- Email: [email protected]
Re-caching of acorns by rodents: Cache management in eastern deciduous forests of North America
Acta Oecologica 92 (2018) 117–122
Contents lists available at ScienceDirect
Acta Oecologica journal homepage: www.elsevier.com/locate/actoec
Re-caching of acorns by rodents: Cache management in eastern deciduous forests of North America
T
Andrew W. Bartlowa,∗, Nathanael I. Lichtib, Rachel Curtisa, Robert K. Swihartb, Michael A. Steelea a b
Department of Biology, Wilkes University, Wilkes-Barre, PA, 18766, USA Department of Forestry and Natural Resources, Purdue University, West Lafayette, IN, 47907, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: Acorn dispersal Long-term caching Quercus Re-caching Sciurus Secondary caching Small mammals
Scatter-hoarding rodents such as tree squirrels selectively cache seeds for subsequent use in widely-spaced caches placed below the ground surface. This behavior has important implications for seed dispersal, seedling establishment, and tree regeneration. Hoarders manage these caches by recovering and eating some seeds, and moving and re-caching others. This process of re-caching, however, is poorly understood. Here, we use radiotelemetry to evaluate re-caching behavior for the management of acorn caches by rodents in eastern deciduous forests. We also test the hypothesis that as seeds are re-cached, the distance from the source increases. Radio transmitters were implanted in Northern red oak (Quercus rubra) acorns and presented to rodents in a natural setting over 3 seasons. We used radio-telemetry to track and document evidence of recovery and re-caching. We tracked a total of 102 acorns. Of the 39 radio-tagged acorns initially cached, 19 (49%) were cached on two or more occasions; one acorn was cached four times. The hypothesis that rodents move seeds to progressively greater distances from the source is not well-supported, suggesting that acorns are being moved within an individual's home range. Given the species of rodents in the study area, gray squirrels (Sciurus carolinensis) are the most likely to be responsible for the caching and re-caching events. Gray squirrels appear to engage in extensive re-caching during periods of long-term food storage, which has important implications for understanding how caching behavior influences acorn dispersal and oak regeneration.
1. Introduction Scatter-hoarding, the process by which mammals and birds store seeds for subsequent use in discrete, widely-spaced cache sites (Smith and Reichman, 1984; Vander Wall, 1990), is considered a primary mechanism of animal-mediated dispersal for many plant species. Scatter-hoarding allows hoarders to minimize food loss to potential pilferers by spacing caches instead of storing food in a single location. This hoarding strategy also results in seedling establishment when scatter-hoarding animals fail to recover their own caches (Cao et al., 2011; Steele et al., 2011). Dispersal of seeds by scatter-hoarding rodents is thus a critical step in the reproduction of many nut-bearing trees and a key process underlying ecosystem function in many terrestrial ecosystems (Steele and Smallwood, 2002). Dispersal of seeds away from parent trees decreases the incidence of density-dependent predation (Janzen, 1970), decreases competition with parents (Janzen, 1970), and often moves seeds to sites more suitable for establishment and recruitment (Wenny and Levey, 1998). Regardless of how well protected caches may be, cache loss is
∗
inevitable. Cache loss can occur because of pilfering by conspecifics and heterospecifics, seed perishability (i.e. germination schedules), insect infestation, or forgetting the cache location (Vander Wall and Smith, 1987). To compensate for these losses, scatter-hoarding animals may over-provision their stores and subsequently fail to recover a portion of cached seeds, thereby contributing to seedling establishment (Vander Wall, 1990). Numerous studies across a variety of temperate, subtropical, and tropical systems show that mammalian scatter-hoarders are highly selective with respect to a variety of seed characteristics (e.g., seed perishability and seed size; Lichti et al., 2014), maintain an advantage at recovering their own caches (Steele et al., 2011; but see Vander Wall and Jenkins, 2003), and may continually manage seed stores during the hoarding period, through a process of cache recovery and re-caching (Vander Wall and Joyner, 1998; Vander Wall, 2002; Jansen et al., 2006). Re-caching, whereby seeds are retrieved from one cache site and moved to a new cache site as a hoarder manages caches, is practiced by both mammals (Vander Wall and Joyner, 1998; Vander Wall, 2002; Jansen et al., 2006; Perea et al., 2011a) and birds (Hutchins and Lanner,
Corresponding author. E-mail address: [email protected] (A.W. Bartlow).
https://doi.org/10.1016/j.actao.2018.08.011 Received 13 March 2018; Received in revised form 24 July 2018; Accepted 26 August 2018 1146-609X/ © 2018 Elsevier Masson SAS. All rights reserved.
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al.
09, and 2010-11. Although also attempted in 2009, frequent equipment failure caused us to abandon the project during that field season. In 2007 and 2008, the study was conducted from mid-November until mid-March of the following year in a mature, deciduous forest dominated by Northern red oak, white oak, chestnut oak (Quercus prinus L.) and sugar maple (Acer saccharum Marsh.), located 3 km south of Mountaintop, Pennsylvania, USA (41°05′N, 75°55′W). The start time in each year of the study coincided with the period when most red oak acorns were being consumed or scatter-hoarded, thereby increasing the probability that our presented seeds would be found and dispersed. During the last field season, the study was conducted from midDecember 2010 until late-April 2011 in a similar stand a few kilometers west of the Mountaintop site (41°08′N, 75°59′W). These two locations are within the same forest and have the same forest structure and the same dispersal agents. A second location was added to increase independence of observations and generality of results. In all 3 seasons, we initiated the study in late fall when there was no evidence of eastern chipmunk (Tamias striatus L.) activity to ensure that caching was performed only by gray squirrels. Red oak acorns were collected beneath 5–10 individual trees near the study area in October of each year of the study and subsequently stored at 3 °C until fitted with radio transmitters. In preparation for experiments, we created a composite acorn sample from at least 5 trees, sorted the acorns to remove insect-damaged or rotted acorns, and pseudo-randomly selected acorns to fit with transmitters. Selected acorns were generally > 7 g because of the minimum size needed to fit the transmitters. Each radio transmitter (A2450, 2.2 g, Advanced Telemetry Systems, Inc.) was prepared by wrapping the antenna in a single layer of Parafilm (Bemis Company, Inc.), which was then wrapped in a tight coil around the body of the transmitter, without crossing the antenna over itself. The transmitter was then wrapped with an additional layer of Parafilm to protect the transmitter and prevent the antenna from unraveling. Each radio tag emitted a unique frequency, allowing acorns to be identified remotely. To fit an acorn with a transmitter, a hole (0.5 cm) was drilled through the basal end (the basal scar) of the acorn using a Dremel drill (Model 200-1/21) to remove much of the cotyledon. A wrapped transmitter was then carefully inserted into the acorn, and pieces of cotyledon were packed tightly into the remaining space to secure the transmitter. The opening in the basal end of the acorn was then filled with a mixture of shell debris and wood filler (Elmer's Wood Filler, E860, Red Oak). The surface was then carefully smoothed over, and after the wood putty dried, sanded lightly to ensure the acorn appeared sound. Rodents selectively consume nuts that appear unsound and preferentially cache those perceived to be sound and less perishable (Hadj-Chikh et al., 1996; Steele et al., 2002). The transmitter in each acorn was switched off by attaching a magnet to the acorn until the acorn was dispersed by rodents. In 2007 and 2010, a sample of the acorns was weighed ( ± 0.01 g) prior to placing the transmitter inside the acorn, and again after the transmitter was inserted and the wood putty had dried. Acorns were presented to rodents in semi-permeable enclosures (1 m × 1 m x 0.33 m) composed of 5 cm × 5 cm wooden frames covered on all sides with 0.5-cm mesh hardware cloth. A small hole (5 cm × 5 cm) at the center and base of each vertical side of the hardware cloth allowed access to only small mammals (i.e. gray squirrels, flying squirrels [Glaucomys volans L.], chipmunks, and mice [Peromyscus leucopus Rafinesque and P. maniculatus Wagner]). Radiotagged acorns were positioned on a 3-x-3 array of magnets that were recessed in a wooden block (5 cm × 5 cm x 2.5 cm) and positioned in the center of the enclosure. Each acorn was carefully positioned over a magnet on the wood block to inactivate the transmitter and save battery life until dispersed. A maximum of 5 radio-tagged acorns were presented at a time, at a single location, along with approximately 100 additional sound red oak and 100 white oak acorns. The additional acorns served to reduce the probability that rodents would associate the
1982; Waite and Reeve, 1992, 1995; Stotz and Balda, 1995; Burns and van Horik, 2007). Although re-caching has been observed in many instances, we know little about the relative frequency with which the behavior occurs and whether it is a part of ongoing cache management strategies in some seed dispersal systems. Re-caching may affect the probability of seedling establishment by influencing how far seeds are dispersed from the parent tree on subsequent re-caching events (Vander Wall and Joyner, 1998; Jansen et al., 2004). During redistribution of caches in some systems, the distance from the parent tree increases as seeds are re-cached by cache owners (Perea et al., 2011a; Zhang et al., 2014). The distance from the parent tree can influence germination and survival of seeds and seedlings by decreasing competition for resources with parents and siblings as well as reducing density-dependent predation (Janzen, 1970; Nathan and MullerLandau, 2000; Hirsch et al., 2012). Shortly after seed fall, seeds may be harvested quickly and cached near the parent tree in an attempt to stop other individuals from collecting seeds (Jenkins and Peters, 1992). Recovery of seeds from the initial caches and placement in new cache sites may result in seeds being taken farther from the parent tree to sites that are more favorable (Hirsch et al., 2012). These new cache locations may be favorable for seeds because of a higher probability of germination and for seed dispersers because they may reduce pilfering and cache loss. This pattern of increasing distance has been attributed to pilferage by conspecifics in other systems (e.g., Jansen et al., 2012). Acorns of Northern red oaks (Quercus rubra L.) are an essential food resource for many animals in eastern deciduous forests. These acorns germinate in the spring, unlike white oak (Quercus alba L.) acorns, and are an important resource throughout the winter. In autumn, red oak acorns are cached preferentially over the more perishable white oak acorns (Smallwood et al., 2001). Rodents (e.g. squirrels, chipmunks, and mice) and birds (e.g. jays) are important predators and dispersal agents of acorns and influence seedling establishment and oak forest regeneration when seeds are cached (Steele and Smallwood, 2002; Kellner and Swihart, 2017) or partially consumed (Perea et al., 2011b; Bartlow et al., 2018). Re-caching of acorns is documented for birds, such as Florida Scrub Jays (Aphelocoma coerulescens) (Kulahci and Bowman, 2011). However, it is not known whether rodents in eastern deciduous forests engage in re-caching behavior during periods of food storage. If re-caching is a common cache management strategy, it is unclear whether the distance from the parent tree increases as acorns are re-cached or whether acorns are redistributed within an individual's home range. Here, we sought to determine if re-caching occurs in the oak-dispersal system of eastern deciduous forests. Recent studies suggest that this is a highly dynamic dispersal process (Yi et al., 2012; Lichti et al., 2014; Sundaram et al., 2017). For example, dispersal patterns of acorns are dependent on oak species composition, seed size, and seed abundance (Moore et al., 2007; Bartlow et al., 2011; Lichti et al., 2014). The predominant dispersal agent, the eastern gray squirrel (Sciurus carolinensis, Gmelin), appears to rely on memory to relocate caches (Jacobs and Liman, 1991; Lavenex et al., 2000) and may potentially manage caches through the food-storing period. In addition to our primary objective of documenting re-caching, we tested the prediction that if acorns are re-cached, they will be moved farther from the source following cache recovery and re-caching. Behavior resulting in increased net dispersal distance may have implications for oak regeneration and design of future seed dispersal studies in these forests. We used radiotelemetry to determine if cached Northern red oak acorns were recovered and immediately consumed or recovered and re-cached. We chose radio-telemetry for this study because it allowed us to track the movement of individual acorns between cache sites without marking the cache sites with flags or tags. 2. Materials and methods This study was conducted in late fall and winter of 2007-08, 2008118
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al.
