Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Action: Provide artificial dens or nest boxes on trees Terrestrial Mammal Conservation

Key messages

Read our guidance on Key messages before continuing



  • Abundance (5 studies): Three of five controlled studies (three also replicated) in the USA, the UK, Canada and Lithuania, found that provision of artificial dens or nest boxes increased abundances of gray squirrels and common dormice. The other two studies found that northern flying squirrel and Douglas squirrel abundances did not increase.
  • Condition (1 study): A replicated, randomized, paired sites, controlled, before-and-after study in Canada found that nest boxes provision did not increase body masses of northern flying squirrel or Douglas squirrel.


  • Use (27 studies): Twenty-seven studies, in Australia, the USA, the UK, Canada, South Africa and Japan found that artificial dens or nest boxes were used by a range of mammal species for roosting and breeding.

Supporting evidence from individual studies


A study in 1940–1947 in a forest site in Michigan, USA (Stuewer 1948) found that artificial dens were used by raccoons Procyon lotor. Over the four years that 15 dens were monitored, 2–13 of them showed signs of being occupied by racoons. Fifteen dens were made of wood and measured 36 × 36 × 31 cm, with entrances measuring 10 × 15 cm. Dens were attached to trees in July 1940, at 7.5–12 m high. They were inspected for signs of racoon use in August, October, and November 1940, May 1941, June 1946, and June 1947.


A replicated, controlled, before-and-after study in 1963–1965 of a forest in Maryland, USA (Burger 1969) found that areas with artificial dens had more gray squirrels Sciurus carolinensis than did areas without dens. No statistical analyses were performed. There were more gray squirrels after dens were installed (1.0–1.8 squirrels/acre) than before installation (0.6–0.9 squirrels/acre). Numbers were stable through this period in plots where dens were not installed (0.8–0.9 squirrels/acre over two years in one plot and 1.0–1.2 squirrels/acre over three years in another). Squirrels were surveyed by live-trapping in five woodland plots (9.5–26 acres extent) in January–February. Three plots were sampled in 1963 and all five in 1964 and 1965. Artificial dens (one den/1.25 acres) were attached to trees in one plot after surveys in 1963 and in two plots after surveys in 1964. Dens comprised half a car tyre, folded and fastened into a kidney-shaped box, with an entrance at the top.


A study in 1974–1977 in a forest plantation site in Utah, USA (Pederson & Heggen 1978) found that nest boxes were used by Abert’s squirrel Sciurus aberti and red squirrel Tamiasciurus hudsonicus. After three years all 12 nest boxes installed were used by Abert's squirrels. Additionally, a red squirrel was detected in one next box, one year after installation. In May 1974, twelve nest boxes (30 × 30 × 40 cm) were placed in a forest area. Boxes were secured 7.6–14 m high, to ponderosa pine Pinus ponderosa, and were checked periodically for signs of use until October 1977.


A replicated study in 1973–1975 of two stands of young hardwood trees in Ohio and Illinois, USA (Nixon & Donohoe 1979) found that nest boxes were used by gray squirrels Sciurus carolinensis at one site and by flying squirrels Glaucomys volans at both sites. At a 21–23-year-old forest stand, gray squirrels did not make active use of any of 10 boxes but flying squirrels occupied 7–10 boxes over six inspections. At a 32–36-year-old forest stand, gray squirrels occupied 7–18 boxes across five inspections and flying squirrels occupied 2–6 boxes. Ten boxes were installed in autumn 1973 in the 21–23-year-old stand, which covered 1.9 ha. They were inspected six times from April 1974 to November 1975. Twenty boxes were installed in April 1973 in the 32–36-year-old stand, which covered 4 ha. They were inspected five times from August 1973 to March 1975.


