(Peromyscus Maniculatus), Red-Backed Vole (Myodes Gapperi), and Woodland Jumping Mouse (Napaeozapus Insignis) Populations

Canadian Journal of Zoology

The short -term impact of abundant fruit upon deer mouse (Peromyscus maniculatus), red-backed vole (Myodes gapperi), and woodland jumping mouse (Napaeozapus insignis) populations

Journal: Canadian Journal of Zoology

Manuscript ID cjz-2015-0234.R1

Manuscript Type: Article

Date Submitted by the Author: 06-Apr-2016 Complete List of Authors: Dracup, Evan;Draft University of New Brunswick, Department of Biology Keppie, D.M.; University of New Brunswick, Forestry and Environmental Management Forbes, Graham; University of New Brunswick, Forestry and Environmental Management

RODENTS;SHREWS < Taxon, CARBOHYDRATES < Organ System, FOREST Keyword: < Habitat

https://mc06.manuscriptcentral.com/cjz-pubs Page 1 of 30 Canadian Journal of Zoology

The short-term impact of abundant fruit upon deer mouse ( Peromyscus

maniculatus ), red-backed vole ( Myodes gapperi ), and woodland jumping

mouse ( Napaeozapus insignis ) populations

Evan C. Dracup a, *, Daniel M. Keppie ab , Graham J. Forbes ab

a Biology Department, University of New Brunswick, P.O. Box 4400, Fredericton, NB, E3B 5A3, Canada

bFaculty of Forestry and Environmental Management, University of New Brunswick, P.O. Box 4400,

Fredericton, NB, E3B 5A3, Canada

*Corresponding author E-mail: [email protected], Ph: 1-506-470-7478

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 2 of 30

Abstract

Fruit has been identified as an important and potentially population restricting food for red-backed

voles (Myodes gapperi (Vigors, 1830)), deer mice (Peromyscus maniculatus (Wagner, 1845)), and

woodland jumping mice (Napaeozapus insignis (Miller, 1891)). We added domestic dried strawberries

and currants—which have native analogues and are preferred foods of these rodents—to white spruce-

plantations from May through August 2011 and 2012 to test fruit and fruit based carbohydrate’s short-

term (1-2 year) impact on these rodent populations. We used mark-recapture to estimate density,

percentages of population that were juvenile and breeding female, average home range size, and body

weight during spring and summer of both years, and fecundity via placental scars from euthanized

females in summer 2012. Fruit enhancement had no apparent effect on our species’, fecundity,

proportion of breeding females or juvenilesDraft during spring and summer of either year, nor were there

differences among these metrics in spring 2012 following 2011 fruit additions. Overall, there were no

impacts to the short-term adult population dynamics for any species during fruit addition. We are led to

believe that short term pulses of fruit and/or fruit based carbohydrate abundance do little to influence

temperate forest small mammal populations.

Keywords: food addition; fruit; small mammals; southern red-backed vole; Myodes gapperi ; woodland jumping mouse; Napaeozapus insignis ; deer mouse; Peromyscus maniculatus

https://mc06.manuscriptcentral.com/cjz-pubs Page 3 of 30 Canadian Journal of Zoology

1. Introduction

Choice of foods by small mammals (e.g., Hamilton 1941; Vickery 1979; Vickery et al. 1994), and

what happens when food availability and quality change (e.g. Gilbert and Krebs 1981; Taitt 1981; Taitt

and Krebs 1981; Krebs et al. 1986; Boutin 1990) have been well explored. Yet, most studies assessing

food effects on small rodents have used artificial food, or food not normally available: oats, peanuts,

sunflower seed, pet food, etc. (Wagg 1963; Fordham 1971; Gilbert and Krebs 1981; Taitt 1981; Taitt and

Krebs 1981; Ims 1987; Schweiger and Boutin 1995; Boonstra et al. 2001). Although logistically easy to

proffer, artificial foods traditionally provide a relatively well-balanced diet (e.g., oats, sunflower seed)

(Kent-Jones and Amos 1967) compared to natural food, which hampers direct relation to natural food

dynamics. Draft Furthermore, small rodent responses to artificial food enhancement have been wide ranging.

Karels and Boonstra (2000), and Karels et al. (2000) added food to arctic ground squirrel ( Spermophilus

parryii plesius Richardson 1825) populations in south-west Yukon and observed 4-7 fold density

increases, and 19 fold increases when predators were excluded, Gilbert and Krebs (1981) added food

(sunflower seeds) over two summers in southern Yukon and observed a doubling to tripling of deer

mouse ( Peromyscus maniculatus (Wagner, 1845)) and northern red-backed vole ( Myodes rutilus (Pallas,

1779)) density from enhanced immigration and juvenile survival; however, adult survival, breeding

period, and animal weight were not changed. Adding whole oats in the Fraser River Delta, south of

Vancouver, generated a doubling in abundance of Townsend’s Vole ( Microtus townsendii (Bachman,

1839)) (Taitt and Krebs 1981) and Peromyscus maniculatus (Taitt 1981). However, aside from the

hibernating arctic ground squirrel, these studies showed limited carry-over effect into the following

spring. Schweiger and Boutin (1995) added sunflower seed over two winters and found northern red-

backed vole populations persisted better through winter into spring with added food, but by autumn,

vole densities in added food sites were again similar to control sites. Along with density effects, home

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 4 of 30

range sizes tend to decrease following food addition (Taitt 1981; Taitt and Krebs 1981; Ims 1987; Hubbs

and Boonstra 1998; Rémy et al. 2013).

Alternative to adding food, tracking natural food abundance and relating it to animal population

dynamics avoids addition of foreign elements, but conclusions inevitably remain correlative

observations that are potentially circumstantial (Krebs and Myers 1974; Boutin 1990). Deer mice,

southern red-backed voles ( Myodes gapperi (Vigors, 1830)), and woodland jumping mice ( Napaeozapus

insignis (Miller, 1891)) eat berries when available, and berries comprise a large portion of their

respective diets (Vickery 1979; Van Horne 1982; West 1982; Vickery et al. 1994). Although deer mice

maintain an expansive diet that is not likely regulated by berries(Falls et al. 2007), red-backed vole and

woodland jumping mouse populations may be influenced considerably by berry abundance (Whitaker Jr.

1963; Vickery 1979; Martell 1981). BoonstraDraft and Krebs (2012) have published summaries on limiting

factors of Myodes spp. population dynamics and the perceived importance that berry crop has for

Myodes spp. , and particularly the northern red-backed vole (Krebs et al. 2010) They predict that

Myodes spp. population fluctuations and demographics are tied to natural patterns of berry crop

(Boonstra and Krebs 2006, 2012; Krebs et al. 2010).