permutation tests (primary vs. secondary, primary vs. tertiary, secondary vs. tertiary, and primary vs. final cache distance) using 10,000 replications per comparison. Acorns from all years were included in these analyses. Bonferroni corrections were used to correct for multiple comparisons. Pairwise permutation tests were also conducted on the time (duration) that acorns remained in primary, secondary, and tertiary caches (residence time) using 10,000 replications per comparison. Again, all acorns during the 3 years were combined for these analyses. Residence time was defined as the number of days elapsed from when an acorn was cached until the acorn was discovered eaten or re-cached. Some acorns were brought back to the lab before they could be consumed or re-cached because of signal failure of the transmitter. The times for these acorns were determined using the last day they were checked before being pulled from the field. All statistical analyses were carried out using the statistical program R (Version 3.4.1; R Core Development Team, 2017).
red oak acorns with transmitters once they were opened. The white oak acorns further increased the probability that the red oak acorns would be cached instead of consumed, because white oak acorns are selectively eaten over red oak acorns, and were expected to satiate the rodents (Lichti et al., 2014). Acorns were presented in 6 boxes in 2007, 5 boxes in 2008, and 6 boxes in 2010. Multiple presentations of radiotagged acorns were completed at some boxes throughout the study. This was important if no acorns were cached during the first (or second) presentations. Having acorns cached at as many boxes as possible increased the number of individual squirrels likely involved in caching. No more than 15 acorns with transmitters were presented at a single box; the majority of boxes (65%) were used to present only 5 acorns. All boxes were separated by a minimum distance of 100 m, and most boxes were greater than 100 m apart. The boxes in 2010 were > 2 km from the boxes in 2007 and 2008. A total of 107 radio-tagged acorns were presented over the duration of the study. Acorn presentation locations were visited 2–3 times weekly to determine when tagged acorns were first dispersed; time constraints prevented more frequent checking. Upon movement of a tagged acorn, it was located with a radio receiver (R1000, Communications Specialists, Inc.) and directional antenna (F150-3FB 09642, AF Atronics). When an acorn was located, the distance and compass bearing from the source box were recorded along with the distance to, and a description of, the nearest landmark (e.g. tree, log, rock, etc.). These data were used to relocate the cached seeds without visibly marking cache sites. Acorns/transmitters were excavated to determine whether tagged acorns were eaten or cached. Acorns were not removed from the hole in which they were buried. Only enough soil was removed to get a clear view of the majority of the acorn. After verifying that a cached acorn was still intact, soil and leaf litter were replaced to show as little disturbance as possible. The acorn was then monitored twice weekly until it was recovered and eaten or re-cached. Upon locating an acorn in a new cache site, the distance and compass bearing from the box was recorded and the above procedures were repeated. Acorns that were found cached in larder hoards were retrieved and removed from the study, because transmitter battery life was short and larder-hoarded acorns were unlikely to be re-cached. We determined the percentage of re-caching events in each year separately and in all years of the study combined. We categorized dispersal events as primary, secondary, tertiary, or quaternary, where primary dispersal refers to the removal of seeds from their enclosures, secondary dispersal indicates removal from an initial (primary) cache, and so on. To test whether seed fates (scatter-hoarded, larder-hoarded, eaten, or not found) differed among these dispersal stages, we conducted a bootstrapped chi-squared test with 10,000 replications for the 2-way contingency table. Acorns for all years were combined for the chi-squared test because of sample size limitations and because we were interested in describing overall patterns of re-caching in this system. We also used bootstrap resampling to estimate the distribution of expected counts in each cell of the 4 × 4 table under the null hypothesis that seed fate probabilities did not change with dispersal stage. To do this, we calculated marginal number of seeds per dispersal stage, and then simulated 10,000 tables, distributing the seeds in each dispersal stage according to a multinomial distribution with the observed marginal fate probabilities. The true cell counts were then compared to these distributions to learn which individual cells drove any significant differences (i.e. p < 0.05) seen in the chi-squared test. We defined dispersal distance as the straight-line distance from the initial enclosure to the current cache, not the distance between successive caches. The final cache distance refers to the distance from the enclosure to the last cache location. For example, the final cache distance for an acorn cached two times would be the distance from the enclosure to the second cache location, and the distance for an acorn cached three times would be the distance from the enclosure to the third cache, etc. To test for differences in dispersal distances among primary, secondary, and tertiary caches, we conducted pairwise
3. Results In 2007, there was a small but significant difference between the masses of untagged and radio-tagged acorns (t = −2.0784, p = 0.043). Acorns housing transmitters were slightly heavier (mean ± SE; untagged acorns = 7.91 g ± 0.17; tagged acorns = 8.40 g ± 0.16), but still within the range of red oak acorn masses (Bartlow et al., 2018). In contrast, the initial and final masses of the acorns did not significantly differ in 2010 (t = 0.1981, p = 0.84) (mean ± SE; untagged acorns = 9.21 g ± 0.34; tagged acorns = 9.12 g ± 0.30). In the 3-year study, 102 of the 107 radio-tagged acorns placed in the enclosures were dispersed and tracked using radio-telemetry. The five acorns not dispersed were eaten inside enclosures. Thirty-nine acorns were found cached in scatter-hoards, and 51 were found eaten after dispersal (Fig. 1). Ten acorns (6 in 2007 and 4 in 2010) were found in larder-hoards after their initial removal and were retrieved and removed from the study. Two were not found after being removed from the enclosures. Instances of re-caching (secondary scatter-hoarding), in which a cached acorn was excavated and re-cached in a new location, were observed for 19 of the 39 primary caches (48.7%, Fig. 1). One acorn was cached four times over a 93-day period before being consumed. Secondary caching was documented in all 3 years of the study: 62.5% (10/16) in 2007, 14.3% (1/7) in 2008, and 50.0% (8/16) in
Fig. 1. Number of acorns dispersed and whether they were consumed or cached for each dispersal stage. Of the 102 acorns presented, 39 were cached after being removed from the boxes (primary dispersal). Of those 39 primary caches, 19 were re-cached (secondary dispersal). Four of the 19 re-cached acorns were cached a third time (tertiary dispersal). One acorn was cached four times (quaternary dispersal). 119
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al.
Fig. 2. Boxplot of the dispersal distances (m) from the source enclosure for each dispersal stage. The number of acorns for each stage are listed below. The distances of primary caches did not significantly differ from secondary cache distances. Tertiary cache distances were significantly farther than primary caches, but not secondary cache distances, after Bonferroni correction. The line in each box plot indicates the median distance from the source. The upper border of the box is the 75% quantile and the lower border of the box is the 25% quantile. (n.s. = p > 0.017 after Bonferroni correction, * = p < 0.017).
Fig. 3. Histogram of duration (days) that acorns were in caches for all dispersal stages combined. There was no significant difference among residence times of acorns in primary, secondary, and tertiary caches.
When analyzed by year, primary cache distances in 2007 (n = 16) were not significantly different from the dispersal distances of final caches of re-cached acorns (n = 10, p = 0.22). A similar pattern was seen in 2010 for primary cache distances (n = 16) and final cache distances of re-cached acorns (n = 8, p = 1.0). Only one acorn was recached in 2008. The durations for which acorns were cached and re-cached before they were consumed ranged from a 1 day to 93 days. There was no evidence of differences in residence times among the dispersal events according to the pairwise permutation tests (p > 0.15). Fig. 3 shows the distribution of residence times of cached acorns.