A study in 1977–1979 in three riverine forest sites in Louisiana and Mississippi, USA (McComb & Noble 1981), found that nest boxes were used by Virginia opossums Didelphis virginiana, southern flying squirrels Glaucomys volans, fox squirrels Sciurus niger, gray squirrels Sciurus carolinensis, golden mice Ochrotomys nuttalli and eastern woodrats Neotoma floridana. Virginia opossums, southern flying squirrels and fox squirrels were more frequently detected in nest boxes than in natural cavities (opossums: 1.2% vs 0.2 of inspections; flying squirrels: 2.1% vs 0.2; fox squirrels: 0.7% vs <0.1%). Gray squirrels were detected with more similar frequencies in nest boxes (1.6 % of inspections) and natural cavities (1.1%). These comparisons were not subjected to statistical tests. Golden mice and eastern woodrats used next boxes rarely (<0.05% of box inspections). Boxes were erected in hardwood and hardwood/pine forests and were of three sizes: large (60 x 30 x 30 cm, 13 cm diameter entrance), medium (45 x 20 x 20 cm, 7.5 cm diameter entrance) and small (30 x 15 x 15 cm, 5 x 7 rectangle entrance). Fifty boxes were installed at two sites and 90 at the other. All boxes had 5–10 cm of pine shavings in the bottom. Boxes and natural cavities were inspected every month from April 1977 to February 1979.


A study in 1977–1980 in a range of agricultural, woodland and suburban areas across two counties in Tennessee, USA (Fowler & Dimmick 1983) found that nest boxes were used by eastern gray squirrels Sciurus carolinensis, southern flying squirrels Glaucomys volans and occasionally opossums Didelphis virginianus. Over three years, gray squirrels were detected in 4–34% of boxes in agricultural sites, 0–19% in woodland and 12–49% in suburban areas. Southern flying squirrels were detected in 0–6% of boxes in agricultural sites, 0–26% in woodland and 0–9% in suburban areas. Opossums were detected only in 2% of boxes in suburban sites during the winter of one year. In 1977, one hundred and fifty wooden nest boxes were erected. Fifty were installed across an unstated number of agricultural sites (at a density of 1 box/1.4 ha), fifty were installed across three woodland sites (1 box/2.0 ha) and fifty were installed across three suburban areas around one city (1 box/2.5 ha). Boxes were 48 cm high, had a 7.6-cm diameter entrance hole and were nailed 4.6–6.1 m high on trees. They were inspected during March-June (spring) and December-February (winter) from 1978 and 1980.


A study in 1979 in a forest in Maryland, USA (Gano & Mosher 1983) found that artificial den cavities were used by southern flying squirrels Glaucomys volans and white-footed mice Peromyscus leucopus. Within 12 months, 84% of artificial cavities had been used by rodents or birds (data provided for both groups combined). Southern flying squirrels nested in the 40 artificial cavities six times and white-footed mice once. In July–August 1979, forty artificial cavities were created in a forest dominated by chestnut oak Quercus prinus. Cavities were created in 37 oaks, two pitch pines Pinus rigida and one white ash Fraxinus americana. Trees averaged 28 cm diameter at breast height. Cavities were 1.5 m above ground, were 15 × 13 cm across and 15 cm deep. The slab of wood initially removed from the tree surface was reattached across the front of the cavity with a 3.8-cm-diameter entrance hole.


A replicated study in 1977–1980 in two forest sites in Victoria, Australia (Menkhorst 1984) found that nest boxes were used by brown antechinus Antechinus stuartii, bobucks Trichosurus caninus, feathertail gliders Acrobates pygmaeus, sugar gliders Petaurus breviceps and greater gliders Petauroides volans. Out of the total of 240 nest boxes across the two sites, brown antechinus used 13 (5%), bobucks used seven (3%), feathertail gliders used 20 (8%), sugar gliders used 16 (7%) and greater gliders used one (<1%). Preference for diameter of entrance hole and height of box was significant for brown antechinus (tended to use 5 cm hole; avoided 8 m height) and sugar glider (tended to use 5 cm hole; selected 8 m height), but no other mammal species. In July 1977, 120 nest boxes were installed in each of two 4-ha forest sites dominated by eucalyptus. Sites were located 6.5 km apart. Boxes were made of 13-mm wide wood, were 22 × 31 cm across and 45 cm high. Entrance hole sizes were 5, 8, 12 or 15-cm in diameter and boxes were attached at heights of 1.5, 4 or 8 m on tree trunks. Nest boxes were installed 20 m apart. Each contained a 50-mm layer of wood shavings. They were inspected fortnightly, for six months after installation and then approximately monthly until January 1980.