We built upon Boonstra and Krebs’ (2012) hypothesis that Myodes spp. demographics are affected by natural berry crop and, in particular, fruit-based carbohydrate content; further we were able investigated fruit based carbohydrate’s impact on the woodland jumping mouse and deer mice, known consumers. To do so, we added domestic fruit that had local natural analogs to study the short- term impacts of abundant fruit and carbohydrates on rodent populations. Deer mice and red-backed voles typically are short-lived (usually <1 year), and can have multiple litters through spring and summer; therefore, if fruit is an important part of their diet, then they should be capable of exhibiting a response during our study period (Manville 1949).

https://mc06.manuscriptcentral.com/cjz-pubs Page 5 of 30 Canadian Journal of Zoology

We hypothesized that fruit enhancement will affect small rodent density and population

characteristics; predicting that berry crop enhancement during spring and summer would (P 1) encourage

higher population densities within the current and following year, and (P 2) that enhanced high-energy

fruit sources will increase individual body condition and juvenile recruitment.

2. Methods

2.1. Study Design

We conducted the experiment in J.D. Irving, Limited’s, Black Brook forest management district

(47 o9’51”, 67 o55’27”) located directly east of St. Leonard, New Brunswick, within the northern end of

the Appalachian Mountain Range (Figure 1). We chose six ≥10 ha newly commercially thinned white

spruce ( Picea glauca (Moench) Voss) plantationsDraft (>75% white spruce), each spatially separated by ≥3.75

km. Plantations were 25-30 years of age after planting, and experienced similar management history:

scarified clear-cut, planted with white spruce, not fertilized, sprayed with herbicide around age 3-5, pre-

commercially thinned between ages 10-15, and commercially thinned between age 25 and 30, one

calendar year prior to this study (for more detailed information on plantation structure see,Dracup et al.

2015; MacLean et al. 2015). The intensive and regimented management history of the plantations made

them closer to true replicates than could be found in highly varied natural forest. Generally, plantations

with this management intensity are more uniform in vegetation structure, species composition, canopy

closure, and woody debris structure than natural stands (Freedman et al. 1994, 1996; Ramovs and

Roberts 2003, 2005; Maclean et al. 2015). Further, plantations with such intensive management

histories and short rotation times generally fail to develop complex shrub layers, which may reduce

fruiting plant abundance below that found in undisturbed sites (Ramovs and Roberts 2003; 2005).

However, commercially thinned plantations are structurally comparable to natural insect outbreaks and

wind-throw events(Kneeshaw et al. 2011). Therefore, an observed rodent response following fruit

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 6 of 30

addition should be better attributed to fruit directly, rather than to varying site structure. Capture rates

of deer mice and red-backed voles in commercially thinned plantations compare well to capture rates

from studies in mature softwood stands (0-13.5/100 trap nights) (Bowman et al. 2001; Fuller et al.

2004).

Two treatment blocks of ≥5 ha were sectioned from each of the six commercially thinned

plantations, ensuring treatment centers of these two blocks were ≥225 m apart. We added dried

commercial strawberries ( Fragaria x ananassa ) and currants ( Ribes nigra ) to one treatment block for an

“Added-Fruit” treatment and the second acted as a control. Strawberries were added throughout the

study, and currants only during summer 2012 as a supplement. Wild strawberries and currants are

known to be selected by deer mice and red-backed voles (Hamilton 1941; Vickery 1979; Martell 1981;

Martell and MacAulay 1981), and are commonDraft native plants within the region (e.g., Fragaria virginiana,

Ribes lacustra ).

We established nine feeding stations 25 m apart in a 75x75 m grid centered within each 100 m 2 animal trapping grid in each Added-Fruit treatment block within the six plantations. Small rodent average daily movements of 33 m were calculated from a concurrent study in similar, neighboring commercially thinned plantation (Dracup et al. 2015); hence, we assumed the average small rodent had access to at least one feeding station within its typical daily movements. Feeding stations were modified Victor Tin Cat multi-capture traps (6.1x16.6x27.9 cm) with treadles removed (Woodstream

Corp., P.O. Box 327, Lititz Pennsylvania, U.S.A., 17543-0327). The small opening in Tin Cat traps allowed mice and voles to enter and leave at will with treadles removed, and prevented access of larger animals.

We added 750 g of dried fruit to each of the six fruit-addition-treatments one week prior to the start of spring trapping, 2011 and 2012. Fruit was subsequently replenished ad libitum on a weekly schedule from May to September, maintaining a constant average of 750 g of added fruit per trapping grid during

https://mc06.manuscriptcentral.com/cjz-pubs Page 7 of 30 Canadian Journal of Zoology

2011, and 1.5 kg during 2012. Feeding stations were cleaned and moved 2-3 m if insect infestations

were discovered.

2.2. Animal Trapping

Each treatment was trapped for five straight days in spring (15 May-20 June) and again in

summer (25 July-31 August), 2011 and 2012, using medium sized (7.5x8.75x22.5 cm) Sherman

galvanized folding live traps (Sherman Traps, 3731 Tallahassee, Florida U.S.A., 32303). We used a square

100-trap grid with 10 m spacing between traps to maximize recapture likelihood (Bergeron and Jodoin

1989; Jorgensen 2004). Unsalted peanuts were used as bait, and a 25 cm 2 piece of unbleached organic

cotton was provided for nesting material during May to reduce cold-temperature mortality of captured

animals. Traps were checked one hour after dawn, and again in mid-afternoon during daily high

temperatures to avoid heat stress. We didDraft not pre-bait traps. Captured mice and voles were identified

to species, weighed, sexed, and assessed for sexual development (testes absent or fully descended;

teats absent or conspicuous/lactating). We ear-tagged individuals with a unique Monel tag (National

Band and Tag #1005-1; 721 York St., Newport Kentucky, U.S.A., 41072-0430) for identification of

individuals. Handling animals followed guidelines set forth by the American Society of Mammalogists

(Gannon and Sikes 2007) and the University of New Brunswick Animal Care Committee (permit numbers

11033 for 2011 and 12028 for 2012) (Animal Care Committee 2011).

2.3. Density Calculation

We used Density 5.0 software (Efford 2012) (hereafter Density ) to calculate deer mouse, red-

backed vole, and woodland jumping mouse density per hectare and average home range radius (m) per

each treatment per trapping session (Krebs et al. 2011; Efford 2012). Animals that move large distances

are assumed to have larger home ranges than animals that move short distances (Borchers and Efford

2008) because rich habitat quality typically provides food, shelter, and mates within close proximity

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 8 of 30

(Bergeron and Jodoin 1987; Bondrup-Nielson and Ims 1988). Density calculates capture likelihood as a descending function moving away from home range center, assuming the more an animal moves the more likely it will be captured (Efford 2004). Density estimated average home range radii per trap-grid per trapping session per species using all captures/recaptures from all animals based on their combined capture probability and distance moved (Efford 2004).

We used maximum likelihood with a Jackknife estimator for heterogeneity of capture probability to calculate density for all scenarios (Burnham and Overton 1978; Hammond and Anthony

2006). We used Chao’s estimator for heterogeneity of capture probability when the capture/recapture rate was too poor to use the Jackknife estimator (Chao 1987; Hammond and Anthony 2006).