2010. Five acorns were moved from scatter-hoards to larder-hoards in 2007. No acorns were moved from scatter-hoards to larder-hoards in 2008 or 2010. A bootstrapped chi-square test on the dispersal stages (primary, secondary, tertiary, quaternary) and seed fates (scatter-hoarded, larderhoarded, eaten, or not found) was significant for the full 4 × 4 contingency table (χ2 = 24.74, p = 0.006). However, the bootstrapped distributions of each cell of the contingency table suggest that this significant difference was due to the fact that acorns were more likely to be ‘not found’ when they were dispersed from higher-order caches, probably due to tag failure. When this category was removed, the bootstrapped chi-square test was not significant (χ2 = 4.27, p = 0.68). The dispersal distances for the red oak acorns in all years of the study were similar to those reported with other methods in this system (Steele et al., 2001a; Moore et al., 2007). We observed 4 tertiary caches at 33, 35, 47, and 136 m from the enclosure. Using the shortest distance of 33 m as a benchmark, these distances were greater than at least 74.4% of primary dispersal events and 63% of the secondary dispersal events (Fig. 2). Assuming that tertiary distances were distributed similarly to primary or secondary distances, the probabilities of randomly drawing all 4 acorns in the top 25.6% or 36.8% of the distribution were 0.0043 and 0.0184, respectively. In addition, the permutation tests show that primary and secondary cache distances were similar (p = 0.29), while tertiary caches were placed at significantly greater distances than primary caches (p = 0.009). The permutation test suggested a difference between the tertiary and secondary cache distances as well, but this difference was not significant after Bonferroni correction (p = 0.03, versus a corrected alpha of 0.017). Primary dispersal distances of acorns that were cached (n = 39) did not differ significantly from final dispersal distances of re-cached acorns (n = 19, p = 0.29) using the permutation test. The final dispersal distance refers to the distance of the last cache site (i.e., third cache site for acorns cached three times). The final dispersal distances of only recached acorns did not differ significantly from their primary cache distances (n = 19 for both groups, p = 0.11). Finally, dispersal distances of acorns that were cached once but not re-cached (n = 20), had similar dispersal distances to the final distances of acorns that were cached two or more times (n = 19, p = 1.0).
4. Discussion Our primary objective was to determine whether re-caching occurs in eastern deciduous forests in North America. Using radio-telemetry, we followed acorns from their initial caches to secondary, tertiary, and quaternary cache sites through the process of cache recovery and recaching. These results provide evidence of re-caching of Northern red oak acorns in eastern deciduous forests, and add to a number of studies that suggest scatter-hoarding involves a variety of caching decisions and cache management strategies to ensure adequate food stores during periods of food shortage (Hadj-Chikh et al., 1996; Steele et al., 2001a; b; Steele and Smallwood, 2002; Sundaram et al., 2015). We argue that the re-caching behavior observed in this study was due primarily to the activity of eastern gray squirrels. The other rodents in this system known to scatter-hoard acorns, eastern chipmunks and white-footed mice (Peromyscus leucopus), are considered predominantly larder-hoarders (Steele and Smallwood, 2002). Indeed, the few instances of larder-hoarding in our study were determined to have been performed by chipmunks, as it was necessary to excavate underground chipmunk burrows to recover transmitters from these larders. These acorns were not regarded as being cached or re-cached. We presented acorns after chipmunk activity became less frequent. However, chipmunks often awaken during warming periods and engage in short periods of foraging even in winter (Frank, 2012). White-footed mice disperse acorns relatively short distances, often in shallow larders and along rocks and fallen logs (Steele and Smallwood, 2002). Several studies suggest that Peromyscus spp. cache fewer acorns compared with larger rodents, such as gray squirrels (Wróbel and Zwolak. 2017). Moreover, the larger acorns used in this study weighed 30–50% of the weight of an average adult white-footed mouse (average weight 120
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al.
final dispersal location may be the last of several caching events for the given acorn. Likewise, the location of a cache may be the first in a series of future recovery and re-caching events. Finally, we suggest that certain aspects of this study likely render our estimates of re-caching rates conservative. The manipulation of the acorn (i.e., internal implantation of a transmitter) may cause acorns to appear less sound, which may cause them to be cached less often than intact acorns. It is also possible that the movement of tagged acorns occurred more than once between observer visits and thus went undetected. This includes acorns that could not be followed until their final fate was determined due to transmitter signal failure. On the other hand, disturbance of the cache without removal of the seed may have increased the likelihood of re-caching since there would have been evidence of potential pilfering. Subsequent studies should determine exactly how many times individual acorns are re-cached. Regardless of whether re-caching was over- or under-estimated, our results show that re-caching occurs in eastern deciduous forests of North America. Recaching behavior may have important implications for understanding seed and nut dispersal in eastern deciduous forests. An important future goal is to better understand the role of both cache owners and pilferers in this system.
approximately 20 g), making it highly unlikely that mice would scatterhoard these acorns (Muñoz and Bonal, 2008). Southern flying squirrels (Glaucomys volans) are reported to occasionally scatter-hoard seeds (Gibbs et al., 2007), but they do so either in trees or in very shallow caches, often just covered by leaf litter (Mull, 1968). This species has never been observed at the sites used in our experiments (Steele, personal observation based on camera trap data). Nonetheless, given that several seed dispersers are found at our study sites, we cannot dismiss the possibility that a small portion of these acorns were cached by white-footed mice or southern flying squirrels. Studies suggest that patterns of re-caching are often the result of pilferage by conspecifics (Vander Wall and Jenkins, 2003; Vander Wall, 2002; Jansen et al., 2012; but see Jenkins and Peters, 1992). Steele et al. (2011) found that eastern gray squirrels have substantial recovery advantages over conspecifics, even in an unusually dense population (see Vander Wall et al., 2008 for similar studies on Tamias amoenus). However, we cannot dismiss the possibility that pilfering conspecifics may have played a role in the results observed here. Cache distances were similar for primary and secondary caches, but farther for tertiary caches. The hypothesis that seeds are moved to progressively greater distances from the source is not well-supported by our findings. If squirrels are the main dispersers, they likely have a recovery advantage over their own caches (Steele et al., 2011). Therefore, most re-caching events probably occur because cache owners are redistributing seeds within their home ranges. Gray squirrels are not territorial; home ranges vary in size and may occasionally overlap (Flyger, 1960; Doebel and McGinnes, 1974). Home ranges can be between one and a few hectares. Some secondary and tertiary caches may represent pilfering events. If pilferers are venturing into their neighbors' home ranges, pilfering caches, and then carrying the seeds back to the core areas of their own home ranges, this could explain a tendency for longer movements to occur in higher-order caches (Jansen et al., 2012). Future work is needed to test this hypothesis. Future studies should also focus on the microsites of the re-cached acorns and their final fates, including if germination rates differ from seeds cached only once. Previous studies on food hoarding by scatter-hoarders suggest that seeds may be revisited to reduce pilferage, to obtain feedback on the status of caches, or to recharge spatial memory (Jansen and Forget, 2001; Samson and Manser, 2016). Spatial memory is important in the cache recovery behavior of gray squirrels and other sciurids (Jacobs and Liman, 1991; Lavenex et al., 1998, 2000; Jacobs and Shiflett, 1999; Gibbs et al., 2007). However, the duration of spatial memory is poorly understood. It has been demonstrated that gray squirrels recover their caches after being removed from their home range for 20–30 days (Steele et al., 2011). However, long periods of cache storage, like those observed in this study (up to 3 months), may require re-visitation of caches to investigate seed quality and to determine if the seed is still suitable for storage. In addition, re-visitation of cache sites would also provide information on cache losses due to pilferage or seed rot and allow the cache owner to recharge memory of multiple cache locations at the same time. This study adds to a series of studies on eastern gray squirrels that suggest that scatter-hoarding by this species is a dynamic process in which seed stores are carefully managed and monitored to reduce cache pilferage and to maximize return rate (Steele et al. 2001a, b; 2011; Steele et al., 2006). Because acorns of red oak species (Section Lobatae) are more preferred for storage over acorns of white oak species (Section Quercus; Steele and Smallwood, 1994; Hadj-Chik et al., 1996; Steele et al., 2001a), these red oak acorns are more likely to be re-cached and managed throughout the winter. This may ultimately influence the final fate and location of dispersed acorns and should be considered important in the oak dispersal system. Knowing that squirrels may engage in this cache management strategy will aid interpretation of results of seed dispersal studies and provide more information about the oak dispersal system. Future studies should take into consideration that the
