ARTICLE IN PRESS
Journal of Arid Environments xxx (2010) 1e6
Contents lists available at ScienceDirect
Journal of Arid Environments
journal homepage: www.elsevier.com/locate/jaridenv
The composition of seed banks on kangaroo rat (Dipodomys spectabilis)
mounds in a Chihuahuan Desert grassland
Terri L. Koontz*, Heather L. Simpson
Department of Biology, 167 Castetter Hall MSC03 2020, 1 University of New Mexico, Albuquerque, NM 87131, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 May 2009
Received in revised form
4 March 2010
Accepted 9 March 2010
Available online xxx
Disturbance is a major factor determining the plant community structure of ecological systems. In
particular, small-scale disturbances from animals can increase heterogeneity and species diversity in
plant communities. Kangaroo rat (Dipodomys spectabilis) mounds are small-scale disturbances that
support unique plant communities compared to surrounding habitats. The seed bank, which is usually an
overlooked part of plant communities, is an important trait of some plant species that allows for their
persistence in unpredictable environments. To further understand how small-scale disturbances by
kangaroo rats affect the plant community in Chihuahuan Desert grassland, we studied the composition of
the seed bank for eight forb taxa on and off kangaroo rat mounds. Kangaroo rat mounds accumulated
more seeds and supported different seed compositions than the adjacent grassland. Also, kangaroo rat
mounds had different microhabitats than areas away from the mound. This suggests that kangaroo rat
mounds facilitate some plant species by providing microhabitats that contain ‘safe sites’ for seeds to
accumulate where adult plant populations can establish and colonize thus increasing plant diversity and
altering plant structure in communities.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Descurainia pinnata
Disturbance
Granivores
Keystone species
Plant communities
Succession
1. Introduction
Seed banks rarely correlate with the above ground vegetation
(Kinucan and Smeins, 1992; Lopez-Marino et al., 2000; Major and
Pyott, 1966; Rabinowitz, 1981; Thompson and Grime, 1979). This
is in part because perennial species often have transient short-lived
seed banks (Thompson et al., 1998) or no seed bank, which means
the long-lived adult populations are maintained above ground
rather than below. Annual plants, however, often produce longlived, persistent seed banks (Guo et al., 1999; Meissner and Facelli,
1999; Milberg and Hansson, 1993; Thompson and Grime, 1979); so,
that if environmental cues are not met for seed germination, then
the population is maintained primarily below ground. This is
particularly important in unpredictable environments like the
deserts of southwestern North America where seed banks maintain
plant species diversity below ground during periods of drought
(Thompson, 2000).
Moore (1980) theorized that frequently and unpredictably
disturbed areas have plant compositions of early successional
species that rely on a large persistent seed bank rather than late
* Corresponding author. Tel.: þ1 505 277 9342; fax: þ1 505 277 5355.
E-mail addresses: [email protected] (T.L. Koontz), [email protected] (H.L.
Simpson).
successional species that establish from invasion (Thompson and
Grime, 1979). As an area becomes available, early successional
species become established. Then through facilitation, tolerance,
or inhibition, communities can reach a climax when early
successional species become less suitable for the new environment. Disturbance destroys this climax stage and returns the
system to early colonization. Specifically, regular unpredictable
tidal fluctuations in a New Jersey freshwater marshland had seed
banks with similar plant compositions as the above ground plant
community (Leck and Graveline, 1979). Grassland in central New
Mexico, which is prone to drought, had similar seed and above
ground plant populations (Henderson et al., 1988). The composition of above and below ground plant populations on frequently
and unpredictably disturbed areas are exceptions to the generalized pattern where the seed bank rarely correlates with the above
ground vegetation.
Soil disturbance resulting from biotic factors can affect the
composition of the plant community both above and below ground.