Populations were assumed to be closed during the five days of trapping, and open between trapping sessions, following White et al. (1982). Draft

2.4. Population Characteristics

We used four interrelated metrics to assess the short-term effects of fruit addition on rodent populations: placental scar count, percent of total population that are breeding females (females with placental scars, pregnant, or lactating), percent of total population that are juvenile (male and female combined), and average adult animal weight. Taken together, these metrics provide a reasonable representation of body condition of individuals, and juvenile recruitment into the population (Martin et al. 1976, West 1982).

Although we were not able to sample fecundity directly, placental scars do provide an index to fecundity (Rolan and Gier 1967; Martin et al. 1976; West 1982). There are sources of error with placental scars, but we assumed that bias in absorption of scar pigment or embryos to be equal between treatments and among plantations (Rolan and Gier 1967; Martin et al. 1976). Females used for scar counts included animals killed incidentally during 2011 and 2012 trapping, and females captured on the

https://mc06.manuscriptcentral.com/cjz-pubs Page 9 of 30 Canadian Journal of Zoology

final day of trapping during 2012 that were euthanized. Females were euthanized with halothane or

chloroform as directed by animal care regulations (Animal Care Committee 2011). Once euthanized, we

counted scars using a dissecting microscope, placing the uterus against a white backdrop (Rolan and

Gier 1967). We used juvenile/sexually mature weight cut-offs determined by Dracup et al. (2015) to

classify each individual as either juvenile (not sexually-mature) or adult (sexually mature): deer mice

mature at 16 g, and red-backed voles and woodland jumping mice at 18 g. Further, male and female

individuals appeared to mature at similar weights (Dracup et al. 2015), and we assumed that male and

female animals gain weight equally when exposed to food addition (Boutin and Larsen 1993).

2.5. Wild Fruit Sampling

Wild berry biomass per hectare was estimated in 2011 and 2012 from 20 systematically placed

10 m 2 circular plots per trapping grid in ControlDraft and Added-Fruit treatments each year. Berries within

the plot were identified to species, counted, and weighed at maturity. Nutrient content for each fruit

species was estimated based on known nutrient compositions (Spinner and Bishop 1950; Smith et al.

1956; Atkeson and Johnson 1979), and then extrapolated to estimate total fruit-based carbohydrate

load per hectare per treatment including carbohydrate content from added fruit. Fruiting-plant ground

cover was estimated from 20 randomly placed 1x1 m quadrats per trapping grid. Sample sizes of berry

biomass plots and cover plots were chosen to achieve a post-hoc power rating ≥0.83 with α = 0.10.

We considered that wild fruit may be attractive to rodents for its water content as opposed to

its nutrient composition because rodent habitation can be limited by water availability (Miller and Getz

1977) and a large portion of wild fruit biomass is water (McManus 1974). We correlated rodent

densities with wild fruit biomass per hectare, and percent ground cover of fruiting plants known to be

used by rodents: Aralia nudicaulis L., Ericacea, Cornus spp., Fragaria spp. , Liliaceae, Ribes spp. , Rubus

spp. , and Sambucus racemosa L. (Vickery 1979; Martell and MacAulay 1981; Martell 1983).

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 10 of 30

2.6. Statistical Analyses

We implemented a randomized block design with six spatially separate blocks (separate

plantations) each containing both treatments to avoid pseudoreplication; the randomized block design

reduces effects from spatial variation across a landscape (Hurlbert 1984). We set a conservative alpha level of 0.10 to interpret null hypotheses because Type II errors are more risky with a small number (i.e.,

6) of treatments blocks (Sokal and Rohlf 1995).

Analyses used SYSTAT 13. We used repeated-measures, general linear mixed models to test animal densities, proportion of juveniles as a percent of total population, proportion of breeding females as a percent of total population, and adult animal weight over spring and summer of 2011 and

2012 (n = 24 for each test: 6 treatment block replicates x 4 trapping sessions [spring 2011, summer

2011, spring 2012, summer 2012]). PopulationsDraft likely differed among our six experimental plantations, so we included plantation as a random factor in our models.

We compared summer animal densities with wild fruiting plant ground cover, fruit biomass (wild and artificial combined), and total fruit carbohydrate content (wild and added fruit combined) using linear regression (n = 12; 6 replicates x 2 years). Statistical power was assessed with power analysis tool in G-Power (Faul et al. 2007).

3. Results

We captured a total of 341 deer mice, 159 in the Control (1.6±1.7 individuals per 100 trap nights

n = 24 [6 replicates, 4 trapping sessions]), 182 in the Added-Fruit treatment (2.0±2.0 individuals per 100 trap nights n = 24); 446 red-backed voles, 217 in the Control (2.3±2.0 individuals per 100 trap nights n =

24), and 229 in the Added-Fruit treatment (2.5±1.9 individuals per 100 trap nights n = 24); and 166 woodland jumping mice, 87 in the Control (0.9±1.2 individuals per 100 trap nights n = 24), and 79 in the

https://mc06.manuscriptcentral.com/cjz-pubs Page 11 of 30 Canadian Journal of Zoology

Added-Fruit treatments (0.9±1.5 individuals per 100 trap nights n = 24). Post-hoc power for treatment

and trap session comparisons was estimated at ß ≥0.98 for deer mice, ß ≥0.96 for red-backed voles, and

ß ≥0.99 for woodland jumping mice. Rodent densities for each species did not differ between the

Added-Fruit and Control treatments in spring or summer 2011 or 2012 (Table 1, Figure 2). Deer mouse

and red-backed vole density increased from spring to summer 2011 equally in both treatments, and

woodland jumping mouse density did not change among trapping sessions (Table 1, Figure 2). There

were no significant interactions between trapping session or treatment for any species density estimate

(Table 1, Figure 2).

Home range radii ( n = 24; 6 plantations, 4 sessions) averaged 24.2±11.9 m and 23.8±9.9 m for

deer mice, respectively in Control and Added-Fruit treatments; and 23.1±11.0 m and 26.5±14.3 m,

respectively, for red-backed voles. NeitherDraft deer mouse nor red-backed vole home range radii differed

meaningfully between treatments (deer mouse: F3,5 = 0.12, p = 0.73; red-backed vole: F3,5 = 2.58, p =

0.12). Too few woodland jumping mice were captured to properly estimate home range size.

Proportion of juveniles did not differ between Added-Fruit and Control treatments for any

species (Table 2), nor did breeding females as a proportion of total animals (Table 3). Proportion of

juvenile deer mice increased through 2011 and 2012 (Table 2); however, proportion of breeding females

did not appear to change among trapping sessions (Table 3). There were proportionately more juvenile

red-backed voles during summer 2011 than during any other trapping session (Table 2), but the

proportion of breeding females did not change among sessions (Table 3). The proportion of juvenile and

breeding female woodland jumping mouse in the population did not differ throughout the trapping

period (Table 2, Table 3).

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 12 of 30

Placental scar counts among breeding females did not differ significantly between treatments for deer mice ( n = 17, F1,16 = 0.29, p = 0.60), or red-backed voles (n = 32 F1,31 = 0.18, p = 0.67). Too few mature female jumping mice were collected (n = 3) to reliably compare placental scar counts.