Author contributions AWB and RC collected and analyzed the data. NL, RKS, and MAS conceived the study. MAS and AWB wrote the manuscript. NL, RKS, and RC provided extensive edits. All authors have approved the final article. Declarations of interest None. Acknowledgments We thank Adam Curtis, Zachary Curtis, Christopher Bartlow, Patrick Lello, Megan Feusner, Peter Dombroski, Gina Robinson, Sarah-Jane Gerstman, and Xianfeng Yi for their contribution to field work. The authors recognize financial support from the US National Science Foundation (NSF-DEB 0642434, DEB-0642634), the Howard Hughes Medical Institute (AB, RC, MS), the Fenner Endowment of the Department of Biology of Wilkes University (AB, RC, MS) a Bullard Fellowship at Harvard Forest (MS), and the US NSF (DEB 1556707), which supported MS during the final preparation of the manuscript. Appendix A. Supplementary data Supplementary data related to this article can be found at https:// doi.org/10.1016/j.actao.2018.08.011. References Bartlow, A.W., Agosta, S.J., Curtis, R., Yi, X., Steele, M.A., 2018. Acorn size and tolerance to seed predators: the multiple roles of acorns as food for seed predators, fruit for dispersal and fuel for growth. Integr. Zool. 13, 251–266. Bartlow, A.W., Kachmar, M., Lichti, N., Swihart, R.K., Stratford, J.A., Steele, M.A., 2011. Does multiple seed loading in Blue Jays result in selective dispersal of smaller acorns? Integr. Zool. 6, 235–243. Burns, K.C., van Horik, J., 2007. Sexual differences in food re-caching by New Zealand robins Petroica australis. J. Avian Biol. 38, 394–398. Cao, L., Xiao, Z., Guo, C., Chen, J., 2011. Scatter-hoarding rodents as secondary seed dispersers of a frugivore-dispersed tree Scleropyrum wallichianum in a defaunated Xishuangbanna tropical forest, China. Integr. Zool. 6, 227–234. Doebel, J.H., McGinnes, B.S., 1974. Home range and activity of a gray squirrel population. J. Wildl. Manag. 38, 860–867. Flyger, V.F., 1960. Movements and home range of the gray squirrel Sciurus carolinensis, in two Maryland woodlots. Ecology 41, 365–369. Frank, C.L., 2012. The relationship between climate warning and hibernation in mammals. In: Storey, K., Tanino, K. (Eds.), Temperature Adaptation in a Changing Climate: Nature at Risk. CABI, New York, NY, pp. 120–130. Gibbs, S.E.B., Lea, S.E.G., Jacobs, L.F., 2007. Flexible use of spatial cues in the southern
121
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al. flying squirrel (Glaucomys volans). Anim. Cognit. 10, 203–209. Hadj-Chikh, L.Z., Steele, M.A., Smallwood, P.D., 1996. Caching decisions by grey squirrels: a test of the handling time and perishability hypotheses. Anim. Behav. 52, 941–948. Hirsch, B.T., Kays, R., Pereira, V.E., Jansen, P.A., 2012. Directed seed dispersal towards areas with low conspecific tree density by a scatter‐hoarding rodent. Ecol. Lett. 15, 1423–1429. Hutchins, H.E., Lanner, R.M., 1982. The central role of Clark's nutcracker in the dispersal and establishment of whitebark pine. Oecologia 55, 192–201. Jacobs, L.F., Liman, E.R., 1991. Grey squirrels remember the locations of buried nuts. Anim. Behav. 41, 103–110. Jacobs, L.F., Shiflett, M.W., 1999. Spatial orientation on a vertical maze in free-ranging fox squirrels (Sciurus Niger). J. Comp. Psychol. 113, 116–127. Jansen, P.A., Forget, P.M., 2001. Scatterhoarding rodents and tree regeneration. In: In: Bongers, F., Charles-Dominique, P., Forget, P.M., Théry, M. (Eds.), Nouragues. Monographiae Biologicae, vol 80 Springer, Dordrecht. Jansen, P.A., Bongers, F., Hemerik, L., 2004. Seed mass and mast seeding enhance dispersal by a neotropical scatter–hoarding rodent. Ecol. Monogr. 74, 569–589. Jansen, P.A., Bongers, F., Prins, H.H.T., 2006. Tropical rodents change rapidly germinating seeds into long-term food supplies. Oikos 113, 449–458. Jansen, P.A., Hirsch, B.T., Emsens, W.J., Zamora-Gutierrez, V., Wikelski, M., Kays, R., 2012. Thieving rodents as substitute dispersers of megafaunal seeds. Proc. Natl. Acad. Sci. Unit. States Am. 109, 12610–12615. Janzen, D.H., 1970. Herbivores and the number of tree species in tropical forests. Am. Nat. 104, 501–528. Jenkins, S.H., Peters, R.A., 1992. Spatial patterns of food storage by Merriam's kangaroo rats. Behav. Ecol. 3, 60–65. Kellner, K.F., Swihart, R.K., 2017. Simulation of oak early life history and interactions with disturbance via an individual-based model, SOEL. PLoS One 12 (6), e0179643. Kulahci, I.G., Bowman, R., 2011. Recaching decisions of Florida scrub‐jays are sensitive to ecological conditions. Ethology 117, 700–707. Lavenex, P., Shiflett, M.W., Lee, R.K., Jacobs, L.F., 1998. Spatial versus nonspatial relational learning in free-ranging fox squirrels (Sciurus Niger). J. Comp. Psychol. 112, 127. Lavenex, P., Steele, M.A., Jacobs, L.F., 2000. The seasonal pattern of cell proliferation and neuron number in the dentate gyrus of wild adult eastern grey squirrels. Eur. J. Neurosci. 12, 643–648. Lichti, N.I., Steele, M.A., Zhang, H., Swihart, R.K., 2014. Mast species composition alters seed fate in North American rodent‐dispersed hardwoods. Ecology 95, 1746–1758. Moore, J.E., McEuen, A.B., Swihart, R.K., Contreras, T.A., Steele, M.A., 2007. Determinants of seed removal distance by scatter‐hoarding rodents in deciduous forests. Ecology 88, 2529–2540. Mull, I., 1968. Behavioral and Physiological Influences on the Distribution of the Flying Squirrel, Glaucomys Volans. Museum of Zoology. University of Michigan. Miscellaneous publications, no. 134. Muñoz, A., Bonal, R., 2008. Are you strong enough to carry that seed? Seed size/body size ratios influence seed choices by rodents. Anim. Behav. 76, 709–715. Nathan, R., Muller-Landau, H.C., 2000. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Trends Ecol. Evol. 15, 278–285. Perea, R., San Miguel, A., Gil, L., 2011a. Acorn dispersal by rodents: the importance of redispersal and distance to shelter. Basic Appl. Ecol. 12, 432–439. Perea, R., San Miguel, A., Gil, L., 2011b. Leftovers in seed dispersal: ecological implications of partial seed consumption for oak regeneration. J. Ecol. 99, 194–201. R Development Core Team, 2017. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria3-900051-07-0. http://www.R-project.org. Samson, J., Manser, M.B., 2016. Caching in the presence of competitors: are Cape ground squirrels (Xerus inauris) sensitive to audience attentiveness? Anim. Cognit. 19, 31–38. Smallwood, P.D., Steele, M.A., Faeth, S.H., 2001. The ultimate basis of the caching
preferences of rodents, and the oak-dispersal syndrome: tannins, insects, and seed germination. Am. Zool. 41, 840–851. Smith, C.C., Reichman, O.J., 1984. The evolution of food caching by birds and mammals. Annu. Rev. Ecol. Systemat. 15, 329–351. Steele, M.A., Bugdal, M., Yuan, A., Bartlow, A., Buzalewski, J., Lichti, N., Swihart, R.K., 2011. Cache placement, pilfering, and a recovery advantage in a seed-dispersing rodent: could predation of scatter hoarders contribute to seedling establishment? Acta Oecol. 37, 554–560. Steele, M.A., Manierre, S., Genna, T., Contreras, T.A., Smallwood, P.D., Pereira, M.E., 2006. The innate basis of food-hoarding decisions in grey squirrels: evidence for behavioural adaptations to the oaks. Anim. Behav. 71, 155–160. Steele, M.A., Smallwood, P.D., 1994. What are squirrels hiding? Nat. Hist. 103, 40–45. Steele, M.A., Smallwood, P.D., 2002. Acorn dispersal by birds and mammals. In: McShea, W., Healy, W. (Eds.), Oak Forest Ecosystems. The Johns Hopkins University Press, Baltimore, pp. 182–195. Steele, M.A., Smallwood, P.D., Spunar, A., Nelsen, E., 2001a. The proximate basis of the oak dispersal syndrome: detection of seed dormancy by rodents. Am. Zool. 41, 852–864. Steele, M.A., Turner, G., Smallwood, P.D., Wolff, J.O., Radillo, J., 2001b. Cache management by small mammals: experimental evidence for the significance of acorn embryo excision. J. Mammal. 82, 35–42. Stotz, N.G., Balda, R.P., 1995. Cache and recovery behavior of wild pinyon jays in northern Arizona. SW. Nat. 40, 180–184. Sundaram, M., Willoughby, J.R., Lichti, N.I., Steele, M.A., Swihart, R.K., 2015. Segregating the effects of seed traits and common ancestry of hardwood trees on eastern gray squirrel foraging decisions. PLoS One 10, e0130942. https://doi.org/10. 1371/journal.pone.0130942. Sundaram, M., Lichti, N.I., Steele, M.A., Dalgleish, H.J., Swihart, R.K., 2017. Frequencydependent hoarding by Sciurus carolinensis occurs with seeds of similar perceived value. J. Mammal. 98, 124–134. Vander Wall, S.B., 1990. Food Hoarding in Animals. University of Chicago Press, Chicago. Vander Wall, S.B., 2002. Secondary dispersal of Jeffrey pine seeds by todent dcatterhoarders: the toles of pilfering, re-caching and a variable environment. In: Levey, D.J., Silva, W.R., Galetti, M. (Eds.), Seed Dispersal and Frugivory: Ecology, Evolution and Conservation. CAB International, New York, NY, pp. 193–208. Vander Wall, S.B., Jenkins, S.H., 2003. Reciprocal pilferage and the evolution of foodhoarding behavior. Behav. Ecol. 14, 656–667. Vander Wall, S.B., Joyner, J.W., 1998. Re-caching of Jeffrey pine (Pinus jeffreyi) seeds by yellow pine chipmunks (Tamias amoenus): potential effects on plant reproductive success. Can. J. Zool/Rev. Canad. Zool. 76, 154–162. Vander Wall, S.B., Downs, C.J., Enders, M.S., Waitman, B.A., 2008. Do yellow-pine chipmunks prefer to recover their own caches? West. N. Am. Nat. 68, 319–325. Vander Wall, S.B., Smith, K.G., 1987. Cache-protecting behavior of food hoarding animals. Pp. 611-644. In: Kamil, A.C., Krebs, J., Pulliam, H.R. (Eds.), Foraging Behavior. Plenum Press, New York, NY. Waite, T.A., Reeve, J.D., 1992. Caching behaviour in the gray jay and the source-departure decision for rate-maximizing scatterhoarders. Behaviour 120, 51–68. Waite, T.A., Reeve, J.D., 1995. Source-use decisions by hoarding gray jays: effects of local cache density and food value. J. Avian Biol. 26, 59–66. Wenny, D.G., Levey, D.J., 1998. Directed seed dispersal by bellbirds in a tropical cloud forest. Proc. Natl. Acad. Sci. U.S.A. 95, 6204–6207. Wróbel, A., Zwolak, R., 2017. Deciphering the effects of disperser assemblages and seed mass on patterns of seed dispersal in a rodent community. Integr. Zool. 12, 457–467. Yi, X., Yang, Y., Curtis, R., Bartlow, A.W., Agosta, S.J., Steele, M.A., 2012. Alternative strategies of seed predator escape by early-germinating oaks in Asia and North America. Ecol. Evol. 2, 487–492. Zhang, H., Steele, M.A., Zhang, Z., Wang, W., Wang, Y., 2014. Rapid sequestration and recaching by a scatter-hoarding rodent (Sciurotamias davidianus). J. Mammal. 95, 480–490.