In particular, biotic small-scale disturbances can create unique
species combinations in plant communities (Bullock, 2000; King,
1977; Peart, 1989; Platt, 1975). For example, porcupine digs,
which are soil pockets containing both litter and seeds (Alkon,
1999), contributed to maintaining a high diversity of annual
plants (Boeken et al., 1995); gopher disturbances increased legumes
while decreasing perennial grass species (Hobbs et al., 2007); and
0140-1963/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jaridenv.2010.03.008
Please cite this article in press as: Koontz, T.L., Simpson, H.L., The composition of seed banks on kangaroo rat (Dipodomys spectabilis) mounds...,
Journal of Arid Environments (2010), doi:10.1016/j.jaridenv.2010.03.008
ARTICLE IN PRESS
2
T.L. Koontz, H.L. Simpson / Journal of Arid Environments xxx (2010) 1e6
seed densities were either higher or lower around ant mounds than
in adjacent habitats depending on ant species and disturbance
location. Specifically, mounds constructed by the ant Formica
exsecta in sub-alpine grassland had fifteen times more total seeds
and slightly higher species richness than the surrounding grassland
(Schutz et al., 2008). In contrast, Dostal (2005) found half the
number of seeds on ant mounds of Tetramorium caespitum
compared to control plots in Slovakian mountain grassland. Given
that many biotic small-scale disturbances can impact the number
and types of plants and their seeds in an area, it is not surprising
that disturbances caused by kangaroo rats also influence plant
communities.
Banner tail kangaroo rat (Dipodomys spectabilis) mounds are
noticeable features of many arid landscapes, creating a mosaic of
disturbed patches with distinct environmental conditions (Guo,
1996; Schroder and Geluso, 1975). In the Chihuahuan Desert, soil
on kangaroo rat mounds had higher concentrations of nitrogen and
lower water availability after rainfall than non-mound soil (Greene
and Reynard, 1932; Moorhead et al., 1988; Mun and Whitford,
1990). The extent to which these disturbances affect the vegetation is so profound that D. spectabilis is considered a keystone
species (Heske et al., 1993). Many studies have addressed vegetation structure and plant species composition on and off kangaroo
rat mounds (Davidson and Lightfoot, 2006, 2008; Fields et al., 1999;
Guo, 1996, 1998; Moorhead et al., 1988; Moroka et al., 1982; Mun
and Whitford, 1990). They have found that annual plant cover is
generally higher on mounds than off (Davidson and Lightfoot,
2008; Moorhead et al., 1988; Moroka et al., 1982). Furthermore,
kangaroo rat mounds create an ecotone where one habitat type, the
mound area, shifts to another, the inter-mound, with the highest
plant species diversity occurring where the two habitats meet (Guo,
1996). Even though there has been extensive research on the
vegetation associated with mounds, no research has examined the
composition of the seed bank on kangaroo rat mounds.
Kangaroo rat activity (mound building and herbivory) may
affect the seed bank in a plant community in several ways. First, the
plant community on mounds typically consists of a wide variety of
annual plant species (Davidson and Lightfoot, 2008; Guo, 1996,
1998). Many plants disperse their seeds via the seed rain close to
the parent plant (Willson and Traveset, 2000). As a consequence,
these annuals may directly deposit a large proportion of their seeds
into the seed bank on the mound. Second, kangaroo rat seed
caching and selective foraging can change seed distributions and
densities in plant communities (Davidson et al., 1980; Reichman,
1979). Banner tail kangaroo rats are larder hoarders that cache
seeds in mounds at depths near or below 30 cm of the soil surface
(Reichman et al., 1985), making them unavailable for germination
(Kemp, 1989). Selective foraging by a kangaroo rat guild (Dipodomys
merriami, Dipodomys ordii, and D. spectabilis) in the Chihuahuan
Desert in Arizona changed the plant community by decreasing
above ground populations of large seeded winter annuals (Brown
and Heske, 1990; Samson et al., 1992). Therefore, to more fully
understand plant populations and the effects of kangaroo rat
activity on plant populations, an investigation of the seed bank is
important.
We surveyed part of the seed bank in Chihuahuan Desert
grassland to understand the impact of banner tail kangaroo rat
mounds on plant communities. Regarding eight dominant forb taxa
in the area, we asked the following questions: 1) Do mounds
compared to off mounds have more seeds in the seed bank? 2) Does
the species composition of the seed bank differ between mounds
and off mounds? We predict that the eight plant taxa will have seed
banks that differ between mounds and off mounds and that
mounds will have high seed numbers of early successional plant
species.