A replicated study in 1982–1984 in woodland at four sites in Western Australia, Australia (Wardell-Johnson 1986) found that nest boxes were used by mardos Antechinus flavipes. Within a 16-year-old regenerating block, all 36 boxes were used at least once, with 2–34 boxes being used across the 18 inspections. Single visits also revealed use of 7/34 boxes in virgin forest and 5/34 in streamside trees, but 0/34 were used in a 50-year-old regenerating block. Thirty-six nest boxes (internal volumes of 0.003–0.017 m3) were erected in each of four areas in June 1982. The 16-year-old block was 47-ha of regenerating karri forest. This was clear-felled in 1966 and prescribed burned in 1967. Boxes were fixed 3–5.5 m up trees. Further sites were virgin forest, retained streamside trees within a four-year-old regenerating block and a 50-year-old regenerating block. Boxes at these sites were set at 4.5–6.5 m height. Boxes were checked in the 16-year-old block monthly, from September 1982 to August 1983, then six further times to May 1984. Boxes at other sites were checked once, in May 1983.


A controlled study in 1986 in a woodland in Somerset, UK (Morris et al. 1990) found that nest boxes increased dormouse Muscardinus avellanarius abundance after 2–3 months. In woodland plots with nest boxes, more dormice were caught (8–11 dormice/plot) than in plots without nest boxes (3–6 dormice/plot). Within a 4-ha woodland, nest boxes were installed in two plots (0.8 and 1.2 ha), and two similar plots did not have nest boxes installed. Boxes, had internal dimensions of 115 ×130 × 120 mm and a 35-mm entrance hole. They were installed in May 1986, with the hole facing the tree, at a density of c.30 boxes/ha. Relative dormouse abundance in each plot was determined from live-trapping over 10 nights, simultaneously in box and non-box plots, in both July and August 1986.


A study in 1994–1997 in a coniferous forest in Lancashire, UK (Shuttleworth 1999) found that red squirrels Sciurus vulgaris used all and bred in some nest boxes. Red squirrels used all boxes within the first three months of placement and used 16-26% of boxes for breeding each year. There was no significant difference in the use of large (18 boxes) and small nest boxes (10 boxes) by breeding females, or in the size of litters in large (2.7 young) and small (2.9) boxes. All age groups and both sexes used boxes. The study site was dominated by Scots pine Pinus sylvestris and Corsican pine Pinus nigra and contained a high density of red squirrels (3.5–4/ha in the spring). Three groups of five small (27 × 30.5 × 48 cm) and five large (32 × 35.5 × 56 cm) timber nest boxes were attached to pine trees a height of 5–8 m in February 1994. Boxes were 50 m apart and filled with hay. In 1995, eight additional large boxes were added. Boxes were waterproofed and had a 7.5-cm-diameter entrance. Boxes were checked monthly from summer 1994 to summer 1997.


A study in 1994–1996 in a forest site Victoria, Australia (Ward 2000) found that nest boxes were used by feathertail gliders Acrobates pygmaeus and agile antechinus Antechinus agilis. Out of 40 nest boxes, feathertail gliders used nine (23%) and agile antechinus used one or two (3-5%). In total, 57 individual feathertail gliders and two agile antechinus used boxes. In January 1994, forty nest boxes were installed in a 7-ha forest area dominated by eucalyptus. Boxes were 50 m apart, had a 15-mm-wide slit as the entrance and were attached to tree trunks at approximately 4.5 m above ground. Nest boxes were checked approximately every two months, between July 1995 and May 1997. Inspections took place during daylight hours and all animals encountered were captured, individually marked and returned to the box.


