No significant differences in habitat use between genders were observed in Arizona [43]. Typical brush mouse habitat in southern Arizona was characterized by 74% tree cover, 60% leaf cover, 21% shrub cover, and 16% rock cover. Additionally, 67% of all brush mice relocated by radiotelemetry in the study were located in a riparian zone with the other observations occurring in uplands and an intermittent stream channel. Brush mice used sites with significantly (P<0.05) more rock cover (19% to 22%) during winter and spring than at other times of the year. Also in spring, plots with the most brush mice also had significantly (P<0.05) more shrub cover (21% vs.14%) and succulents (9% vs. 2%) than random plots [43].
In a New Mexico study, 27% of brush mice captured were taken around rocky areas, 10% under Gambel oak (Q. gambelii), 9% under juniper, and 7% each around woodpiles and sacahuista [114]. Less frequently, brush mice were taken under or around pinyon pine, gray oak (Q. grisea), apache plume, rabbitbrush, mountain-mahogany, white fir (Abies concolor), Douglas-fir (Pseudotsuga menziesii), locust (Robinia spp.), prickly-pear, cholla (Opuntia spp.), chokecherry (Prunus spp.), bricklebush (Brickellia spp.), and grape (Vitis spp.) [114].
In southern California, brush mice were captured on leaf mold in an oak hardwood association of coast live oak, white alder (Alnus rhombifolia), whiteflower currant, and sumac (Rhus spp.) [110]. They were also trapped under logs and dense vegetation and on wet seepage slopes next to a creek [110]. In coastal California, brush mice were captured primarily under coast live oak, false-willow (Baccharis douglasii), California buckeye, and California bay [58].
Fallen logs and rock outcrops provide nest sites for the brush mouse [17]. They may also construct nests in tree hollows and underground burrows [113]. Brush mice living in mine shafts or caves build nests similar to those of goldfinches [48]. Davis [26] noted that the brush mouse nest is a "globular structure" and is constructed primarily of dried grasses within natural cavities. Brush mouse nest sites in California were found on sites with a high density coast live oak (Q. agrifolia) overstory and an open understory with low vegetation and ground cover [57].
Most small mammals killed in fires likely die from suffocation rather than high temperatures or catching on fire [22,65], although many exceptions have been documented. For instance, several days after fire passed through a chaparral-covered canyon in California, 2 mice (Peromyscus spp.) carcasses were discovered [22]. The deer mice were not touched by fire, but a Virginia opossum (Didelphis virginiana), a desert cottontail (Sylvilagus audubonii), and a house mouse (Mus musculus) were burnt [22]. It is unclear whether the carcasses were charred before or after death.
Observations of small mammals before, during, and after fire suggest that most small mammals are able to escape fire by burrowing into the soil [99]. However, rodents may be susceptible to the heat effects of fire when soil temperatures are high. For instance, soil temperatures reaching or exceeding 145 °F (63 °C) appear to almost always be lethal to small mammals [52]. Rodents buried up to 6 inches (15 cm) died from lethal temperatures resulting from fire. Temperatures were especially high under fallen logs and near burning stumps. Rodents also died in rock crevices and under ceanothus when the temperature reached at least 140 °F (60 °C). Similarly, rats buried 1.5-5 inches (4-13 cm) below the soil surface in an open sunny site outside the burned area died when temperatures from solar radiation alone reached at least 145 °F (60 °C). However, a rat that was buried 7 inches (18 cm) under the surface survived where temperatures only reached 138 °F (59 °C) [52]. No data were found on brush mouse survival in burrows during fire.
In addition to lethal temperatures, small mammals may perish from exposure to flames. Deer mice (P. maniculatus), pinyon mice (P. truei), Townsend's chipmunks (Tamias townsendii), Trowbridge's shrews (Sorex trowbridgii), and dusky-footed woodrats (Neotoma fuscipes) were seen fleeing from burning slash piles in a prescribed fire in California [106]. Woodrats in the fire waited until their cover was fully engulfed in flames before escaping. Some woodrats caught on fire from burning pitch that dropped onto them while they were hiding. Such individuals appeared to run in a panicked manner, starting small fires along the way. Woodrats not on fire tended to run from one patch of cover to the next in a less erratic fashion. Several woodrats perished either from catching on fire or by refusing to leave their homes [106]. Although observations of mouse (Peromyscus spp.) behavior during the fire was limited, it is possible that they may behave in a similar manner to the woodrats.