Adult deer mouse body mass averaged 18.1 g ±1.3, and did not differ between spring and summer or between the Added-Fruit treatment and Control (Table 4). Average adult woodland jumping mouse body mass averaged 23.7 g ±1.2, and did not differ between treatments, but differed significantly between spring (21.0 g ±1.1) and summer (26.4 g ±1.2) (Table 4). Adult red-backed vole body mass averaged 21.8 g ±1.1 and did not differ between treatments or seasons (Table 4).

Berry producing plants typically comprised 10% of ground coverage and included 13 species (Table 5).Nutrient content of wild fruit availableDraft to rodents in Black Brook, averaged by wet weight, was 4% carbohydrate, 16% fat, and 8% protein (Spinner and Bishop 1950, Johnson et al. 1985), while domestic strawberries and currants we added contained, respectively, 89% and 74% carbohydrate, 1% and 0.5% fat, and 1% and 4% protein (Bulk Barns Foods Ltd. 2012). Natural fruit biomass was highly variable between years and among plantations, being absent in half of the plantations across the two years (Table 6). On average, Added-Fruit treatments possessed more total fruit-based carbohydrate than paired Controls during 2011 and 2012 with a mean of 1880 g carbohydrates/ha from added and natural fruit combined compared to a mean of 278 g carbohydrates/ha from natural fruit in controls

(Table 6). Density of the three rodent species was not meaningfully related to either percent ground cover of fruiting plants (R2 ≤ 0.08, p ≥ 0.21), natural fruit wet weight (R2 ≤ 0.02, p ≥ 0.51), or combined

carbohydrate content of natural and artificial fruit (R2 ≤0.03, p ≥ 0.44).

https://mc06.manuscriptcentral.com/cjz-pubs Page 13 of 30 Canadian Journal of Zoology

4. Discussion

Previous food addition experiments witnessed 2-3 fold increase of mouse and vole abundances

within a year (Fordham 1971; Gilbert and Krebs 1981; Taitt 1981; Ims 1987; Boonstra et al. 2001).

However, we found no apparent response to fruit addition among our three rodent species. Although

natural fruit abundance varied considerably among plantations, the Added-Fruit treatment raised

average carbohydrate load by at least 1000 g/ha above that which occurred naturally, and was

consistently present during 2011 and 2012 from May through end of August. We cannot determine if

enough fruit was added to elicit a response from rodents, but a lot of fruit and additional carbohydrates

were introduced into the system. We expected to see an increased numberDraft of pregnant or lactating females in the added food treatment as carbohydrates are a preferred nutrient during that stage (Day et al. 2002). However, there

was no observed difference in pregnant or lactating female density in Control and Added-Fruit

treatments throughout spring and summer 2011 and 2012, so any benefit fruit proffered to pregnant or

nursing females had no apparently effect. Reproductively active animals were caught at the start and

end of trapping in 2011 and 2012, so we did not capture the entire breeding period. However, breeding

season length is determined primarily by availability of animal protein (Tabacaru et al. 2010), so our fruit

addition probably would not have changed its length.

Jumping mice require large energy stores for hibernation (Sheldon 1938), thus, we expected the

increased carbohydrate load in the Added-Fruit treatment to offer a substantial benefit. Hibernating

animals need to balance their intake of polyunsaturated fatty acids and Vitamin E to reduce hibernation

stress (Frank et al. 1998). Therefore, jumping mice preparing to hibernate may not have been as

concerned with gathering carbohydrates leading up to hibernation, as they were with balancing their

diet for hibernation.

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 14 of 30

The fruit stations were discrete individual locations that may have been guarded by aggressive female red-backed voles against other mice and voles (Rémy et al. 2013). Female red-backed voles are aggressively territorial toward other species and each other when breeding (Lovejoy 1973; Bondrup-

Nielson 1986; Ims 1987; Löfgren 1995). If aggressive individuals were able to defend territories around feeding stations then we would observe a fattening of individuals, which might not necessarily affect the local animal population (Boutin 1990). Anecdotally, 17 individuals weighed ≥30 g (10 voles, 6 jumping mice, and 1 deer mouse), and 7 heavy voles and 5 heavy jumping mice were caught in the Added-Fruit treatment, which tends to support feeding stations conferring benefits upon a few individuals. Our feeding stations conferred additional fruit to small and specific areas while naturally abundant fruit, or dispersed added food (e.g. Boonstra and Krebs 2006), may be more difficult for individuals to control, and could lead to large-scale population trendsDraft occurring (Ims 1987; Löfgren 1995; Rémy et al. 2013).

Previous food addition experiments commonly use whole oats, sunflower seeds, or pet chow: all of which are nutritionally well balanced. By weight, sunflower seeds are 51% fat, 20% carbohydrate, and 21% protein; whole oats are 7% fat, 66% carbohydrate, and 17% protein (United States Department of Agriculture 2012). Although our dried strawberries and currants had natural analogs, the dried fruit were mostly comprised of carbohydrate, while natural berries have a more balanced nutritional profile.

Furthermore, flowers and unripe fruit and seeds possess a unique and scarce suite of nutrients not normally available to herbivores, and may be very important to animal diets (White 2011). Our artificial fruit ignored a ripening stage, which may be one reason that the animals did not appear to respond to our fruit. Although we added fruit, it was nutritionally different than wild fruit, and may not have had the same influence on our rodent populations that natural fruit would have, and may be why our results are different than results reported in Krebs et al. (2010). Additionally, Bergeron and Jodoin (1987, 1989) found that mouse and vole abundance correlated well to plants high in protein. They speculated that protein is difficult to acquire for herbivores, so high protein content in vegetation should be strongly

https://mc06.manuscriptcentral.com/cjz-pubs Page 15 of 30 Canadian Journal of Zoology

selected. If protein is a limiting factor to population growth, then adding carbohydrate rich fruit may

not stimulate a population, whereas oats, sunflower seed, and pet chow may do so.

We did not directly measure consumption of fruit by rodents (e.g., motion activated camera,

etc.). Nevertheless, we think mice and voles were responsible for most fruit consumed. To determine if

animals were entering feeding stations we left trapping mechanisms active in several feeding stations

for the first two weeks of trapping in May 2011, and deer mice, red-backed voles, and jumping mice

were all captured in feeding stations. We observed insects feeding on the fruit, and ants infrequently

nested inside the feeding stations, but these losses were infrequent and seemed to be minor.

Numerous bird species and mammals, such as black bear ( Ursus americanus (Pallas, 1780)), and striped

skunk ( Mephitis mephitis (Schreber, 1776)) were present in the forest, but did not appear to visit or

damage feeding stations. Draft

In general, we did not observe a strong impact on animal populations as a result of

fruit/carbohydrate addition. Therefore, we are led to conclude that carbohydrate laden fruit on its own

does not impose any short term restrictions to animal populations in this environment. We did not

continue fruit addition through autumn and winter 2011, so we could not directly explore the

importance of fruit supply during winter. Food sources for rodents are limited over winter, but a

proportion of berries are able to survive under snow pack (West 1982), and we recommend future work

to explore overwinter effects of fruit addition as Schweiger and Boutin (1995) found positive population

effects from overwinter addition of sunflower seed. Further, our fruit was added via feeding stations to

small areas, and may have benefitted too few animals to cause an effect. We suggest a larger area

broadcast application rather than using feeding stations to avoid effects from territoriality and

aggression.