A study in 1990–1993 in a rainforest in New South Wales, Australia (Bladon et al. 2002) found that nest boxes were used by eastern pygmy-possums Cercartetus nanus. Over the first 16 months, the average monthly capture rate of eastern pygmy-possums was 33.5/100 nest box checks. Twenty-one months after the study commenced, part of the area was cleared and the average monthly capture rate dropped to 7.8/100 nest box checks. Ninety-eight individual pygmy-possums were caught in boxes over the study. The study was conducted in a 4-ha early regrowth rainforest plot at 1,200 m altitude. Between 28 and 55 nest boxes (the quantity changing through the study) were attached to tree trunks, 1.5–2.0 m above ground and 10–20 m apart. Boxes were made from 18-mm-wide pine wood, and were 17 × 17 cm and 25 cm tall, with a 1.5-cm-wide opening across the front under the lid. In February 1992, 1.4 ha of the study area was cleared by bulldozing and burning. Boxes were checked at least monthly, between June 1990 and December 1992, and in April 1993.


A replicated, randomized, controlled study in 1992–1998 in a forest in Washington, USA (Carey 2002) found that artificial breeding sites were used by northern flying squirrels Glaucomys sabrinus but did not increase their abundance. Average northern flying squirrel abundance in sites with artificial dens (0.51–0.80 squirrels/ha) was not significantly higher than in sites without artificial dens (0.42–0.48 squirrels/ha). During 11 inspections of the 256 dens, a total of 349 northern flying squirrels, 201 Douglas’ squirrels Tamiasciurus douglasii and 16 Townsend's chipmunk Tamias townsendii were detected. By the end of the study 74-80% of next boxes and 34-50% of artificial cavities were used. In 1992, 16 nest boxes (20 × 22 cm across and 22 cm tall, with a 3.8 × 3.8-cm entrance) and 16 artificial cavities (10 ×15 cm across and 18–33 cm tall with a 3.8 × 3.8 cm or 4.5-cm-diameter entrance) were added to eight of 16 Douglas-fir Pseudotsuga menziesii stands. Forest stands were 13 ha and located in four areas (≤4 km apart). Each area had two stands with supplementary dens and two stands without supplementary dens (each ≥ 80 m apart). Supplementary dens were 6 m high and were inspected once in summer and once in winter, from summer 1993 to summer 1998. Flying squirrels were trapped during 49,152 trap nights in 1997–1998, with two Tomahawk live traps at each of 64 samplings stations, in each stand.


A replicated study in 1996–2000 in three forest plantations and one native forest in Queensland, Australia (Smith & Agnew 2002) found that nest boxes were used by feathertail gliders Acrobates pygmaeus, sugar gliders Petaurus breviceps, squirrel gliders Petaurus norfolcensis and yellow‐footed marsupial mice Antechinus flavipes at three of four sites. Between 0 and 40% of nest boxes were occupied at each check within each of the three plantations. No boxes were used in the native forest. Out of 96 boxes, feathertail gliders used 16 (17%), sugar gliders used 10 (10%), squirrel gliders used four (4%) and yellow‐footed marsupial mice used one (1%). The study was conducted in three 2–18-year-old eucalyptus plantations (1.2–1.5 ha) and one native forest dominated by >30 year-old eucalyptus (1.8 ha). At each site, 24 boxes were attached to trees, 3 m or 6 m above ground and 2–25 m apart. Nest boxes (40 cm long, 20 cm wide, ≤18.5 cm deep) were made from laminated plywood and had a 15–20 mm wide slot at the bottom. Boxes were checked 5–9 times between April 1996 and November 2000.