2. Materials and methods
2.1. Study site
We collected data at Sevilleta National Wildlife Refuge Longterm Ecological Research (LTER) site in Socorro County in central
New Mexico, latitude 34180 5000 N, longitude 106 420 3600 W. The
Sevilleta is located along a transition zone among several biomes:
Great Plains Grassland, Colorado Plateau Shrub Steppe, Conifer
Woodland and Chihuahuan Desert. The Chihuahuan Desert grassland is dominated by black grama (Bouteloua eriopoda) with associated species consisting of creosote bush (Larrea tridentata), yucca
(Yucca glauca), snakeweed (Gutierrezia sarothrae) and several
annual and perennial forbs. Long-term total precipitation from
September 2000 to August 2001 was 293.9 mm where the mean
annual precipitation for 2000 and 2001 in the area was 249.9 mm.
Most of this precipitation in 2001 occurred during the summer in
August. Long-term monthly mean temperatures for 2001 were
0.5 C and 23.8 C in January and August, respectively. We used
meteorological data obtained from the Sevilleta LTER website,
http://sev.lternet.edu/project_details.php?id¼SEV001, for precipitation and temperature assessments.
2.2. Collecting soil samples
We randomly selected 10 active kangaroo rat mounds over an
area of 1.1 km2. We collected soil on the kangaroo rat mound at the
base of the mound (B, termed ‘base’), 1 m away from the base to
capture the disturbed area surrounding the mound (S, termed
‘surrounding’), and approximately 5 m from the surrounding
sample to represent off mound locations (Fig. 1). Off mound locations were at the edge of the nearest black grama clump (E, termed
‘edge’) and in an open undisturbed area between black grama
clumps (I, termed ‘inter-space’). We sampled the edge and interspace of black grama since this is the dominant grass species in the
Chihuahuan Desert and should represent a large proportion of the
microhabitats that are off the mound. We collected subsamples in
the four cardinal directions resulting in a total of 16 subsamples per
mound (4 subsamples by 4 mound locations). Some subsamples for
some mounds were destroyed. To ensure that the data for mounds
with missing samples were orthogonal, we used a lesser number of
samples so that the number of samples was the same for each
mound location.
We collected samples using a soil auger with 10 cm diameter
and 2 cm depth during the last week of August 2001 when most
spring annuals would have already dispersed their seeds. We
recorded percent cover of undisturbed and disturbed soil, gravel,
litter and vegetation within a 30 cm 30 cm area at each
subsample location to understand the potential effects of these
physical variables on seed accumulation. We calculated the
percentage of undisturbed soil by subtracting the sum of the other
physical variable percentages from 100%. We then described
microhabitats using the combination of these five physical variables. This should not be confused with Harper’s et al. (1961) term
of a ‘safe site’, which is an area at a much smaller scale where a seed
can successfully germinate and a species can establish.
2.3. Processing samples
We dried soil samples in an oven for 48 h at 50 C and sieved
them using the finest possible sieve to capture small seeds and
a larger sieve to exclude large particles. To further separate seeds
from remaining soil we floated samples in a salt solution (sodium
bicarbonate (5 g), sodium hexametaphosphate (10 g), and magnesium sulphate (25 g) in 200 ml of tap water) and dried them
Please cite this article in press as: Koontz, T.L., Simpson, H.L., The composition of seed banks on kangaroo rat (Dipodomys spectabilis) mounds...,
Journal of Arid Environments (2010), doi:10.1016/j.jaridenv.2010.03.008
ARTICLE IN PRESS
T.L. Koontz, H.L. Simpson / Journal of Arid Environments xxx (2010) 1e6
3
Fig. 1. Diagram of where soil samples were collected on and off kangaroo rat mounds. Circles with letters inside represent areas where soil samples were taken (on mound:
B ¼ base and S ¼ surrounding; off mound: E ¼ edge and I ¼ inter-space). Areas hatched with vertical stripes indicate locations on the mound and horizontal stripes indicate
locations off the mound.