122
Contents lists available at ScienceDirect
Acta Oecologica journal homepage: www.elsevier.com/locate/actoec
Re-caching of acorns by rodents: Cache management in eastern deciduous forests of North America
T
Andrew W. Bartlowa,∗, Nathanael I. Lichtib, Rachel Curtisa, Robert K. Swihartb, Michael A. Steelea a b
Department of Biology, Wilkes University, Wilkes-Barre, PA, 18766, USA Department of Forestry and Natural Resources, Purdue University, West Lafayette, IN, 47907, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: Acorn dispersal Long-term caching Quercus Re-caching Sciurus Secondary caching Small mammals
Scatter-hoarding rodents such as tree squirrels selectively cache seeds for subsequent use in widely-spaced caches placed below the ground surface. This behavior has important implications for seed dispersal, seedling establishment, and tree regeneration. Hoarders manage these caches by recovering and eating some seeds, and moving and re-caching others. This process of re-caching, however, is poorly understood. Here, we use radiotelemetry to evaluate re-caching behavior for the management of acorn caches by rodents in eastern deciduous forests. We also test the hypothesis that as seeds are re-cached, the distance from the source increases. Radio transmitters were implanted in Northern red oak (Quercus rubra) acorns and presented to rodents in a natural setting over 3 seasons. We used radio-telemetry to track and document evidence of recovery and re-caching. We tracked a total of 102 acorns. Of the 39 radio-tagged acorns initially cached, 19 (49%) were cached on two or more occasions; one acorn was cached four times. The hypothesis that rodents move seeds to progressively greater distances from the source is not well-supported, suggesting that acorns are being moved within an individual's home range. Given the species of rodents in the study area, gray squirrels (Sciurus carolinensis) are the most likely to be responsible for the caching and re-caching events. Gray squirrels appear to engage in extensive re-caching during periods of long-term food storage, which has important implications for understanding how caching behavior influences acorn dispersal and oak regeneration.
1. Introduction Scatter-hoarding, the process by which mammals and birds store seeds for subsequent use in discrete, widely-spaced cache sites (Smith and Reichman, 1984; Vander Wall, 1990), is considered a primary mechanism of animal-mediated dispersal for many plant species. Scatter-hoarding allows hoarders to minimize food loss to potential pilferers by spacing caches instead of storing food in a single location. This hoarding strategy also results in seedling establishment when scatter-hoarding animals fail to recover their own caches (Cao et al., 2011; Steele et al., 2011). Dispersal of seeds by scatter-hoarding rodents is thus a critical step in the reproduction of many nut-bearing trees and a key process underlying ecosystem function in many terrestrial ecosystems (Steele and Smallwood, 2002). Dispersal of seeds away from parent trees decreases the incidence of density-dependent predation (Janzen, 1970), decreases competition with parents (Janzen, 1970), and often moves seeds to sites more suitable for establishment and recruitment (Wenny and Levey, 1998). Regardless of how well protected caches may be, cache loss is
∗
inevitable. Cache loss can occur because of pilfering by conspecifics and heterospecifics, seed perishability (i.e. germination schedules), insect infestation, or forgetting the cache location (Vander Wall and Smith, 1987). To compensate for these losses, scatter-hoarding animals may over-provision their stores and subsequently fail to recover a portion of cached seeds, thereby contributing to seedling establishment (Vander Wall, 1990). Numerous studies across a variety of temperate, subtropical, and tropical systems show that mammalian scatter-hoarders are highly selective with respect to a variety of seed characteristics (e.g., seed perishability and seed size; Lichti et al., 2014), maintain an advantage at recovering their own caches (Steele et al., 2011; but see Vander Wall and Jenkins, 2003), and may continually manage seed stores during the hoarding period, through a process of cache recovery and re-caching (Vander Wall and Joyner, 1998; Vander Wall, 2002; Jansen et al., 2006). Re-caching, whereby seeds are retrieved from one cache site and moved to a new cache site as a hoarder manages caches, is practiced by both mammals (Vander Wall and Joyner, 1998; Vander Wall, 2002; Jansen et al., 2006; Perea et al., 2011a) and birds (Hutchins and Lanner,
Corresponding author. E-mail address: [email protected] (A.W. Bartlow).
https://doi.org/10.1016/j.actao.2018.08.011 Received 13 March 2018; Received in revised form 24 July 2018; Accepted 26 August 2018 1146-609X/ © 2018 Elsevier Masson SAS. All rights reserved.
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al.
09, and 2010-11. Although also attempted in 2009, frequent equipment failure caused us to abandon the project during that field season. In 2007 and 2008, the study was conducted from mid-November until mid-March of the following year in a mature, deciduous forest dominated by Northern red oak, white oak, chestnut oak (Quercus prinus L.) and sugar maple (Acer saccharum Marsh.), located 3 km south of Mountaintop, Pennsylvania, USA (41°05′N, 75°55′W). The start time in each year of the study coincided with the period when most red oak acorns were being consumed or scatter-hoarded, thereby increasing the probability that our presented seeds would be found and dispersed. During the last field season, the study was conducted from midDecember 2010 until late-April 2011 in a similar stand a few kilometers west of the Mountaintop site (41°08′N, 75°59′W). These two locations are within the same forest and have the same forest structure and the same dispersal agents. A second location was added to increase independence of observations and generality of results. In all 3 seasons, we initiated the study in late fall when there was no evidence of eastern chipmunk (Tamias striatus L.) activity to ensure that caching was performed only by gray squirrels. Red oak acorns were collected beneath 5–10 individual trees near the study area in October of each year of the study and subsequently stored at 3 °C until fitted with radio transmitters. In preparation for experiments, we created a composite acorn sample from at least 5 trees, sorted the acorns to remove insect-damaged or rotted acorns, and pseudo-randomly selected acorns to fit with transmitters. Selected acorns were generally > 7 g because of the minimum size needed to fit the transmitters. Each radio transmitter (A2450, 2.2 g, Advanced Telemetry Systems, Inc.) was prepared by wrapping the antenna in a single layer of Parafilm (Bemis Company, Inc.), which was then wrapped in a tight coil around the body of the transmitter, without crossing the antenna over itself. The transmitter was then wrapped with an additional layer of Parafilm to protect the transmitter and prevent the antenna from unraveling. Each radio tag emitted a unique frequency, allowing acorns to be identified remotely. To fit an acorn with a transmitter, a hole (0.5 cm) was drilled through the basal end (the basal scar) of the acorn using a Dremel drill (Model 200-1/21) to remove much of the cotyledon. A wrapped transmitter was then carefully inserted into the acorn, and pieces of cotyledon were packed tightly into the remaining space to secure the transmitter. The opening in the basal end of the acorn was then filled with a mixture of shell debris and wood filler (Elmer's Wood Filler, E860, Red Oak). The surface was then carefully smoothed over, and after the wood putty dried, sanded lightly to ensure the acorn appeared sound. Rodents selectively consume nuts that appear unsound and preferentially cache those perceived to be sound and less perishable (Hadj-Chikh et al., 1996; Steele et al., 2002). The transmitter in each acorn was switched off by attaching a magnet to the acorn until the acorn was dispersed by rodents. In 2007 and 2010, a sample of the acorns was weighed ( ± 0.01 g) prior to placing the transmitter inside the acorn, and again after the transmitter was inserted and the wood putty had dried. Acorns were presented to rodents in semi-permeable enclosures (1 m × 1 m x 0.33 m) composed of 5 cm × 5 cm wooden frames covered on all sides with 0.5-cm mesh hardware cloth. A small hole (5 cm × 5 cm) at the center and base of each vertical side of the hardware cloth allowed access to only small mammals (i.e. gray squirrels, flying squirrels [Glaucomys volans L.], chipmunks, and mice [Peromyscus leucopus Rafinesque and P. maniculatus Wagner]). Radiotagged acorns were positioned on a 3-x-3 array of magnets that were recessed in a wooden block (5 cm × 5 cm x 2.5 cm) and positioned in the center of the enclosure. Each acorn was carefully positioned over a magnet on the wood block to inactivate the transmitter and save battery life until dispersed. A maximum of 5 radio-tagged acorns were presented at a time, at a single location, along with approximately 100 additional sound red oak and 100 white oak acorns. The additional acorns served to reduce the probability that rodents would associate the
1982; Waite and Reeve, 1992, 1995; Stotz and Balda, 1995; Burns and van Horik, 2007). Although re-caching has been observed in many instances, we know little about the relative frequency with which the behavior occurs and whether it is a part of ongoing cache management strategies in some seed dispersal systems. Re-caching may affect the probability of seedling establishment by influencing how far seeds are dispersed from the parent tree on subsequent re-caching events (Vander Wall and Joyner, 1998; Jansen et al., 2004). During redistribution of caches in some systems, the distance from the parent tree increases as seeds are re-cached by cache owners (Perea et al., 2011a; Zhang et al., 2014). The distance from the parent tree can influence germination and survival of seeds and seedlings by decreasing competition for resources with parents and siblings as well as reducing density-dependent predation (Janzen, 1970; Nathan and MullerLandau, 2000; Hirsch et al., 2012). Shortly after seed fall, seeds may be harvested quickly and cached near the parent tree in an attempt to stop other individuals from collecting seeds (Jenkins and Peters, 1992). Recovery of seeds from the initial caches and placement in new cache sites may result in seeds being taken farther from the parent tree to sites that are more favorable (Hirsch et al., 2012). These new cache locations may be favorable for seeds because of a higher probability of germination and for seed dispersers because they may reduce pilfering and cache loss. This pattern of increasing distance has been attributed to pilferage by conspecifics in other systems (e.g., Jansen et al., 2012). Acorns of Northern red oaks (Quercus rubra L.) are an essential food resource for many animals in eastern deciduous forests. These acorns germinate in the spring, unlike white oak (Quercus alba L.) acorns, and are an important resource throughout the winter. In autumn, red oak acorns are cached preferentially over the more perishable white oak acorns (Smallwood et al., 2001). Rodents (e.g. squirrels, chipmunks, and mice) and birds (e.g. jays) are important predators and dispersal agents of acorns and influence seedling establishment and oak forest regeneration when seeds are cached (Steele and Smallwood, 2002; Kellner and Swihart, 2017) or partially consumed (Perea et al., 2011b; Bartlow et al., 2018). Re-caching of acorns is documented for birds, such as Florida Scrub Jays (Aphelocoma coerulescens) (Kulahci and Bowman, 2011). However, it is not known whether rodents in eastern deciduous forests engage in re-caching behavior during periods of food storage. If re-caching is a common cache management strategy, it is unclear whether the distance from the parent tree increases as acorns are re-cached or whether acorns are redistributed within an individual's home range. Here, we sought to determine if re-caching occurs in the oak-dispersal system of eastern deciduous forests. Recent studies suggest that this is a highly dynamic dispersal process (Yi et al., 2012; Lichti et al., 2014; Sundaram et al., 2017). For example, dispersal patterns of acorns are dependent on oak species composition, seed size, and seed abundance (Moore et al., 2007; Bartlow et al., 2011; Lichti et al., 2014). The predominant dispersal agent, the eastern gray squirrel (Sciurus carolinensis, Gmelin), appears to rely on memory to relocate caches (Jacobs and Liman, 1991; Lavenex et al., 2000) and may potentially manage caches through the food-storing period. In addition to our primary objective of documenting re-caching, we tested the prediction that if acorns are re-cached, they will be moved farther from the source following cache recovery and re-caching. Behavior resulting in increased net dispersal distance may have implications for oak regeneration and design of future seed dispersal studies in these forests. We used radiotelemetry to determine if cached Northern red oak acorns were recovered and immediately consumed or recovered and re-cached. We chose radio-telemetry for this study because it allowed us to track the movement of individual acorns between cache sites without marking the cache sites with flags or tags. 2. Materials and methods This study was conducted in late fall and winter of 2007-08, 2008118
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al.