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 16 of 30

Acknowledgments

Funding was provided by Natural Sciences and Engineering Research Council, New Brunswick

Wildlife Trust Fund, and New Brunswick Innovation Fund. J.D. Irving, Limited, deserves special thanks

for in-kind forestry support as well as access to their lands and resources, and especially appreciation to

G. Adams, G. Pelletier, and P. Poitras for help with project design. We thank Cristina Gantor, Justin

Miller, Stephanie Seymour, Amanda Valois, and Marika Wheeler for their assistance with fieldwork. We

would also like to thank R. Boonstra for reviewing this manuscript.

Draft

https://mc06.manuscriptcentral.com/cjz-pubs Page 17 of 30 Canadian Journal of Zoology

Works Cited

Animal Care Committee. 2011. Animal Care, Ethics & Safety. University of New Brunswick, Fredericton, New Brunswick. Atkeson, T.D., and Johnson, A.S. 1979. Succession of small mammals on pine plantations in the Georgia Piedmont. Am. Midl. Nat. 101 : 385–392. Bergeron, J., and Jodoin, L. 1987. Defining “high quality” food resources of herbivores: the case for meadow voles (Microtus pennsylvanicus ). Oecologia, 71 : 510–517. Bergeron, J., and Jodoin, L. 1989. Patterns of resource use, food quality, and health status of voles (Microtus pennsylvanicus ) trapped from fluctuating populations. Oecologia, 79 : 306–314. Bondrup-Nielson, S. 1986. Investigation of spacing behaviour of Clethrionomys gapperi by experimentation. J. Anim. Ecol. 55 : 269–279. Bondrup-Nielson, S., and Ims, R.A. 1988. Predicting stable and cyclic populations of Clethrionomys . Oikos, 52 : 178–185. Boonstra, R., Krebs, C., Gilbert, S., and Schweiger, S. 2001. Voles and Mice. In Ecosystem Dynamics of the Boreal Forest - The Kluane Project. Oxford Press, Toronto, Ontario. p. 544. Boonstra, R., and Krebs, C.J. 2006. Population limitation of the northern red-backed vole in the boreal forests of northern Canada. J. Anim. Ecol.Draft 75 : 1269–1284. Boonstra, R., and Krebs, C.J. 2012. Population dynamics of red-backed voles ( Myodes ) in North America. Oecologia, 168 : 601–620. Borchers, D.L., and Efford, M.G. 2008. Spatially explicit maximum likelihood methods for capture- recapture studies. Biometrics, 64 : 377–85. Boutin, S. 1990. Food supplementation experiments with terrestrial vertebrates: patterns, problems, and the future. Can. J. Zool. 68 : 203–220. Boutin, S., and Larsen, K.W. 1993. Does food availability affect growth and survival of males and females differently in a promiscuous small mammal, Tamiasciurus hudsonicus . J. Anim. Ecol. 62 : 364–370. Bowman, J.C., Forbes, G.J., and Dilworth, T.G. 2001. Landscape context and small-mammal abundance in a managed forest. For. Ecol. Manage. 140 : 249–255. Bulk Barns Foods Ltd. 2012. Nutrition Facts. Richmond Hill. Burger, A.E. 1987. Fruiting and frugivory of Cornus canadensis in boreal forest in Newfoundland. Oikos, 49 : 3–10. Burnham, K., and Overton, W. 1978. Estimation of the size of a closed population when capture probabilities vary among animals. Biometrika, 65 : 625–633. Chao, A. 1987. Estimating the population size for capture-recapture data with unequal catchability. Biometrics 43 : 783–791. Conaway, C.H. 1955. Embryo resorption and placental scar formation in the rat. J. Mammal. 36 : 516– 532. Day, D.E., Mintz, E.M., and Bartness, T.J. 2002. Diet choice exaggerates food hoarding, intake and pup

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 18 of 30

survival across reproduction. Physiol. Behav. 75 : 143–157. Dracup, E.C., Keppie, D.M., and Forbes, G.J. 2015. Woodland mouse and vole response to increased structural diversity following midrotation commercial thinning in spruce plantations. Can. J. For. Res. 45 : 1121–1131. Efford, M.G. 2004. Density estimation in live-trapping studies. Oikos, 106 : 598–610. Efford, M.G. 2012. DENSITY 5.0: Software for spatially explicit capture-recapture. Department of Mathmatics and Statistics, University of Otago, Dunedin, New Zealand. Available from http://www.otago.ac.nz/density. Falls, J.B., Falls, E.A., and Fryxell, J.M. 2007. Fluctuations of deer mice in Ontario in relation to seed crops. Ecol. Monogr. 77 : 19–32. Faul, F., Erdfelder, E., Lang, A.-G., and Buchner, A. 2007. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav. Res. Methods 39 : 175–191. Fordham, R. 1971. Field populations of deermice with supplemental food. Ecology, 52 : 138–146. Frank, C.L., Dierenfeld, E.S., and Storey, K.B. 1998. The relationship between lipid peroxidation, hibernation, and food selection in mammals. Am. Zool. 38 : 341–349. Freedman, B., Woodley, S., and Loo, J. 1994. Forestry practices and biodiversity, with particular reference to the Maritime Provinces Draftof eastern Canada. Environ. Rev. 2: 33–77. Freedman, B., Zelazny, V., Beaudette, D., Fleming, T., Flemming, S., Forbes, G.J., Gerrow, J., Johnson, G., and Woodley, S. 1996. Biodiversity implications of changes in the quantity of dead organic matter in managed forests. Environ. Rev. 4: 238–265. Fuller, A., Lachowski, H., and Harrison, D.J. 2004. Stand scale effects of partial harvesting and clearcutting on small mammals and forest structure. For. Ecol. Manage. 191 : 373–386. Gannon, W., and Sikes, R. 2007. Guidelines of the American Society of Mammalogists for the use of wild mammals in research. J. Mammal. 88 : 809–823. Getz, L.L. 1962. Notes on the water balance of the redback vole. Ecology, 43 : 565–566. Gilbert, B.S., and Krebs, C.J. 1981. Effects of extra food on Peromyscus and Clethrionomys populations in the southern Yukon. Oecologia, 51 : 326–331. Hamilton, W. 1941. The food of small forest mammals in eastern United States. J. Mammal. 22 : 250– 263. Hammond, E., and Anthony, R. 2006. Mark-Recapture estimates of population parameters for selected species of small mammals. J. Mammal. 87 : 618–627. Van Horne, B. 1982. Niches of adult and juvenile deer mice ( Peromyscus maniculatus ) in seral stages of coniferous forest. Ecology, 63 : 992–1003. Van Horne, B. 1983. Density as a misleading indicator of habitat quality. J. Wildl. Manage. 47 : 893–901. Hubbs, A., and Boonstra, R. 1998. Effects of food and predators on the home-range sizes of Arctic ground squirrels ( Spermophilus parryii ). Can. J. Zool. 76 : 592–596. Hurlbert, S. 1984. Pseudoreplication and the design of ecological field experiments. Ecol. Monogr. 54 : 187–211.

