging date, and rayama, Fukuoka, Japan (33°31ЈN, 130°23ЈE). The nest predation when it occurred. We determined a study area consists of a mosaic of Cryptomeria nest was successful when at least one of the chicks japonica plantations and young broad-leaved trees in a nest was over 15 d old. We deduced the preda- *To whom correspondence should be addressed at: E-mail: [email protected] † Present address: Department of Life Sciences, Faculty of Science, Rikkyo University, Nishi-ikebukuro 3–34–1, Tokyo 171–8501, Japan DOI: 10.1303/aez.2005.575 575 576 N. YAMAGUCHI et al. Fig. 2. What happened to nests in nest boxes: 1999 (un- treated boxes [nϭ106]; open bars) and 2000 (treated boxes [nϭ121]; black bars). “Others” represents failure nests be- cause of human (not authors) disturbance and dropping of nest boxes due to bad weather. Fig. 1. Profile of the nest box. A small wooden plate (H: 14 mm; W: 45 mm; L: 80 mm) was attached inside. tor in the following manner. If neither the box nor the nesting material had been damaged (or thrown about), we concluded that the predator was a snake. If the box surface had been scratched and the en- trance hole gnawed, we concluded that the predator was a marten. If the entrance hole was enlarged and encircled with peck marks, we concluded that the predator was a bird (probably a crow or a woodpecker). Fig. 3. Percent of nests predated by each type of predator Statistical analysis. We used Fisher’s exact for the two years. Numbers of depredated nests are 49 in 1999 probability test to compare the usage of nest boxes and 34 in 2000, respectively. for treated (protected) boxes with that for untreated boxes and to compare the proportions of depre- 29.2% to 43.8% (Fig. 2). The increase in breeding 2ϭ ϭ dated nests for each type of predator between success was significant (c 3 12.0, p 0.01). Preda- treated boxes and untreated boxes. We tested tion by martens declined remarkably (pϭ0.03) and whether the fates of the nests in treated boxes dif- predation by snakes did not change at all (Fig. 3). fered from those in untreated boxes using the chi- Usage of nesting boxes did not differ between squared test. Mean values are given with ϮSD. All untreated and treated boxes. This result suggests analyses were performed with the software pack- that the wooden plate attached within the nest age of “R” (http://www.r-project.org). boxes did not decrease the attractiveness of the nest boxes for the two tit species (although we did not control for the effect of year). The wooden plates RESULTS AND DISCUSSION attached probably did not disturb fledgings, since In 1999 and 2000, 92 of 126 (73.0%) and 118 of the mean length of nestling stages did not differ be- 182 (64.8%) nest boxes, respectively were used by tween the two years (NY, unpubl. data). the Great Tit or the Varied Tit. Some next boxes Predation by martens declined remarkably after were used twice in a breeding season (14 in 1999; the treatment. Martens certainly inhabited the three in 2000). The proportion of the nest boxes study area in both years, as we often observed their used did not differ between untreated boxes footprints and droppings. Assuming that the den- (erected in 1999) and treated boxes (erected in sity of martens did not change between the two 2000) (pϭ0.14). years, we suggest that the attachment of wooden Between the two seasons, the percentage of pre- plates inside may constitute an effective protection dated nests decreased from 46.2% to 28.1%, and against predation by martens. the percentage of successful nests increased from All boxes predated by martens (the boxes’ sur- Nest Box Protection against Marten 577 face had been scratched and the entrance hole REFERENCES gnawed) had nest materials drawn out from nest holes. The marten probably grasped nest material Dunn, E. (1977) Predation by weasels (Mustela nivalis) on with its mouth or forelimbs and pulled it out with breeding tits (Parus spp.) in relation to the density of tits and rodents. J. Anim. Ecol. 46: 633–652. nestlings or eggs. So the Japanese marten can Møller, A. P. (1989) Parasites, predators and nest boxes: probably access clutches or broods in boxes, only facts and artifacts in nest box studies of birds? Oikos when nest materials are full and the distance from 56: 421–423. the nest entrance to the nestlings is small, since the Møller, A. P. (1992) Nest boxes and the scientific rigour of martens’ head is too large to enter a nest hole of experimental studies. Oikos 63: 309–311. 30 mm in diameter (mean head widthϭ50.05 mmϮ Nilsson, S. G. (1984a) Clutch size and breeding success of ϭ the pied flycatcher Ficedula hypoleuca in natural tree- 5.66, n 27). The wooden plates lengthen the dis- holes. Ibis 126: 407–410. tance between nest entrance and nest materials and Nilsson, S. G. (1984b) The evolution of nest-site selection prevent martens from reaching nestlings or eggs di- among hole-nesting birds: The importance of nest preda- rectly. tion and competition. Ornis Scand. 15: 167–175. Perrins, C. M. (1965) Population fluctuations and clutch-size ACKNOWLEDGEMENTS in the great tit, Parus major L. J. Anim. Ecol. 34: 601– 647. We are most grateful to Kazuhiro Eguchi for critical com- Robertson, R. J. and W. B. Rendell (1990) A comparison of ments throughout the study. We would like to thank Tomoaki the breeding ecology of a secondary cavity nesting bird, Sameshima for giving advice on field techniques. Two anony- the tree swallows (Tachycineta bicolor), in nest boxes and mous reviewers provided helpful comments on the manuscript. natural cavities. Can. J. Zool. 68: 1046–1052. We thank the staff of the nature reserve of Fukuoka City and Sonerud, G. A. (1985) Nest hole shift in tengmalm’s owl Ae- the Wild Bird Society of Japan at Mt. Aburayama for cooper- golius funereus as defense against nest predation involv- ating with our field investigation. The Tsushima Wildlife Con- ing long-term memory in the predator. J. Anim. Ecol. servation Center of the Ministry of Environment provided us 54: 179–192. the head width measurement data of the Japanese marten..