permutation tests (primary vs. secondary, primary vs. tertiary, secondary vs. tertiary, and primary vs. final cache distance) using 10,000 replications per comparison. Acorns from all years were included in these analyses. Bonferroni corrections were used to correct for multiple comparisons. Pairwise permutation tests were also conducted on the time (duration) that acorns remained in primary, secondary, and tertiary caches (residence time) using 10,000 replications per comparison. Again, all acorns during the 3 years were combined for these analyses. Residence time was defined as the number of days elapsed from when an acorn was cached until the acorn was discovered eaten or re-cached. Some acorns were brought back to the lab before they could be consumed or re-cached because of signal failure of the transmitter. The times for these acorns were determined using the last day they were checked before being pulled from the field. All statistical analyses were carried out using the statistical program R (Version 3.4.1; R Core Development Team, 2017).
red oak acorns with transmitters once they were opened. The white oak acorns further increased the probability that the red oak acorns would be cached instead of consumed, because white oak acorns are selectively eaten over red oak acorns, and were expected to satiate the rodents (Lichti et al., 2014). Acorns were presented in 6 boxes in 2007, 5 boxes in 2008, and 6 boxes in 2010. Multiple presentations of radiotagged acorns were completed at some boxes throughout the study. This was important if no acorns were cached during the first (or second) presentations. Having acorns cached at as many boxes as possible increased the number of individual squirrels likely involved in caching. No more than 15 acorns with transmitters were presented at a single box; the majority of boxes (65%) were used to present only 5 acorns. All boxes were separated by a minimum distance of 100 m, and most boxes were greater than 100 m apart. The boxes in 2010 were > 2 km from the boxes in 2007 and 2008. A total of 107 radio-tagged acorns were presented over the duration of the study. Acorn presentation locations were visited 2–3 times weekly to determine when tagged acorns were first dispersed; time constraints prevented more frequent checking. Upon movement of a tagged acorn, it was located with a radio receiver (R1000, Communications Specialists, Inc.) and directional antenna (F150-3FB 09642, AF Atronics). When an acorn was located, the distance and compass bearing from the source box were recorded along with the distance to, and a description of, the nearest landmark (e.g. tree, log, rock, etc.). These data were used to relocate the cached seeds without visibly marking cache sites. Acorns/transmitters were excavated to determine whether tagged acorns were eaten or cached. Acorns were not removed from the hole in which they were buried. Only enough soil was removed to get a clear view of the majority of the acorn. After verifying that a cached acorn was still intact, soil and leaf litter were replaced to show as little disturbance as possible. The acorn was then monitored twice weekly until it was recovered and eaten or re-cached. Upon locating an acorn in a new cache site, the distance and compass bearing from the box was recorded and the above procedures were repeated. Acorns that were found cached in larder hoards were retrieved and removed from the study, because transmitter battery life was short and larder-hoarded acorns were unlikely to be re-cached. We determined the percentage of re-caching events in each year separately and in all years of the study combined. We categorized dispersal events as primary, secondary, tertiary, or quaternary, where primary dispersal refers to the removal of seeds from their enclosures, secondary dispersal indicates removal from an initial (primary) cache, and so on. To test whether seed fates (scatter-hoarded, larder-hoarded, eaten, or not found) differed among these dispersal stages, we conducted a bootstrapped chi-squared test with 10,000 replications for the 2-way contingency table. Acorns for all years were combined for the chi-squared test because of sample size limitations and because we were interested in describing overall patterns of re-caching in this system. We also used bootstrap resampling to estimate the distribution of expected counts in each cell of the 4 × 4 table under the null hypothesis that seed fate probabilities did not change with dispersal stage. To do this, we calculated marginal number of seeds per dispersal stage, and then simulated 10,000 tables, distributing the seeds in each dispersal stage according to a multinomial distribution with the observed marginal fate probabilities. The true cell counts were then compared to these distributions to learn which individual cells drove any significant differences (i.e. p < 0.05) seen in the chi-squared test. We defined dispersal distance as the straight-line distance from the initial enclosure to the current cache, not the distance between successive caches. The final cache distance refers to the distance from the enclosure to the last cache location. For example, the final cache distance for an acorn cached two times would be the distance from the enclosure to the second cache location, and the distance for an acorn cached three times would be the distance from the enclosure to the third cache, etc. To test for differences in dispersal distances among primary, secondary, and tertiary caches, we conducted pairwise
3. Results In 2007, there was a small but significant difference between the masses of untagged and radio-tagged acorns (t = −2.0784, p = 0.043). Acorns housing transmitters were slightly heavier (mean ± SE; untagged acorns = 7.91 g ± 0.17; tagged acorns = 8.40 g ± 0.16), but still within the range of red oak acorn masses (Bartlow et al., 2018). In contrast, the initial and final masses of the acorns did not significantly differ in 2010 (t = 0.1981, p = 0.84) (mean ± SE; untagged acorns = 9.21 g ± 0.34; tagged acorns = 9.12 g ± 0.30). In the 3-year study, 102 of the 107 radio-tagged acorns placed in the enclosures were dispersed and tracked using radio-telemetry. The five acorns not dispersed were eaten inside enclosures. Thirty-nine acorns were found cached in scatter-hoards, and 51 were found eaten after dispersal (Fig. 1). Ten acorns (6 in 2007 and 4 in 2010) were found in larder-hoards after their initial removal and were retrieved and removed from the study. Two were not found after being removed from the enclosures. Instances of re-caching (secondary scatter-hoarding), in which a cached acorn was excavated and re-cached in a new location, were observed for 19 of the 39 primary caches (48.7%, Fig. 1). One acorn was cached four times over a 93-day period before being consumed. Secondary caching was documented in all 3 years of the study: 62.5% (10/16) in 2007, 14.3% (1/7) in 2008, and 50.0% (8/16) in
Fig. 1. Number of acorns dispersed and whether they were consumed or cached for each dispersal stage. Of the 102 acorns presented, 39 were cached after being removed from the boxes (primary dispersal). Of those 39 primary caches, 19 were re-cached (secondary dispersal). Four of the 19 re-cached acorns were cached a third time (tertiary dispersal). One acorn was cached four times (quaternary dispersal). 119
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al.
Fig. 2. Boxplot of the dispersal distances (m) from the source enclosure for each dispersal stage. The number of acorns for each stage are listed below. The distances of primary caches did not significantly differ from secondary cache distances. Tertiary cache distances were significantly farther than primary caches, but not secondary cache distances, after Bonferroni correction. The line in each box plot indicates the median distance from the source. The upper border of the box is the 75% quantile and the lower border of the box is the 25% quantile. (n.s. = p > 0.017 after Bonferroni correction, * = p < 0.017).
Fig. 3. Histogram of duration (days) that acorns were in caches for all dispersal stages combined. There was no significant difference among residence times of acorns in primary, secondary, and tertiary caches.
When analyzed by year, primary cache distances in 2007 (n = 16) were not significantly different from the dispersal distances of final caches of re-cached acorns (n = 10, p = 0.22). A similar pattern was seen in 2010 for primary cache distances (n = 16) and final cache distances of re-cached acorns (n = 8, p = 1.0). Only one acorn was recached in 2008. The durations for which acorns were cached and re-cached before they were consumed ranged from a 1 day to 93 days. There was no evidence of differences in residence times among the dispersal events according to the pairwise permutation tests (p > 0.15). Fig. 3 shows the distribution of residence times of cached acorns.