(Malone, 1967). From this organic material, a subsample was taken
to represent the sample. We counted and tested seeds from these
subsamples for viability. We identified eight target taxa based on
their high occurrences in the above ground vegetation at an adjacent study site (T. Koontz and H. Simpson, personal observation).
Four of the taxa (Cryptantha crassisepala, Descurainia pinnata, Phacelia integrifolia, and Plantago patagonica) are spring annuals, three
(Astragalus missouriensis, Lesquerella fendleri, and Oenothera spp.)
are perennial forbs that flower in the spring, and one (Sphaeralcea
spp.) is a perennial forb that flowers in the fall. Since we could not
distinguish species for the seeds of Sphaeralcea and Oenothera taxa
at our site, we analyzed these seeds at the genus level. We assumed
that spring annuals are early successional species and perennial
species are not. We made identifications under a dissecting
microscope using a reference collection compiled by Sevilleta
biologists along with seeds that we had collected. We tested for
viability using the pressure method (see Meissner and Facelli, 1999
and Roberts and Ricketts, 1979 for details on the pressure method).
Due to time constraint we only counted the number of viable seeds
from the eight taxa.
mound locations were significantly different from each other for
total seed numbers. To determine whether the seed composition
differed among mound locations, we used a multivariate analysis
of variance (MANOVA) where ranks of total seed numbers for the
eight target taxa were dependent variables and the independent
variable was mound location. We also tested whether five physical variables (disturbed and undisturbed soil, gravel, litter, and
vegetation) differed among mound locations. Again, a MANOVA
was used for this analysis where ranks of percent cover for the
physical variables were the dependent variables and mound
2.4. Statistical analysis
We used an analysis of variance (ANOVA) to assess differences
in total number of seeds among the four different kangaroo rat
mound locations. We added seed counts from the subsamples for
each of the four mound locations (base, surrounding, edge, and
inter-space) on each mound to calculate the total number of
seeds. Since neither total number of seeds nor any transformations of the data had normal distributions, we used the
GLIMMIX procedure in SAS that does not require normal distributions of populations. For the model, the total seed number was
the dependent variable and mound location (mound: base and
surrounding; off mound: edge and inter-space) was the independent variable. We did not pool the data together for mound
and off mound samples since the ANOVA indicated that the four
Fig. 2. Mean total seed numbers per mound for each of the four mound locations
(B ¼ base, S ¼ surrounding, E ¼ edge, and I ¼ inter-space). Bars indicate the standard
error. Only seed abundances of Descurainia pinnata and Cryptantha crassisepala had
significant differences among mound locations. D. pinnata had significantly more seeds
at the base and surrounding locations than off mound locations (ANOVA: F3,36 ¼ 51.28;
P < 0.0001). C. crassisepala had significantly more seeds at the surrounding mound
locations than all other mound locations (ANOVA: F3,36 ¼ 5.54; P ¼ 0.0031).
Please cite this article in press as: Koontz, T.L., Simpson, H.L., The composition of seed banks on kangaroo rat (Dipodomys spectabilis) mounds...,
Journal of Arid Environments (2010), doi:10.1016/j.jaridenv.2010.03.008
ARTICLE IN PRESS
4
T.L. Koontz, H.L. Simpson / Journal of Arid Environments xxx (2010) 1e6
location was the independent variable. Statistical analyses were
performed in SAS version 9.2 (SAS Institute Inc., Cary, North
Carolina).
3. Results
Average total seed numbers differed significantly among mound
locations (F3,36 ¼ 31.30, P < 0.0001). We extracted 10 658 seeds of
the eight target taxa: 9269 from mounds and 1389 seeds off
mounds (see Fig. 1 to reference mound locations). The base had
significantly more seeds (n ¼ 6662) than all other mound locations,
and surrounding mound areas had significantly more seeds
(n ¼ 2607) than both off mound locations (edge and inter-space).
The inter-space had the lowest number of seeds (n ¼ 377), but this
was not significantly different from the number of seeds (n ¼ 1012)
at the edge.