Immediately after the fire, approximately one third of the original number of mice captured on the site were recaptured [106]. Nearly all were collected from the edge of the burn. Individuals captured within the burn were collected from islands which did not catch fire. Thus, all the mice that were not able to find cover outside the fire appear to have either perished or fled the area entirely [106]. Fire effects on the brush mouse may be comparable because of the biological similarities between brush mice and other Peromyscus species.
Brush mice are semiarboreal and can be found foraging in shrubs and trees for leaves and fruits [35,49,51,56]. Females were captured more often than males foraging in canyon live oaks (Q. chrysolepis) (P<0.05) [69]. The individuals with the longest tails appear to spend more time climbing than those with shorter tails [35].
Acorns are commonly eaten by brush mice wherever they are available [10,17,26,54,57]. Arthropods and cutworms (Protorthodes rufula) are also eaten throughout the year [31,41,54,56,98]. A variety of fruits and seeds from Douglas-fir, ponderosa pine (Pinus ponderosa), pinyon, California buckeye, manzanita (A. patula and A. viscida), silktassel (Garrya spp.), oneseed juniper (Juniperus monosperma), hackberries (Celtis spp.), New Mexico groundsel (Senecio neomexicanus var. neomexicanus), trailing fleabane (Erigeron flagellaris), annual sunflower (Helianthus annuus), broom snakeweed, common dandelion (Taraxacum officinale), western yarrow (Achillea millefolium), white sweetclover (Melilotus albus), threenerve goldenrod (Solidago velutina), prickly-pear, desert wheatgrass (Agropyron desertorum), Kentucky bluegrass (Poa pratensis), and blue grama (Bouteloua gracilis) are eaten throughout the year when available [17,26,31,41,54,56,57,98]. Other plant parts, such as leaves, stems, flowers, pollen cones and new sprouts are typically eaten in lower quantities than other foods [31,41,54,56,98]. Fungi are typically consumed when other foods are scarce [54,56]. Infrequently, stomach contents of brush mice contained pieces of mammals, birds, and fence lizards (Sceloporus spp.) [54].
Brush mice have been observed caching pinyon pine seeds [78,79]. This observation suggests that the brush mouse may play a role in seed dispersal for some plant species.
During the 2.5 weeks after a prescribed fire in California, 13 mice (P. maniculatus or P. truei) were captured [106], or approximately one third the original number of captures before the fire. The deep ash covering the burned area prevented the movement of the mice. Several days after the fire, heavy rain hardened the surface of the ash. Within 5 days, 51 mice were captured within the burned area [106]. Thus, fire-altered habitat appears to restrict the movement of mice until postfire conditions change.
Brush mice in New Mexico preferred sites with significantly lower tree cover (P<0.01), lower conifer density and basal area (P<0.001) and less litter (P<0.001) than random sites [111]. This suggests that fire may benefit brush mice by keeping tree cover and litter depth low. However, they also preferred sites with significantly higher shrub and oak sapling and seedling densities (P<0.001) [111]. This result suggests that fires removing low shrub and tree cover would be detrimental to the brush mouse. In an Arizona study, small mammals were sampled in burned and unburned chaparral and pine-oak forest from 14 months to 3 years after fire [73]. The wildfire burned more than 90% of the vegetation that was present before the fire. More than twice as many brush mice were captured in the unburned forest habitat than any other habitat sampled. The lowest number of captures came from the unburned chaparral. The burned habitats may have recovered sufficiently 3 years after fire, allowing brush mice to reinvade the area [73].