2010. Five acorns were moved from scatter-hoards to larder-hoards in 2007. No acorns were moved from scatter-hoards to larder-hoards in 2008 or 2010. A bootstrapped chi-square test on the dispersal stages (primary, secondary, tertiary, quaternary) and seed fates (scatter-hoarded, larderhoarded, eaten, or not found) was significant for the full 4 × 4 contingency table (χ2 = 24.74, p = 0.006). However, the bootstrapped distributions of each cell of the contingency table suggest that this significant difference was due to the fact that acorns were more likely to be ‘not found’ when they were dispersed from higher-order caches, probably due to tag failure. When this category was removed, the bootstrapped chi-square test was not significant (χ2 = 4.27, p = 0.68). The dispersal distances for the red oak acorns in all years of the study were similar to those reported with other methods in this system (Steele et al., 2001a; Moore et al., 2007). We observed 4 tertiary caches at 33, 35, 47, and 136 m from the enclosure. Using the shortest distance of 33 m as a benchmark, these distances were greater than at least 74.4% of primary dispersal events and 63% of the secondary dispersal events (Fig. 2). Assuming that tertiary distances were distributed similarly to primary or secondary distances, the probabilities of randomly drawing all 4 acorns in the top 25.6% or 36.8% of the distribution were 0.0043 and 0.0184, respectively. In addition, the permutation tests show that primary and secondary cache distances were similar (p = 0.29), while tertiary caches were placed at significantly greater distances than primary caches (p = 0.009). The permutation test suggested a difference between the tertiary and secondary cache distances as well, but this difference was not significant after Bonferroni correction (p = 0.03, versus a corrected alpha of 0.017). Primary dispersal distances of acorns that were cached (n = 39) did not differ significantly from final dispersal distances of re-cached acorns (n = 19, p = 0.29) using the permutation test. The final dispersal distance refers to the distance of the last cache site (i.e., third cache site for acorns cached three times). The final dispersal distances of only recached acorns did not differ significantly from their primary cache distances (n = 19 for both groups, p = 0.11). Finally, dispersal distances of acorns that were cached once but not re-cached (n = 20), had similar dispersal distances to the final distances of acorns that were cached two or more times (n = 19, p = 1.0).
4. Discussion Our primary objective was to determine whether re-caching occurs in eastern deciduous forests in North America. Using radio-telemetry, we followed acorns from their initial caches to secondary, tertiary, and quaternary cache sites through the process of cache recovery and recaching. These results provide evidence of re-caching of Northern red oak acorns in eastern deciduous forests, and add to a number of studies that suggest scatter-hoarding involves a variety of caching decisions and cache management strategies to ensure adequate food stores during periods of food shortage (Hadj-Chikh et al., 1996; Steele et al., 2001a; b; Steele and Smallwood, 2002; Sundaram et al., 2015). We argue that the re-caching behavior observed in this study was due primarily to the activity of eastern gray squirrels. The other rodents in this system known to scatter-hoard acorns, eastern chipmunks and white-footed mice (Peromyscus leucopus), are considered predominantly larder-hoarders (Steele and Smallwood, 2002). Indeed, the few instances of larder-hoarding in our study were determined to have been performed by chipmunks, as it was necessary to excavate underground chipmunk burrows to recover transmitters from these larders. These acorns were not regarded as being cached or re-cached. We presented acorns after chipmunk activity became less frequent. However, chipmunks often awaken during warming periods and engage in short periods of foraging even in winter (Frank, 2012). White-footed mice disperse acorns relatively short distances, often in shallow larders and along rocks and fallen logs (Steele and Smallwood, 2002). Several studies suggest that Peromyscus spp. cache fewer acorns compared with larger rodents, such as gray squirrels (Wróbel and Zwolak. 2017). Moreover, the larger acorns used in this study weighed 30–50% of the weight of an average adult white-footed mouse (average weight 120
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al.
final dispersal location may be the last of several caching events for the given acorn. Likewise, the location of a cache may be the first in a series of future recovery and re-caching events. Finally, we suggest that certain aspects of this study likely render our estimates of re-caching rates conservative. The manipulation of the acorn (i.e., internal implantation of a transmitter) may cause acorns to appear less sound, which may cause them to be cached less often than intact acorns. It is also possible that the movement of tagged acorns occurred more than once between observer visits and thus went undetected. This includes acorns that could not be followed until their final fate was determined due to transmitter signal failure. On the other hand, disturbance of the cache without removal of the seed may have increased the likelihood of re-caching since there would have been evidence of potential pilfering. Subsequent studies should determine exactly how many times individual acorns are re-cached. Regardless of whether re-caching was over- or under-estimated, our results show that re-caching occurs in eastern deciduous forests of North America. Recaching behavior may have important implications for understanding seed and nut dispersal in eastern deciduous forests. An important future goal is to better understand the role of both cache owners and pilferers in this system.
approximately 20 g), making it highly unlikely that mice would scatterhoard these acorns (Muñoz and Bonal, 2008). Southern flying squirrels (Glaucomys volans) are reported to occasionally scatter-hoard seeds (Gibbs et al., 2007), but they do so either in trees or in very shallow caches, often just covered by leaf litter (Mull, 1968). This species has never been observed at the sites used in our experiments (Steele, personal observation based on camera trap data). Nonetheless, given that several seed dispersers are found at our study sites, we cannot dismiss the possibility that a small portion of these acorns were cached by white-footed mice or southern flying squirrels. Studies suggest that patterns of re-caching are often the result of pilferage by conspecifics (Vander Wall and Jenkins, 2003; Vander Wall, 2002; Jansen et al., 2012; but see Jenkins and Peters, 1992). Steele et al. (2011) found that eastern gray squirrels have substantial recovery advantages over conspecifics, even in an unusually dense population (see Vander Wall et al., 2008 for similar studies on Tamias amoenus). However, we cannot dismiss the possibility that pilfering conspecifics may have played a role in the results observed here. Cache distances were similar for primary and secondary caches, but farther for tertiary caches. The hypothesis that seeds are moved to progressively greater distances from the source is not well-supported by our findings. If squirrels are the main dispersers, they likely have a recovery advantage over their own caches (Steele et al., 2011). Therefore, most re-caching events probably occur because cache owners are redistributing seeds within their home ranges. Gray squirrels are not territorial; home ranges vary in size and may occasionally overlap (Flyger, 1960; Doebel and McGinnes, 1974). Home ranges can be between one and a few hectares. Some secondary and tertiary caches may represent pilfering events. If pilferers are venturing into their neighbors' home ranges, pilfering caches, and then carrying the seeds back to the core areas of their own home ranges, this could explain a tendency for longer movements to occur in higher-order caches (Jansen et al., 2012). Future work is needed to test this hypothesis. Future studies should also focus on the microsites of the re-cached acorns and their final fates, including if germination rates differ from seeds cached only once. Previous studies on food hoarding by scatter-hoarders suggest that seeds may be revisited to reduce pilferage, to obtain feedback on the status of caches, or to recharge spatial memory (Jansen and Forget, 2001; Samson and Manser, 2016). Spatial memory is important in the cache recovery behavior of gray squirrels and other sciurids (Jacobs and Liman, 1991; Lavenex et al., 1998, 2000; Jacobs and Shiflett, 1999; Gibbs et al., 2007). However, the duration of spatial memory is poorly understood. It has been demonstrated that gray squirrels recover their caches after being removed from their home range for 20–30 days (Steele et al., 2011). However, long periods of cache storage, like those observed in this study (up to 3 months), may require re-visitation of caches to investigate seed quality and to determine if the seed is still suitable for storage. In addition, re-visitation of cache sites would also provide information on cache losses due to pilferage or seed rot and allow the cache owner to recharge memory of multiple cache locations at the same time. This study adds to a series of studies on eastern gray squirrels that suggest that scatter-hoarding by this species is a dynamic process in which seed stores are carefully managed and monitored to reduce cache pilferage and to maximize return rate (Steele et al. 2001a, b; 2011; Steele et al., 2006). Because acorns of red oak species (Section Lobatae) are more preferred for storage over acorns of white oak species (Section Quercus; Steele and Smallwood, 1994; Hadj-Chik et al., 1996; Steele et al., 2001a), these red oak acorns are more likely to be re-cached and managed throughout the winter. This may ultimately influence the final fate and location of dispersed acorns and should be considered important in the oak dispersal system. Knowing that squirrels may engage in this cache management strategy will aid interpretation of results of seed dispersal studies and provide more information about the oak dispersal system. Future studies should take into consideration that the