https://mc06.manuscriptcentral.com/cjz-pubs Page 19 of 30 Canadian Journal of Zoology

Ims, R. 1987. Responses in spatial organization and behaviour to manipulations of the food resource in the vole Clethrionomys rufocanus . J. Anim. Ecol. 56 : 585–596. Johnson, R., Willson, M., Thompson, J., and Bertin, R. 1985. Nutritional values of wild fruits and consumption by migrant frugivorous birds. Ecology, 66 : 819–827. Jones, C., McShea, W., Conroy, M., and Kunz, T. 1996. Chapter 8: Capturing Mammals. In Measuring and Monitoring Biological Diversity: Standard Methods for Mammals. Smithsonian Institution Press, Washington, D.C. Jorgensen, E. 2004. Small mammal use of microhabitat reviewed. J. Mammal. 85 : 531–539. Karels, T.J., and Boonstra, R. 2000. Concurrent density dependence and independence in populations of arctic ground squirrels. Nature, 408 : 460–463. Karels, T.J., Byrom, A.E., Boonstra, R., and Krebs, C.J. 2000. The interactive effects of food and predators on reproduction and overwinter survival of Arctic ground squirrels. J. Anim. Ecol. 69 : 235–247. Kent-Jones, D., and Amos, A. 1967. Modern Cereal Chemistry. In 6th edition. Food Trade Press, London. Kneeshaw, D., Harvey, B., Reyes, G., Caron, M., and Barlow, S. 2011. Spruce budworm, windthrow and partial cutting: Do different partial disturbances produce different forest structures? For. Ecol. Manage. 262 : 482–490. Krebs, C.J., Boonstra, R., Gilbert, B.S., Reid,Draft D., Kenney, A., and Hofer, E. 2011. Density estimation for small mammals from livetrapping grids: rodents in northern Canada. J. Mammal. 92 : 974–981. Krebs, C.J., Cowcill, K., Boonstra, R., and Kenney, A. 2010. Do changes in berry crops drive population fluctuations in small rodents in the southwestern Yukon? J. Mammal. 91 : 500–509. Krebs, C.J., Gilbert, B.S., Boutin, S., Sinclair, A., and Smith, J. 1986. Population biology of snowshoe hares. I. Demography of food-supplemented populations in the southern Yukon. J. Anim. Ecol. 55 : 963– 982. Krebs, C.J., and Myers, J.H. 1974. Population cycles in small mammals. Adv. Ecol. Res. 8: 267–399. Löfgren, O. 1995. Spatial organization of cyclic Clethrionomys females: occupancy of all available space at peak densities? Oikos, 72 : 29–35. Lovejoy, D. 1973. Ecology of the woodland jumping mouse ( Napaeozapus insignis ) in New Hampshire. Can. Field-Nat. 87 : 145–149. MacLean, D.A., Dracup, E., Gandiaga, F., Haughian, S.R., Mackay, A., Nadeau, P., Omari, K., Adams, G., Frego, K.A., Keppie, D., Moreau, G., and Villard, M.-A. 2015. Experimental manipulation of habitat structures in intensively managed spruce plantations to increase their value for biodiversity conservation. For. Chron. 91 : 161–175. Manville, R.H. 1949. A study of small mammal populations in northern Michigan. Misc. Publ. Mich. Mus. Zool.No. 73 : 1–83. Martell, A. 1981. Food habits of southern red-backed voles Clethrionomys gapperi in northern Ontario Canada. Can. Field-Nat. 95 : 325–328. Martell, A. 1983. Demography of southern red-backed voles ( Clethrionomys gapperi ) and deer mice (Peromyscus maniculatus ) after logging in north-central Ontario. Can. J. Zool. 61 : 958–969.

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 20 of 30

Martell, A., and MacAulay, A. 1981. Food habits of deer mice ( Peromyscus maniculatus ) in northern Ontario. Can. Field-Nat. 95 : 319–324. Martin, K., Stehn, R., and Richmond, M. 1976. Reliability of placental scar counts in the prairie vole. J. Wildl. Manage. 40 : 264–271. McManus, J. 1974. Bioenergetics and water requirements of the redback vole, Clethrionomys gapperi . J. Mammal. 55 : 30–44. Miller, D.H., and Getz, L.L. 1977. Factors influencing local distribution and species diversity of forest small mammals in New England. Can. J. Zool. 55 : 806–814. Morris, R. 1955. Population studies on some small forest mammals in eastern Canada. J. Mammal. 36 : 21–35. Ramovs, B. V., and Roberts, M.R. 2003. Understory vegetation and environment response to tillage, forest harvesting, and conifer plantation development. Ecol. Appl. 13 : 1682–1700. Ramovs, B. V., and Roberts, M.R. 2005. Response of plant functional groups within plantations and naturally regenerated forests in southern New Brunswick , Canada. Can. J. For. Res. 35 : 1261–1267. Rémy, A., Odden, M., Richard, M., Stene, M.T., Le Galliard, J.-F., and Andreassen, H.P. 2013. Food distribution influences social organization and population growth in a small rodent. Behav. Ecol. 24 : 832–841. Draft Rolan, R., and Gier, H. 1967. Correlation of embryo and placental scar counts of Peromyscus maniculatus and Microtus ochrogaster . J. Mammal. 48 : 317–319. Ross-Davis, A.L., and Frego, K.A. 2002. Comparison of plantations and naturally regenerated clearcuts in the Acadian Forest: forest floor bryophyte community and habitat features. Can. J. Bot. 80 : 21–33. Schweiger, S., and Boutin, S. 1995. The effects of winter food addition on the population dynamics of Clethrionomys rutilus . Can. J. Zool. 73 : 419–426. Sheldon, C. 1938. Vermont jumping mice of the genus Napaeozapus . J. Mammal. 19 : 444–453. Smith, F., Beeson, K., and Price, W. 1956. Chemical composition of herbage browsed by deer in two wildlife management areas. J. Wildl. Manage. 20 : 359–367. Sokal, R., and Rohlf, J. 1995. Biometry. In 3rd edition. W. H. Freeman, New York, New York. Spinner, G.P., and Bishop, J.S. 1950. Chemical analysis of some wildlife foods in Connecticut. J. Wildl. Manage. 14 : 175–180. Tabacaru, C.A., Millar, J.S., Longstaffe, F.J., and Ansell, A.K. 2010. Seasonal breeding in relation to dietary animal protein in deer mice ( Peromyscus maniculatus ). Can. J. Zool. 88 : 520–526. Taitt, M. 1981. The effect of extra food on small rodent populations: I. Deermice ( Peromyscus maniculatus ). J. Anim. Ecol. 50 : 111–124. Taitt, M.J., and Krebs, C.J. 1981. The effects of extra food on small rodent populations: II. Voles (Microtus townsendii ). J. Anim. Ecol. 50 : 125–137. Turner, B., Perrin, M., and Iverson, S. 1975. Winter coexistence of voles in spruce forest: relevance of seasonal changes in aggression. Can. J. Zool. 53 : 1004–1011. United States Department of Agriculture. 2012. National Nutrient Database for Standard Reference.