Fire may alter the habitat structure and/or the community composition, which may influence brush mouse populations [67]. Brush mice are able to recolonize areas that were burned previously. For instance, in a chaparral habitat of southern California, brush mice recolonized sites 3 years and 4 months after wildfire or prescribed fire [116]. Presumably, the fires severely altered the chaparral habitat so that it was unsuitable to brush mice for more than 3 years. Furthermore, brush mice occupied fire-caused secondary forest growth in dry forest habitats of southern Mexico [88]. Additionally, brush mice were captured in a former mixed conifer woodland dominated by white fir, Douglas-fir, incense-cedar (Calocedrus decurrens), sugar pine (Pinus lambertiana), and ponderosa pine that had succeeded to a California black oak (Q. kelloggii)-manzanita-ceanothus-Fremont silktassel (Garrya fremontii) association 19-38 years after fire [53].
FIRE REGIMES: The following table provides fire return intervals for plant communities and ecosystems where the brush mouse is important. Find fire regime information for the plant communities in which this species may occur by entering the species name in the FEIS home page under "Find FIRE REGIMES".
Community or Ecosystem Dominant Species Fire Return Interval Range (years) silver fir-Douglas-fir Abies amabilis-Pseudotsuga menziesii var. menziesii >200 [3] California chaparral Adenostoma and/or Arctostaphylos spp. 77] silver sagebrush steppe Artemisia cana 5-45 [47,83,118] sagebrush steppe Artemisia tridentata/Pseudoroegneria spicata 20-70 [77] basin big sagebrush Artemisia tridentata var. tridentata 12-43 [89] coastal sagebrush Artemisia californica <35 to <100 saltbush-greasewood Atriplex confertifolia-Sarcobatus vermiculatus <35 to <100 desert grasslands Bouteloua eriopoda and/or Pleuraphis mutica 5-100 [77] plains grasslands Bouteloua spp. <35 blue grama-buffalo grass Bouteloua gracilis-Buchloe dactyloides <35 [77,118] grama-galleta steppe Bouteloua gracilis-Pleuraphis jamesii <35 to <100 blue grama-tobosa prairie Bouteloua gracilis-Pleuraphis mutica <35 to <100 California montane chaparral Ceanothus and/or Arctostaphylos spp. 50-100 [77] curlleaf mountain-mahogany* Cercocarpus ledifolius 13-1,000 [6,93] mountain-mahogany-Gambel oak scrub Cercocarpus ledifolius-Quercus gambelii <35 to <100 blackbrush Coleogyne ramosissima <35 to <100 Arizona cypress Cupressus arizonica <35 to 200 western juniper Juniperus occidentalis 20-70 Rocky Mountain juniper Juniperus scopulorum <35 creosotebush Larrea tridentata <35 to <100 [77] Engelmann spruce-subalpine fir Picea engelmannii-Abies lasiocarpa 35 to >200 blue spruce* Picea pungens 35-200 [3] pinyon-juniper Pinus-Juniperus spp. <35 [77] Rocky Mountain bristlecone pine P. aristata 9-55 [28,29] Mexican pinyon Pinus cembroides 20-70 [72,104] Rocky Mountain lodgepole pine* Pinus contorta var. latifolia 25-340 [3,12,13,105] Sierra lodgepole pine* Pinus contorta var. murrayana 35-200 [3] Colorado pinyon Pinus edulis 10-400+ [37,44,59,77] Jeffrey pine Pinus jeffreyi 5-30 Pacific ponderosa pine* Pinus ponderosa var. ponderosa 1-47 [3] interior ponderosa pine* Pinus ponderosa var. scopulorum 2-30 [3,8,64] Arizona pine Pinus ponderosa var. arizonica 2-15 [8,24,94] galleta-threeawn shrubsteppe Pleuraphis jamesii-Aristida purpurea <35 to <100 [77] Rocky Mountain Douglas-fir* Pseudotsuga menziesii var. glauca 25-100 [3,4,5] coastal Douglas-fir* Pseudotsuga menziesii var. menziesii 40-240 [3,76,87] California mixed evergreen Pseudotsuga menziesii var. menziesii-Lithocarpus densiflorus-Arbutus menziesii <35 California oakwoods Quercus spp. <35 [3] oak-juniper woodland (Southwest) Quercus-Juniperus spp. <35 to <200 [77] coast live oak Quercus agrifolia 2-75 [45] canyon live oak Quercus chrysolepis <35 to 200 blue oak-foothills pine Quercus douglasii-P. sabiniana <35 Oregon white oak Quercus garryana <35 [3] California black oak Quercus kelloggii 5-30 shinnery Quercus mohriana <35 interior live oak Quercus wislizenii <35 [77] *fire return interval varies widely; trends in variation are noted in the species reviewThe activities of humans since the arrival of Europeans have created more brushlands than previously existed, thus creating favorable habitat for the brush