Author contributions AWB and RC collected and analyzed the data. NL, RKS, and MAS conceived the study. MAS and AWB wrote the manuscript. NL, RKS, and RC provided extensive edits. All authors have approved the final article. Declarations of interest None. Acknowledgments We thank Adam Curtis, Zachary Curtis, Christopher Bartlow, Patrick Lello, Megan Feusner, Peter Dombroski, Gina Robinson, Sarah-Jane Gerstman, and Xianfeng Yi for their contribution to field work. The authors recognize financial support from the US National Science Foundation (NSF-DEB 0642434, DEB-0642634), the Howard Hughes Medical Institute (AB, RC, MS), the Fenner Endowment of the Department of Biology of Wilkes University (AB, RC, MS) a Bullard Fellowship at Harvard Forest (MS), and the US NSF (DEB 1556707), which supported MS during the final preparation of the manuscript. Appendix A. Supplementary data Supplementary data related to this article can be found at https:// doi.org/10.1016/j.actao.2018.08.011. References Bartlow, A.W., Agosta, S.J., Curtis, R., Yi, X., Steele, M.A., 2018. Acorn size and tolerance to seed predators: the multiple roles of acorns as food for seed predators, fruit for dispersal and fuel for growth. Integr. Zool. 13, 251–266. Bartlow, A.W., Kachmar, M., Lichti, N., Swihart, R.K., Stratford, J.A., Steele, M.A., 2011. Does multiple seed loading in Blue Jays result in selective dispersal of smaller acorns? Integr. Zool. 6, 235–243. Burns, K.C., van Horik, J., 2007. Sexual differences in food re-caching by New Zealand robins Petroica australis. J. Avian Biol. 38, 394–398. Cao, L., Xiao, Z., Guo, C., Chen, J., 2011. Scatter-hoarding rodents as secondary seed dispersers of a frugivore-dispersed tree Scleropyrum wallichianum in a defaunated Xishuangbanna tropical forest, China. Integr. Zool. 6, 227–234. Doebel, J.H., McGinnes, B.S., 1974. Home range and activity of a gray squirrel population. J. Wildl. Manag. 38, 860–867. Flyger, V.F., 1960. Movements and home range of the gray squirrel Sciurus carolinensis, in two Maryland woodlots. Ecology 41, 365–369. Frank, C.L., 2012. The relationship between climate warning and hibernation in mammals. In: Storey, K., Tanino, K. (Eds.), Temperature Adaptation in a Changing Climate: Nature at Risk. CABI, New York, NY, pp. 120–130. Gibbs, S.E.B., Lea, S.E.G., Jacobs, L.F., 2007. Flexible use of spatial cues in the southern
121
Acta Oecologica 92 (2018) 117–122
A.W. Bartlow et al. flying squirrel (Glaucomys volans). Anim. Cognit. 10, 203–209. Hadj-Chikh, L.Z., Steele, M.A., Smallwood, P.D., 1996. Caching decisions by grey squirrels: a test of the handling time and perishability hypotheses. Anim. Behav. 52, 941–948. Hirsch, B.T., Kays, R., Pereira, V.E., Jansen, P.A., 2012. Directed seed dispersal towards areas with low conspecific tree density by a scatter‐hoarding rodent. Ecol. Lett. 15, 1423–1429. Hutchins, H.E., Lanner, R.M., 1982. The central role of Clark's nutcracker in the dispersal and establishment of whitebark pine. Oecologia 55, 192–201. Jacobs, L.F., Liman, E.R., 1991. Grey squirrels remember the locations of buried nuts. Anim. Behav. 41, 103–110. Jacobs, L.F., Shiflett, M.W., 1999. Spatial orientation on a vertical maze in free-ranging fox squirrels (Sciurus Niger). J. Comp. Psychol. 113, 116–127. Jansen, P.A., Forget, P.M., 2001. Scatterhoarding rodents and tree regeneration. In: In: Bongers, F., Charles-Dominique, P., Forget, P.M., Théry, M. (Eds.), Nouragues. Monographiae Biologicae, vol 80 Springer, Dordrecht. Jansen, P.A., Bongers, F., Hemerik, L., 2004. Seed mass and mast seeding enhance dispersal by a neotropical scatter–hoarding rodent. Ecol. Monogr. 74, 569–589. Jansen, P.A., Bongers, F., Prins, H.H.T., 2006. Tropical rodents change rapidly germinating seeds into long-term food supplies. Oikos 113, 449–458. Jansen, P.A., Hirsch, B.T., Emsens, W.J., Zamora-Gutierrez, V., Wikelski, M., Kays, R., 2012. Thieving rodents as substitute dispersers of megafaunal seeds. Proc. Natl. Acad. Sci. Unit. States Am. 109, 12610–12615. Janzen, D.H., 1970. Herbivores and the number of tree species in tropical forests. Am. Nat. 104, 501–528. Jenkins, S.H., Peters, R.A., 1992. Spatial patterns of food storage by Merriam's kangaroo rats. Behav. Ecol. 3, 60–65. Kellner, K.F., Swihart, R.K., 2017. Simulation of oak early life history and interactions with disturbance via an individual-based model, SOEL. PLoS One 12 (6), e0179643. Kulahci, I.G., Bowman, R., 2011. Recaching decisions of Florida scrub‐jays are sensitive to ecological conditions. Ethology 117, 700–707. Lavenex, P., Shiflett, M.W., Lee, R.K., Jacobs, L.F., 1998. Spatial versus nonspatial relational learning in free-ranging fox squirrels (Sciurus Niger). J. Comp. Psychol. 112, 127. Lavenex, P., Steele, M.A., Jacobs, L.F., 2000. The seasonal pattern of cell proliferation and neuron number in the dentate gyrus of wild adult eastern grey squirrels. Eur. J. Neurosci. 12, 643–648. Lichti, N.I., Steele, M.A., Zhang, H., Swihart, R.K., 2014. Mast species composition alters seed fate in North American rodent‐dispersed hardwoods. Ecology 95, 1746–1758. Moore, J.E., McEuen, A.B., Swihart, R.K., Contreras, T.A., Steele, M.A., 2007. Determinants of seed removal distance by scatter‐hoarding rodents in deciduous forests. Ecology 88, 2529–2540. Mull, I., 1968. Behavioral and Physiological Influences on the Distribution of the Flying Squirrel, Glaucomys Volans. Museum of Zoology. University of Michigan. Miscellaneous publications, no. 134. Muñoz, A., Bonal, R., 2008. Are you strong enough to carry that seed? Seed size/body size ratios influence seed choices by rodents. Anim. Behav. 76, 709–715. Nathan, R., Muller-Landau, H.C., 2000. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Trends Ecol. Evol. 15, 278–285. Perea, R., San Miguel, A., Gil, L., 2011a. Acorn dispersal by rodents: the importance of redispersal and distance to shelter. Basic Appl. Ecol. 12, 432–439. Perea, R., San Miguel, A., Gil, L., 2011b. Leftovers in seed dispersal: ecological implications of partial seed consumption for oak regeneration. J. Ecol. 99, 194–201. R Development Core Team, 2017. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria3-900051-07-0. http://www.R-project.org. Samson, J., Manser, M.B., 2016. Caching in the presence of competitors: are Cape ground squirrels (Xerus inauris) sensitive to audience attentiveness? Anim. Cognit. 19, 31–38. Smallwood, P.D., Steele, M.A., Faeth, S.H., 2001. The ultimate basis of the caching
preferences of rodents, and the oak-dispersal syndrome: tannins, insects, and seed germination. Am. Zool. 41, 840–851. Smith, C.C., Reichman, O.J., 1984. The evolution of food caching by birds and mammals. Annu. Rev. Ecol. Systemat. 15, 329–351. Steele, M.A., Bugdal, M., Yuan, A., Bartlow, A., Buzalewski, J., Lichti, N., Swihart, R.K., 2011. Cache placement, pilfering, and a recovery advantage in a seed-dispersing rodent: could predation of scatter hoarders contribute to seedling establishment? Acta Oecol. 37, 554–560. Steele, M.A., Manierre, S., Genna, T., Contreras, T.A., Smallwood, P.D., Pereira, M.E., 2006. The innate basis of food-hoarding decisions in grey squirrels: evidence for behavioural adaptations to the oaks. Anim. Behav. 71, 155–160. Steele, M.A., Smallwood, P.D., 1994. What are squirrels hiding? Nat. Hist. 103, 40–45. Steele, M.A., Smallwood, P.D., 2002. Acorn dispersal by birds and mammals. In: McShea, W., Healy, W. (Eds.), Oak Forest Ecosystems. The Johns Hopkins University Press, Baltimore, pp. 182–195. Steele, M.A., Smallwood, P.D., Spunar, A., Nelsen, E., 2001a. The proximate basis of the oak dispersal syndrome: detection of seed dormancy by rodents. Am. Zool. 41, 852–864. Steele, M.A., Turner, G., Smallwood, P.D., Wolff, J.O., Radillo, J., 2001b. Cache management by small mammals: experimental evidence for the significance of acorn embryo excision. J. Mammal. 82, 35–42. Stotz, N.G., Balda, R.P., 1995. Cache and recovery behavior of wild pinyon jays in northern Arizona. SW. Nat. 40, 180–184. Sundaram, M., Willoughby, J.R., Lichti, N.I., Steele, M.A., Swihart, R.K., 2015. Segregating the effects of seed traits and common ancestry of hardwood trees on eastern gray squirrel foraging decisions. PLoS One 10, e0130942. https://doi.org/10. 1371/journal.pone.0130942. Sundaram, M., Lichti, N.I., Steele, M.A., Dalgleish, H.J., Swihart, R.K., 2017. Frequencydependent hoarding by Sciurus carolinensis occurs with seeds of similar perceived value. J. Mammal. 98, 124–134. Vander Wall, S.B., 1990. Food Hoarding in Animals. University of Chicago Press, Chicago. Vander Wall, S.B., 2002. Secondary dispersal of Jeffrey pine seeds by todent dcatterhoarders: the toles of pilfering, re-caching and a variable environment. In: Levey, D.J., Silva, W.R., Galetti, M. (Eds.), Seed Dispersal and Frugivory: Ecology, Evolution and Conservation. CAB International, New York, NY, pp. 193–208. Vander Wall, S.B., Jenkins, S.H., 2003. Reciprocal pilferage and the evolution of foodhoarding behavior. Behav. Ecol. 14, 656–667. Vander Wall, S.B., Joyner, J.W., 1998. Re-caching of Jeffrey pine (Pinus jeffreyi) seeds by yellow pine chipmunks (Tamias amoenus): potential effects on plant reproductive success. Can. J. Zool/Rev. Canad. Zool. 76, 154–162. Vander Wall, S.B., Downs, C.J., Enders, M.S., Waitman, B.A., 2008. Do yellow-pine chipmunks prefer to recover their own caches? West. N. Am. Nat. 68, 319–325. Vander Wall, S.B., Smith, K.G., 1987. Cache-protecting behavior of food hoarding animals. Pp. 611-644. In: Kamil, A.C., Krebs, J., Pulliam, H.R. (Eds.), Foraging Behavior. Plenum Press, New York, NY. Waite, T.A., Reeve, J.D., 1992. Caching behaviour in the gray jay and the source-departure decision for rate-maximizing scatterhoarders. Behaviour 120, 51–68. Waite, T.A., Reeve, J.D., 1995. Source-use decisions by hoarding gray jays: effects of local cache density and food value. J. Avian Biol. 26, 59–66. Wenny, D.G., Levey, D.J., 1998. Directed seed dispersal by bellbirds in a tropical cloud forest. Proc. Natl. Acad. Sci. U.S.A. 95, 6204–6207. Wróbel, A., Zwolak, R., 2017. Deciphering the effects of disperser assemblages and seed mass on patterns of seed dispersal in a rodent community. Integr. Zool. 12, 457–467. Yi, X., Yang, Y., Curtis, R., Bartlow, A.W., Agosta, S.J., Steele, M.A., 2012. Alternative strategies of seed predator escape by early-germinating oaks in Asia and North America. Ecol. Evol. 2, 487–492. Zhang, H., Steele, M.A., Zhang, Z., Wang, W., Wang, Y., 2014. Rapid sequestration and recaching by a scatter-hoarding rodent (Sciurotamias davidianus). J. Mammal. 95, 480–490.
122