https://mc06.manuscriptcentral.com/cjz-pubs Page 21 of 30 Canadian Journal of Zoology

Available from http://www.ars.usda.gov. Vickery, W. 1979. Food consumption and preferences in wild populations of Clethrionomys gapperi and Napaeozapus insignis . Can. J. Zool. 57 : 1536–1542. Vickery, W., Daoust, J.-L., Wartiti, A. El, and Peletier, J. 1994. The effect of energy and protein content on food choice by deer mice, Peromyscus maniculatus (Rodentia). Anim. Behav. 47 : 55–64. Wagg, J. 1963. Notes on food habits of small mammals of the white spruce forest. For. Chron. 39 : 436– 445. West, S.D. 1982. Dynamics of colonization and abundance in central Alaskan populations of the northern red-backed vole, Clethrionomys rutilus . J. Mammal. 63 : 128–143. Whitaker Jr., J.O. 1963. Food, habitat and parasites of the woodland jumping mouse in central New York. J. Mammal. 44 : 316–321. White, G.C., Anderson, D.R., Burnham, K.P., and Otis, D. 1982. Capture-recapture and removal methods for sampling closed populations. United States Government Printing Press, Los Alamos, New Mexico. White, T.C.R. 2011. The significance of unripe seeds and animal tissues in the protein nutrition of herbivores. Cambridge Philos. Soc. 86 : 217–224. Draft

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 22 of 30

5. Tables

Table 1: Estimated mean±S.E. density (6 replicates per trapping session) of deer mice (Peromyscus maniculatus ), southern red-backed voles (Myodes gapperi ), and woodland jumping mice (Napaeozapus insignis ) per treatment during spring and summer 2011 and 2012, with RM ANOVA results, Black Brook, north-west New Brunswick.

Mean Density/ha ±S.E. per Trapping Session Treatment treatment Treatment Session x Season Block Effect deer mouse Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 1.7 ± 0.8 0.8 ± 0.7 0.26 0.64 1.68 0.00 0.89 0.59 0.93 0.53 Summer 2011 2.7 ± 0.9 5.7 ± 1.3 Spring 2012 8.2 ± 1.7 9.3 ± 1.7 Summer 2012 14.0 ± 4.9 14.5 ± 4.3 Draft southern red-backed vole Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 2.7 ± 0.9 2.2 ± 0.9 0.14 0.72 10.26 0.00 1.36 0.72 3.41 0.11 Summer 2011 8.7 ± 2.3 9.3 ± 2.0 Spring 2012 13.5 ± 4.0 15.2 ± 3.6 Summer 2012 11.3 ± 3.6 11.5 ± 3.9 woodland jumping mouse Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 2.0 ± 0.8 3.2 ± 0.9 0.03 0.88 0.41 0.75 0.79 0.68 4.20 0.07 Summer 2011 3.8 ± 2.1 2.5 ± 1.3 Spring 2012 6.3 ± 2.5 5.7 ± 4.3 Summer 2012 2.3 ± 1.3 1.8 ± 1.2

https://mc06.manuscriptcentral.com/cjz-pubs Page 23 of 30 Canadian Journal of Zoology

Table 2: Estimated mean±S.E. percent (6 replicates per trapping session) of juveniles among total individuals ( n = range of individuals captured within trap session) deer mice(Peromyscus maniculatus ), southern red-backed voles(Myodes gapperi ), and woodland jumping mice (Napaeozapus insignis ) with RM ANOVA results, Black Brook, north-west New Brunswick.

Trapping Session Mean proportion of juveniles ±S.E. per treatment Treatment Treatment Session x Season Block Effect

deer mouse Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 0.0 ± 0.0 (n 0-4) 16.7 ± 16.7 (n 0-4) 1.32 0.32 16.27 <0.01 0.48 0.70 1.27 0.40 Summer 2011 18.1 ± 8.7 (n 0-6) 34.0 ± 16.0 (n 2-11) Spring 2012 28.6 ± 8.2 (n 3-13) 22.7 ± 11.2 (n 4-14) Summer 2012 32.2 ± 9.8 (n 0-28) 51.2 ± 11.5 (n 2-29)

southern red-backed vole Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 4.2 ± 5.6 (n 0-6) 2.8 ± 2.8 (n 0-5) 0.35 0.58 4.04 0.03 0.93 0.45 2.85 0.14 Summer 2011 15.3 ± 2.8 (n 4-18) 10.4Draft ± 8.2 (n 4-15) Spring 2012 5.0 ± 0.0 (n 2-27) 0.6 ± 0.6 (n 0-22) Summer 2012 11.1 ± 11.1 (n 0-24) 0.0 ± 0.0 (n 1-23)

woodland jumping mouse Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 5.6 ± 4.2 (n 0-5) 6.7 ± 6.7 (n 0-6) 0.07 0.81 0.79 0.52 0.25 0.86 1.94 0.24 Summer 2011 4.1 ± 5.1 (n 0-13) 12.1 ± 4.4 (n 0-8) Spring 2012 <0.1 ± 2.5 (n 2-18) 15.2 ± 4.4 (n 0-27) Summer 2012 11.1 ± 4.7 (n 0-8) 9.2 ± 4.6 (n 0-6)

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 24 of 30

Table 3: Estimated mean±S.E. percent (6 replicates per trapping session) of reproductively active females among total individuals ( n = range of individuals captured within trap session) deer mice(Peromyscus maniculatus ), southern red-backed voles(Myodes gapperi ), and woodland jumping mice (Napaeozapus insignis ) with RM ANOVA results, Black Brook, north-west New Brunswick.