A replicated study in 1998–2002 of two Eucalyptus regnans-forests in Victoria, Australia (Lindenmayer et al. 2003) found that nest boxes were used by four arboreal marsupial species, with large high boxes used more than smaller or lower boxes. No statistical analyses were performed. Leadbeater’s possum Gymnobelideus leadbeateri, mountain brushtail possum Trichosurus cunninghami, common ringtail possum Pseudocheirus peregrinus and eastern pygmy possum Cercartetus nanus were recorded. There were 38 records of presence of these species in large high boxes, 16 in small high boxes, 10 in large low boxes and 18 in small low boxes. In each of two forests, 12 locations were selected. Each had four trees in a 20 × 20 m square. At each location, a large high, large low, small high and small low box was installed in October–November 1998, one on each tree. Large and small box volumes were 0.038 m3 and 0.019 m3 respectively. High and low boxes were set at 8 m and 3 m height respectively. Boxes were checked 10 times to January 2002. Mammal occupancy was determined by animal presence, or hairs left on sticky devices.


A replicated, randomized, paired sites, controlled, before-and-after study in 1996–1999 in three forest sites in British Columbia, Canada (Ransome & Sullivan 2004) found that nest boxes were used by northern flying squirrels Glaucomys sabrinus and Douglas squirrels Tamiasciurus douglasii but did not increase their abundance or body mass. Northern flying squirrels occupied 68–83% of boxes with Douglas squirrels occupying 0–29%. However, two years after boxes were erected, the abundance and body mass of northern flying squirrels did not differ significantly between plots with nest boxes (abundance: 9.8/ha; body mass: 134 g) and plots without nest boxes (abundance: 7.7/ha; body mass: 128 g). At the same time, the abundance and body mass of Douglas squirrels also did not differ significantly between plots with nest boxes (abundance: 15.1/ha; body mass: 198 g) and plots without nest boxes (abundance: 20.1/ha; body mass: 207 g). In February–March 1997, thirty nest boxes (12.8 × 13.6 × 15.5 cm), 100 m apart in a 5×6 grid and 5.5 m above ground, were mounted in each of three 30-ha plots. Three other 30-ha plots had no nest boxes. In each plot, squirrels were trapped every 5–6 weeks during the snow-free period, from June 1996 to March 1999, using 80 baited Tomahawk live traps, at 40-m intervals in an 8×10 grid.


A replicated study in 1993–1994 in 20 forest sites in Victoria, Australia (Harper et al. 2005) found that nest boxes were used by common brushtail possums Trichosurus vulpecula and common ringtail possums Pseudocheirus peregrinus. Over one year, common brushtail possums were detected in 43% (52) and common ringtail possums in 33% (40) of the available 120 nest boxes. The average occupancy rate of nest boxes per monthly survey was 9% for common brushtail possums and 10% for common ringtail possums. In July 2003, 120 nest boxes were installed in 20 randomly selected (from 44) forest fragments (<2 ha) within a 183-km2 study area. Boxes were of two designs (12 or 25-mm-wide plywood; 30 × 30 x 27.5 or 30 cm high), had a 10-cm diameter entrance hole and were attached to tree trunks approximately 4 m above the ground. Nest boxes were installed 50 m apart, on either side of a 100-m transect crossing the centre of each fragment. Nest box monitoring commenced eight weeks after installation and each box was inspected monthly over one year.


A replicated study in 2002–2003 in four forest sites in New South Wales, Australia (Harris & Goldingay 2005) found that nest boxes were used by eastern pygmy-possums Cercartetus nanus and brown antechinus Antechinus stuartii. Five individual pygmy-possums (three of which were encountered twice) at one site and five brown antechinus were detected over 264 nest box inspections. Additionally, nesting materials characteristic of pygmy-possums was detected in eight nest boxes at the one site and brown antechinus in 11 nest boxes across the sites. The study was conducted in four 1-ha sites within a 2,000-ha forest reserve. In July-November 2002, forty nest boxes were attached to tree trunks, 1–2 m above the ground. Boxes had a 15-mm-wide entry slot and were placed 10–20 m apart. Boxes were checked eight times, with visits in alternate months in 2002 and then monthly.