mouse [10]. Brush mouse populations appear positively correlated to the amount of downed woody debris present on a site [95]. In New Mexico, bulldozed and thinned sites had 2.5-3 times more slash (10% to 13% slash cover) than untreated sites (3% slash cover). These same sites had at least twice as many brush mice captured as the untreated site. A site that was bulldozed, piled, and burned (6% slash cover) also had a higher brush mouse population than the untreated site, but lower than the bulldozed or thinned only sites [95]. Similarly, a habitat assessment in California suggests that the removal of downed woody debris and the reduction of tree and shrub canopy cover to 40% during fuels treatments would have a negative impact on the brush mouse [15].
Brush mice are primary carriers of Sin Nombre virus, which causes hantavirus pulmonary syndrome in humans [42]. Because their presence in and around buildings has been documented [42], reducing habitat suitability by the removal of vegetation and wood piles around man-made structures may reduce the transmission of the disease to humans.
Vegetation in brush mouse habitats may vary from location to location, but brush mice are consistently captured in areas with medium to high densities of shrubs and tree cover under 16 feet (5 m) in height [2,25,31,36,48,49,73,75,82,103,111,114]. In California, mature chaparral (cover ≥50%) appears to provide more suitable habitat for brush mice than young, open chaparral (cover <50%) [82]. Similarly, in Arizona, Duran [31] captured brush mice most frequently in shrub live oak and birchleaf mountain-mahogany (Cercocarpus betuloides) understory habitats with 45% to 50% plant cover. Fewer brush mice were captured in habitats with less plant cover [31]. Holbrook [50] observed that after vegetation crowns were removed in a manzanita (Arctostaphylos spp.)-oak shrubland, brush mice avoided the newly-opened space. In another study, brush mice were strongly restricted to habitats in which gaps between rocks or ceanothus (Ceanothus spp.) on the site were less than 4 feet (1.2 m) [81].
In addition to shrub density, the height of cover appears to influence brush mouse distribution within a site. An average understory height of 5-6.5 feet (1.5-2.0 m) was preferred by brush mice over lower understory cover [97]. In another study, brush mouse presence was positively correlated with microhabitats of shrub cover up to 10 feet (3 m) tall, logs over 3 inches (7.5 cm) in diameter, and understory trees 10-33 feet (3-10 m) in height, but negatively correlated with grass-forb microhabitats [16].
Brush mice are also commonly captured at locations with a high proportion of rock cover and/or slash piles in habitats characterized by chaparral-mountain shrub, oak/shrub, oak-juniper-pinyon pine, juniper-pinyon pine, and oak-pine communities, as well as riparian habitats [1,2,25,27,30,36,38,40,41,48,49,50,51,66,73,103,110,111,114]. The brush mouse in Texas has been found in all major habitats present (desert, grassland, riparian, and montane) although it is typically associated with rock outcrops within these habitats [25]. Riparian sites with abundant brush mouse populations had high shrub cover, high frequency of debris piles with low grass, litter, and tree cover [32]. In a Mexico study, a canyon was dominated by exposed rock, grasses, pines, hardwoods, and brush [30]. In western Texas, brush mice favored fallen logs and brush piles [26]. Modi [71] discovered that brush mice were common in riparian zones dominated by pecan (Carya illinoensis), American sycamore (Platanus occidentalis) and live oak (Quercus virginiana), in an oak community with a partially open canopy and dense understory, and in a pine forest with little understory and scattered boulders. In New Mexico, brush mouse populations were significantly higher (P<0.05) on sites that were bulldozed or thinned (98 and 115 captures, respectively) than untreated or bulldozed and burned sites [95]. Populations were lowest on sites that had not been treated (45 captures). Sites that had increased slash from bulldozing and burning had more brush mice (57 captures) than the untreated sites, but the difference was not significant (P>0.05). No influence of canopy cover on brush mice was observed in the study [95].