Mean proportion of reproductively active females ±S.E. Trapping Session Treatment per treatment Treatment Session x Season Block Effect

deer mouse Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 8.3 ± 8.3 (n 0-4) 0.0 ± 0.0 (n 0-4) 4.00 0.10 2.45 0.10 2.08 0.15 17.80 0.01 Summer 2011 4.2 ± 4.2 (n 0-6) 9.1 ± 4.8 (n 2-11) Spring 2012 11.5 ± 4.3 (n 3-13) 6.5 ± 3.3 (n 4-14) Summer 2012 3.8 ± 2.3 (n 0-28) 0.9 ± 0.9 (n 2-29) F F F southern red-backed vole Control Draft Added Fruit F1,5 p 3,15 p 3,15 p 3,5 p Spring 2011 5.6 ± 5.6 (n 0-6) 12.2 ± 7.8 (n 0-5) 3.53 0.12 2.43 0.10 1.02 0.41 5.88 0.04 Summer 2011 11.1 ± 8.2 (n 4-18) 2.1 ± 2.1 (n 4-15) Spring 2012 14.9 ± 8.4 (n 2-27) 0.0 ± 0.0 (n 0-22) Summer 2012 16.7 ± 16.7 (n 0-24) 10.0 ± 10.0 (n 1-23)

woodland jumping mouse Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 0.0 ± 0.0 (n 0-5) 3.3 ± 3.3 (n 0-6) 3.46 0.12 0.87 0.48 1.20 0.34 9.48 0.01 Summer 2011 6.1 ± 4.6 (n 0-13) 11.2 ± 5.1 (n 0-8) Spring 2012 21.1 ± 15.9 (n 2-18) 10.4 ± 3.4 (n 0-27) Summer 2012 16.7 ± 7.5 (n 0-8) 14.4 ± 5.0 (n 0-6)

https://mc06.manuscriptcentral.com/cjz-pubs Page 25 of 30 Canadian Journal of Zoology

Table 4: Estimated mean ±S.E. adult weight (g) of deer mice(Peromyscus maniculatus ), southern red-backed voles(Myodes gapperi ), and woodland jumping mice (Napaeozapus insignis ) per treatment during spring and summer with RM ANOVA results, Black Brook, north- west New Brunswick.

Season Mean weight ±S.E. per treatment Treatment Treatment Session x Season Block Effect

deer mouse Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 18.9 ± 1.5 (n 5) 15.5 ± 2.3 (n 2) 0.22 0.64 1.79 0.16 0.80 0.50 0.54 0.75 Summer 2011 15.8 ± 1.4 (n 6) 19.4 ± 1.0 (n 13) Spring 2012 19.2 ± 0.8 (n 23) 19.5 ± 0.7 (n 24) Summer 2012 19.6 ± 1.0 (n 14) 17.2 ± 1.6 (n 14) F F F southern red-backed vole Control Draft Added Fruit F1,5 p 3,15 p 3,15 p 3,5 p Spring 2011 21.5 ± 1.7 (n 7) 21.7 ± 2.0 (n 5) 0.91 0.34 0.94 0.42 0.59 0.62 3.70 <0.01 Summer 2011 20.7 ± 1.1 (n 15) 20.3 ± 1.0 (n 19) Spring 2012 21.8 ± 0.7 (n 41) 23.0 ± 0.7 (n 42) Summer 2012 22.4 ± 0.7 (n 49) 23.3 ± 0.9 (n 27)

woodland jumping mouse Control Added Fruit F1,5 p F3,15 p F3,15 p F3,5 p Spring 2011 20.1 ± 1.5 (n 5) 20.8 ± 1.2 (n 7) 0.02 0.88 18.50 <0.01 1.08 0.36 0.74 0.59 Summer 2011 24.3 ± 1.4 (n 5) 25.7 ± 1.2 (n 7) Spring 2012 22.4 ± 0.8 (n 21) 20.6 ± 0.9 (n 21) Summer 2012 26.7 ± 0.9 (n 20) 28.8 ± 1.3 (n 17)

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 26 of 30

Table 5: Average and maximum ground coverage (%) of berry producing plants per treatment ( n = 120; 20 plots per treatment; 6 treatment blocks) during 2011 and 2012, Black Brook, north-west New Brunswick

Average Cover (% ±S.E.) Maximum Cover (%) Species Control Added Fruit Control Added Fruit Aralia nudicaulis L. 1.42 ± 0.59 2.23 ± 0.41 22 25 Cornus Canadensis L. 2.32 ± 1.43 2.67 ± 1.40 45 42 Ericacea 1 0.00 ± 0.00 0.03 ± 0.03 0 2 Fragaria virginiana Dcne. 0.03 ± 0.2 1.95 ± 1.07 1 32 Liliaceae 2 2.10 ± 0.37 1.92 ± 0.42 8 12 Ribes spp. 3 1.05 ± 0.65 6.68 ± 3.81 27 60 Rubus spp. 4 0.82 ± 0.51 1.08 ± 0.71 18 13 Sambucus racemosa L. 0.22 ± 0.15 0.90 ± 0.90 6 30 1: Gaultheria hispidula (L.) Muhl., Gaultheria procumbens L.,Draft Vaccinium angustifolium Ait., Vaccinium myrtilloides Michx. 2: Clintonia borealis (Ait.) Raf., Maianthemum canadense Desf., Maianthemum racemosum (L.), Streptopus lanceolatus (Ait.), Trillium erectum L., Trillium undulatum Willd. 3: Ribes glandulosum Grauer (glandular), Ribes lacustra (Pers.) Poir. 4: Rubus alleghaniensis Porter, Rubus idaeus L., Rubus pubescens Raf.

https://mc06.manuscriptcentral.com/cjz-pubs Page 27 of 30 Canadian Journal of Zoology

Table 6: Weight (g) of fruit (combined natural and added) (g) (n = 20 quadrats) and carbohydrate load per hectare per treatment block during 2011 and 2012, Black Brook, north-west New Brunswick.

Control Added-Fruit Plantation Fruit Weight (g) Fruit Carbs. (g) Fruit Weight (g) Fruit Carbs. (g) 2011 2012 2011 2012 2011 2012 2011 2012 A 0 0 0 0 5679.0 2173.3 3684.7 1019.1 B 4001.2 2142.9 440.1 235.7 2583.3 3520.0 2144.2 1647.2 C 0 214.4 0 23.6 2035.8 2678.3 1483.9 1254.6 D 0 18371.0 0 2020.8 3702.8 22083.0 2147.3 3397.5 E 0 0 0 0 2654.9 2846.7 1792.0 1333.1 F 1171.6 4471.9 128.9 491.9 10985.4 6072.2 3068.4 1747.9

Draft

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 28 of 30

6. List of Figures

Figure 1: Study plantations within J.D. Irving Limited's Black Brook Forest District, north-west New Brunswick, Canada.

Figure 2: Average density of deer mice(Peromyscus maniculatus ), red-backed voles(Myodes gapperi ), and woodland jumping mice (Napaeozapus insignis ) during spring and summer 2011 and 2012 between Control and Added Fruit treatments ( n = 6 each), within Black Brook, north-west New Brunswick, Canada.

Draft

https://mc06.manuscriptcentral.com/cjz-pubs Page 29 of 30 Canadian Journal of Zoology

D C A B

Kilometres 0 3 6 9 12

F Black Brook Draft New Brunswick

Figure 1: Study plantations within J.D. Irving Limited's Black Brook Forest District, north-west New Brunswick, Canada.

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 30 of 30

Deer mouse

15 Control Added Food

0 Red-backed vole

10 Density (ha) 5 Draft

0 Woodland jumping mouse

12 0 2011 2011 2 2012 ing er pr pring S S Summ Summer Figure 2: Average density of deer mice, red-backed voles, and woodland jumping mice during spring and summer 2011 and 2012 between Control and Added Fruit treatments (n = 6 each), within Black Brook, north-west New Brunswick, Canada.

https://mc06.manuscriptcentral.com/cjz-pubs