A controlled, before-and-after study in 1985–1989 and 2000–2003 in a forest site in Lithuania (Juškaitis 2005) found that after more nest boxes were provided, common dormouse Muscardinus avellanarius density approximately doubled. Dormouse density was higher when there were 16 boxes/ha (0.9–3.0 dormice/ha) than when there were 4 boxes/ha (0.3–1.5 dormice/ha). Dormouse density did not increase in an area where next box provision remained at 4 boxes/ha (after: 0.6–0.9 individuals/ha; before: 0.7–1.3 individuals/ha). The study was conducted in 60 ha of a 40-50-year-old forest. In 1985–1999 wooden nest boxes (12 × 12 × 24 cm) were installed in a 50 × 50 m grid (276 boxes, 4 boxes/ha). In 2001, eighty-five additional nest boxes were added to a 6.25-ha section of the forest to form a 25 × 25 m grid (increasing box density to 16 boxes/ha). Boxes were inspected twice each month from April until October in 1985–1989 and 2000–2003.


A replicated study in 2005–2007 in five eucalyptus plantation sites in New South Wales and Queensland, Australia (Goldingay et al. 2007) found nest boxes were used by five marsupial species with different frequencies, depending on box type. Feathertail gliders Acrobates pygmaeus used 15 of 45 available small rear-entry boxes, 10 large slit-entrance boxes and nine wedge-shaped boxes, but did not use any medium rear-entry boxes. Squirrel gliders Petaurus norfolcensis used 18 of 45 medium rear-entry boxes and three large slit-entry boxes. Yellow-footed antechinus Antechinus flavipes used two large slit-entry boxes and one medium rear-entry boxes. Brown antechinus Antechinus stuartii used three small rear-entry boxes and brush-tailed phascogales Phascogale tapoatafa used one large slit-entry box. Nest boxes were of four types, small rear-entry boxes (height×width×depth: 23×14×14 cm, 25-mm-diameter entrance), large slit-entrance boxes (48×28×18.5 cm, 1.5×15 cm entrance on the side), wedge-shaped boxes (19×16×12.5–5 cm, 2×16 cm entrance at the base) and medium rear-entry boxes (40×14.5×14 cm, 45-mm-diameter entrance). They were installed in February–March 2005 and March 2006, 3 m above ground, in 45 plots. Each plot had one of each box type (180 boxes in total). Boxes were surveyed five times over 22 months.


A study in 1993–2005 of restored sites within bauxite mined areas in the jarrah Eucalyptus marginata forest of Western Australia, Australia (Nichols & Grant 2007) found that nest boxes within restoration areas were used by western pygmy possums Cercartetus concinnus, mardo Antechinus flavipes and brush-tailed phascogale Phascogale tapoatafa. Western pygmy possum used nest boxes placed in 8–10-year-old restoration sites. Mardo and brush-tailed phascogale also used nest boxes and possibly bred in them (no further details provided). Mined areas were revegetated using various techniques. In 1993–1994, mammal nest boxes were placed in a range of sites. Control of non-native red foxes Vulpes vulpes was also carried out for several years from 1994. Nest box designs and monitoring protocols are not described.