Besides high tree, shrub, and rock densities, brush mice appear to prefer locations with low grass cover [32,81]. At the same time, grasses are often present in the understory indicating that grasses do not exclude brush mice [25,40]. Brush mice utilized grazed and ungrazed pastures and ceanothus plots [81], but they were concentrated around rocky outcrops and vegetation continuous with the rock outcrops. No brush mice were captured in the grasslands more than 20 feet (6 m) from rocks, shrubs, or trees [81]. Litter depth also appears negatively correlated to brush mouse presence [97,111]. For instance, brush mice in Arizona were captured in litter depths of only 0.9 inches (2.4 cm) [111].
Brush mice also utilize fire-affected habitats. In one study, brush mice were captured in burned and unburned chaparral as well as burned and unburned pine-oak forest. The highest number of captures were recorded in unburned forest while the lowest captures occurred in the unburned chaparral [73]. These results are somewhat inconsistent with other observations which show the brush mouse favoring dense chaparral habitat. Small mammal capture data in the study were collected from 14 months to 3 years after fire [73]. The time frame of sampling after fire may influence the perceived response of the brush mouse to postfire habitats.
Elevation, in addition to habitat characteristics, may play a role in habitat suitability in some areas. For instance, in the northern Sierra Nevada of California, brush mice were captured in brush habitats at 3,500-5,000 feet (1,000-1,500 m), but not at 6,500 feet (2,000 m) [53]. Aspect may influence the distribution of brush mice on a site as well. For example, in New Mexico, 51% of all brush mice captured were taken on south-facing slopes, 24% on west-facing slopes, with 13% and 12% of mice captured on east- and north-facing slopes, respectively [114]. The south-facing canyon slopes may provide more cover for brush mice due to higher numbers of shrubs [114].
Site characteristics of brush mouse habitat:
State/Region Elevation Slope/Aspect Precipitation Citation Arizona 1,000-1,500 feet (300-450 m), 2,300-8,300 feet (700-2,500 m) canyon and valley bottoms, canyon and talus slopes, rolling hills, level uplands, east and south aspects 15.5-28 inches (393-710 mm) [16,18,21,31,34,41,42,43,50,73,74,85,86] California 1,500-6,600 feet (450-2,000 m), rare up to 9,800 feet (3,000 m) canyon bottoms, north slopes favored, but found on all aspects, brushy hillsides 10-44 inches (260-1110 mm) [18,53,54,56,57,97,110] Colorado up to 8,300 feet (2,500 m) hill slopes and valleys no data [1,103] New Mexico 5,200-8,000 feet (1,585-2,400 m), rare at 8,600 ft (2,600 m) south aspects often preferred, but all aspects utilized, low ridges and slopes, north aspects on hot arid sites, gradual slopes 12.5-18 inches (317-460 mm) [18,35,36,114] Utah 3,750-7,005 feet (1,150-2,135 m) canyons and mesas 7.5 inches (191 mm) [2,66] Texas 4,250-7,970 feet (1,295-2,430 m) south aspects, level plain no data [25,26,27,38,71,91]Although brush mice are found on a variety of slopes, including flat mesas and gradual slopes, they seem to prefer locations with very steep slopes, such as hillsides, mountain sides, and canyons (including some slopes with >45% gradient) over more gradual slopes in the same areas [1,30,40,41,73,97,101,111]. Findley [35] reported that brush mice were captured on hillsides in an oak/sacahuista (Nolina spp.) community. In another study, brush mice were common in canyon bottoms, on hillsides, and in arroyos (water channels in arid regions) characterized by oak woodlands [36]. Brush mice have also been captured along the sides of brush covered canyons and burned slopes of an oak/brush association [114].