A study in 2003–2007 in a forest reserve in Eastern Cape, South Africa (Madikiza et al. 2010) found that nest boxes were used by woodland dormice Graphiurus murinus and Mozambique thicket rats Grammomys cometes. Out of 70 nest boxes, at least 49 (70%) were occupied by dormice and seven (10%) by thicket rats. Dormouse nest box occupation was lowest during winter (3% of boxes) and peaked in spring (8%) and summer (9% of boxes). Over one year, at least 66 dormice used between one and 16 next boxes (average 4). More adult females (17) than adult males (11) used nest boxes, but they were used by similar numbers of adults (30) and juveniles (36). Between March 2003 and January 2006, seventy wooden nest boxes (11.5 × 13 × 12 cm) were erected across a 2.5-ha area. Boxes had a 3-cm-diameter entrance hole facing the tree trunk. Boxes were installed 1.1–2.4 m above the ground, in trees with an average trunk diameter at nest box height of 90 cm. Boxes were monitored 57 times (average 4.4 times/month) between June 2006 and June 2007. Captured dormice were individually marked to determine recaptures.


A study in 2003–2006 of 16 woodland fragments in Queensland, Australia (Ball et al. 2011) found that 20% of nest boxes were used by squirrel gliders Petaurus norfolcensis. In total, 11 out of 56 nest boxes were occupied at least once by squirrel gliders, with presence detected 15 times out of 318 box visits. No squirrel gliders were found in boxes until ≥18 months after placement. Four of the boxes were occupied by five female gliders with young. In 16 woodland remnants (from <50 ha to >1,000 ha in extent), 56 nest boxes were erected in September–December 2003. Boxes were 40 cm high, 25 cm wide and 18 cm deep. They were installed ≥3 m above the ground. There were 2–6 boxes/site, with the number dependent on site size. Boxes were checked at six-month intervals from summer 2003 to summer 2006.


A study in 2008–2011 in a forest area in North Carolina, USA (Kelly et al. 2013) found that nest boxes were used by northern flying squirrels Glaucomys sabrinus. Sixteen northern flying squirrels were caught at nest boxes. The study was conducted in a forest area dominated by eastern hemlock Tsuga Canadensis. The number of nest boxes used was not detailed. Nest boxes measured 30 × 18 × 15 cm, had a 5 × 5-cm entrance, and were attached 3.6 m up the trunks of trees using nails and wire. They were monitored in winters of 2008 to 2011 and in spring 2009. Captured flying squirrels were individually tagged.


A study in 2004–2005 in a forest reserve in Nagano Prefecture, Japan (Nakamura-Kojo et al. 2014) found nest boxes were used by Japanese dormouse Glirulus japonicus. Of 200 nest boxes, at least 127 (64%) were occupied by dormice. Thirty-nine individuals used the nest boxes (total 82 captures), 23 males and 16 females. The number of dormice captured in nest boxes peaked in August 2004 and June 2005 (14 captured/month) and October in both years (10-13). Pup-rearing was observed twice in nest boxes. The average diameter at breast height of trees with used nest boxes (33 cm) was smaller than unused boxes (51 cm). In early 2004, two hundred nest boxes were installed at equal distances across a 3.8-ha area of dense deciduous forest. Nest boxes were constructed from 12-mm-wide pinewood boards with a 35 x 35 mm square entrance at one side. Boxes were attached to trees with a diameter at breast height <40 cm, at a height of 1.0–1.2 m. Boxes were checked 2–4 times/month (total 76 times) between April 2004 and October 2005. Captured dormice were individually marked. Nest boxes were considered occupied when either dormice were present or when nesting materials were found.


A replicated study in 2003–2014 in one urban and two rural forest sites in New South Wales and Queensland, Australia (Goldingay et al. 2015) found that nest boxes were used by six species of arboreal marsupial. Within the rural landscapes nest boxes were occupied by sugar gliders Petaurus breviceps (29% of available boxes, use affected by design), brown antechinus Antechinus stuartii (23%, use unaffected by design), mountain brushtail possums Trichosurus caninus (1%) and feathertail gliders Acrobates pygmaeus (1%). Within an urban landscape, nest boxes were occupied by common brushtail possum Trichosurus sp. (20% of available boxes), common ringtail possum Pseudocheirus peregrinus (4%), and sugar gliders (4%). Use of some nest boxes influenced by design (see original paper for details). All boxes accessible to squirrel gliders Petaurus norfolcensis at two sites were used by them over a 10-year period (6-21 adults/year in boxes; total 61 individuals). Nest boxes of five different types (11–42 × 15–29 × 26–45 cm, 3.5–21-cm diameter entrance) were installed 3–6 m above ground. In the rural landscape, five boxes in each of 32 plots (25 x 25 m; ≥ 200 m apart) were installed across nine sites (>1 km apart). At the urban site a total of 188 boxes were installed across 20 sites. Boxes were erected in 2003–2007 and inspected three times in 2008–2009 at the rural sites and once in August 2010 at the urban site. In 2005–2009, 16 additional boxes were installed or adapted for squirrel gliders across two sites and were inspected usually once/year in 2005-2014.


A study in 2003–2016 in a coniferous forest plantation in Dumfries and Galloway, UK (Croose et al. 2016) found that pine martens Martes martes occupied and, in most years, bred in den boxes. Each year, 30–70% of available den boxes were occupied by pine martens. Martens used 5–20% of den boxes for breeding, in 10 of the 12 years monitored. The study was conducted in an 800-km2 forest into which 12 martens were reintroduced in 1980–1981. Fifty den boxes (55 cm high, 51 cm wide, 24 cm deep) were fitted to trees at approximately 4 m high. Ten boxes were installed in 2003 and 40 in 2013. Boxes were made of wood, had two entrances and had 10 cm depth of softwood shavings inside the chamber. Boxes were checked for martens, signs of use by martens and marten kits, once/year in 2004-2016 (excluding 2013).


A study in 2010–2013 of planted and remnant woodland patches at 30 sites in New South Wales, Australia (Lindenmayer et al. 2016) found that nest boxes were used by five native and one non-native mammal species. Use of boxes was detected for yellow-footed antechinus Antechinus flavipes (two detections), sugar glider Petaurus breviceps (two detections), common brushtail possum Trichosurus vulpecula (52 detections), common ringtail possum Pseudocheirus peregrinus (eight detections) and lesser long-eared bat Nyctophilus geoffroyi (four detections). The introduced black rat Rattus rattus was also detected on 24 occasions. One each of five nest box designs was placed at 30 sites. Sites comprised seven connected woodland plantations, nine isolated woodland plantations (>70 m from native vegetation), eight connected remnant woodlands, and six isolated remnant woodlands (>70m from native vegetation). Boxes were erected in February 2010 and checked in October 2010, December–January of 2010–2011, October 2011 and December–January of 2012–2013. Mammals were identified from live animals or from signs, such as faeces.


A study in 2010–2013 in a eucalypt forest in New South Wales, Australia (Lindenmayer et al. 2017) found that nest boxes were used by a range of native and non-native mammal species. Yellow-footed antechinus Antechinus flavipes were found in 12–14% of nest boxes, common brushtail possum Trichosurus vulpecula in 11–13%, and common ringtail possum Pseudocheirus peregrinus in 3–7%. Brush tailed phascogale Phascogale tapoatafa, squirrel glider Petaurus norfolcensis, and sugar glider Petaurus breviceps were all found in <1% of nest boxes. The non-native black rat Rattus rattus was found in 4–14% of boxes and the house mouse Mus musculus in 0–2% of boxes. On an unspecified date, 587 nest boxes were installed in a woodland. Animal presence, or signs of presence, were recorded during six surveys in 2010–2013.

Referenced papers

Please cite as:

Littlewood, N.A., Rocha, R., Smith, R.K., Martin, P.A., Lockhart, S.L., Schoonover, R.F., Wilman, E., Bladon, A.J., Sainsbury, K.A., Pimm S. and Sutherland, W.J. (2020) Terrestrial Mammal Conservation: Global Evidence for the Effects of Interventions for terrestrial mammals excluding bats and primates. Synopses of Conservation Evidence Series. University of Cambridge, Cambridge, UK.