Density and Home Range
Kalcounis-Ruppell [57] discovered brush mouse population densities were 40-72 mice per hectare in coastal California. Similarly, Kalcounis-Ruppell and Millar [58] determined that densities of brush mice in coastal California were 42-89 mice per hectare. Brush mouse densities appear greatly influenced by weather. Densities were 17-20 per acre one year, but were reduced to 6 mice per acre following a severe winter [41]. At another study site, brush mice increased from 4 individuals per acre to 12 per acre after a mild winter [41]. Populations in a canyon in Mexico were estimated at a mean of 10.8 brush mice per acre or 6.0 males and 4.8 females per acre [30]. The range was 9.0-14.0 brush mice per acre [30].
A study utilizing radiotelemetry and trapping data in Arizona estimated mean home range size for male brush mice at 0.72-1.6 acres (0.29-0.64 hectare), and 0.32-0.79 acre (0.13-0.32 hectare) for females [86]. In a another study at the same location, home ranges for male brush mice ranged from means of 1.2-1.5 acres (0.47-0.62 hectare) and for females, means of 0.64-1.2 acres (0.26-0.49 hectare) [85]. Mean home range for male and female brush mice in Arizona, based on radiotelemetry, was 0.30 acre (0.12 hectare) [42].
The scientific name of brush mouse is Peromyscus boylii Baird [7,115]. It is a
member of the mouse, rat, and vole family, Muridae [9]. Subspecies in the United
States include:
P. b. subsp. rowleyi (Allen) [46]
P. b. subsp. utahensis Durrant [46]
In the Timing Of Major Life History Events section
of this review, some reproductive information specific to the brush mouse was
not available. In these cases, information on the entire genus is presented.
Additionally, in the Fire Effects And Use section,
observations of mice from the genus Peromyscus are presented, although the
mice were not identified to species in those publications.
Brush mice are nocturnal [26,31,54,63,70,100]. Brush mice appear to limit activity during cooler months in the Sierra Nevada of California [100], although they are not known to hibernate or enter torpor [113]. In southern areas, such as Arizona, Texas, and northern Mexico, brush mice are active year-round [25,26,38,48,92].
Brown [19] reported that brush mice reach sexual maturity at 12-19 weeks. However, Clark [23] reported that female brush mice may reach sexual maturity in approximately 5-9 weeks. In Arizona, females born in late May or June were found pregnant during August of the same year [48]. In western Texas, Arizona, southern Utah, and Mexico, breeding appears to occur nearly year-round [25,38,48,66,92]. However, in northern California, Jameson [55] found that breeding in brush mice peaked twice each year, once in late spring and again in late summer. This result may be related to food availability [55,56].
Female brush mice carry 1-6 embryos each [11,17,48,55,66,100] with an average of 3-4 embryos [11,17,48,55]. Lactating females may become pregnant [48], although lactation may slow the development of embryos in Peromyscus species [102]. Bradley and Schmidly [17] reported that the gestation period of the brush mouse is around 23 days. Females may have multiple litters per year [17,25]. The amount of time between litters is likely similar to other Peromyscus species and may be 25-31 days [100]. Young are weaned at 3-4 weeks of age [17].
Brush mice have a promiscuous mating system [57,85]. Four out of 7 litters in a California study were fathered by several males [57]. Males and females did not share nests and mating pairs did not remain together for long periods of time [57].
Few Peromyscus spp. mice live longer than 6 months in the wild, but individuals may live up to 4-5 years in captivity [80].
Brush mice have a negative response to fire initially. However, after vegetation begins to recover, brush mice are able to recolonize burned habitats. Over the long term, fire is beneficial to brush mice unless plant cover is greatly reduced or habitat structure is altered. Detrimental habitat changes may include a reduction in shrub height or shrub cover (see Preferred Habitat).
Frequent fires would likely be necessary to keep brush mouse populations low. This goal would be valuable when the spread of hantavirus to humans was a concern. On the other hand, if increasing brush mouse populations as a food source for predators such as the spotted owl were necessary, then less frequent fires may be desirable.