Seeds of all eight taxa were found at each mound location
(Fig. 2) and seed composition of the eight taxa differed significantly
among mound locations (MANOVA, P < 0.0001). The first canonical
axis (CAN1), which accounted for 93% of total variation, indicated
that mound locations supported a different seed assemblage than
off mound locations. The second canonical axis (CAN2, 5% total
variation) indicated differences in seed bank composition between
the two off mound locations (Fig. 3A). D. pinnata accounted for 77%
of seeds extracted and explained most of the variation in seed
composition among mound locations since it had the highest
loading on the first canonical axis. When we compared the number
of D. pinnata seeds extracted among locations, base locations had
Fig. 3. Canonical coefficients for (A) the seed composition for the eight plant taxa during summer 2001 (MANOVA: F8,31 ¼ 24.05, P < 0.0001) and (B) the five physical variables that
were used to describe the mound microhabitats where soil samples were collected (MANOVA: F5,34 ¼ 42.67, P < 0.0001). Ellipses represent two standard deviations from the mean
for both CAN1 and CAN2. Note that CAN1 groups different seed compositions and microhabitats between on mound locations and off mound locations and CAN2 groups different
microhabitats between the off mound locations. CAN2 for seed composition shows only a slight grouping for seed composition between off mound samples. Roy’s greatest root
statistics are reported for MANOVAs.
Please cite this article in press as: Koontz, T.L., Simpson, H.L., The composition of seed banks on kangaroo rat (Dipodomys spectabilis) mounds...,
Journal of Arid Environments (2010), doi:10.1016/j.jaridenv.2010.03.008
ARTICLE IN PRESS
T.L. Koontz, H.L. Simpson / Journal of Arid Environments xxx (2010) 1e6
over three-times more seeds than surrounding locations, 29.7
times more than edge locations and 67.3 times more than interspace locations (Fig. 2).
The five physical variables also differed significantly among
mound locations (MANOVA, P < 0.0001), demonstrating that
mound locations contained different microhabitats. The first
canonical axis (71% of total variation) indicated different microhabitats between mound (base and surrounding) and off mound
(edge and inter-space). The second canonical axis (28% of total
variation) indicated different microhabitats among mound, edge
and inter-space (Fig. 3B). Gravel, litter and disturbed soil had the
highest loadings on the first canonical axis. Mound microhabitats
had higher percent cover of these three attributes than off mound
microhabitats (Fig. 4). Most of the variation between edge and
inter-space microhabitats was explained by vegetation cover, with
four-fold greater vegetation cover in edge microhabitats compared
with inter-space microhabitats (Fig. 4).
4. Discussion
We found that kangaroo rat mounds influenced the size and
composition of a subset of the seed bank in this arid grassland.
Mounds had nearly seven-times more seeds than the adjacent
grassland. The base of kangaroo rat mounds had 64% of total seeds,
surrounding areas of mounds had 25% of total seeds, and off mound
locations had only 12% of total seeds. Mounds contained larger seed
banks of four taxa (C. crassisepala, D. pinnata, Oenothera spp. and
Sphaeralcea spp.) compared to off mounds. Of these four taxa, only
the two annuals (C. crassisepala and D. pinnata) had significantly
larger seed banks on the mound where D. pinnata had six-times
more seeds on the mound than any of the other seven taxa (Fig. 2).
The mechanism of seed dispersal is an important factor in
determining the number of seeds in the seed bank and their spatial
distribution. D. pinnata is a small seeded, weedy annual that can
produce copious amounts of seeds. In 2001, we observed that above
ground plant populations on kangaroo rat mounds were dominated
by D. pinnata. Other studies support this observation, indicating
that above ground populations of D. pinnata are more prevalent on
disturbed, nitrogen rich kangaroo rat mounds (Davidson and
Lightfoot, 2008; Guo, 1996, 1998). Our results show that D. pinnata seeds were highest on the mound where most parental plants
Fig. 4. Mean percent cover for the five physical variables per mound for each of the
four mound locations (B ¼ base, S ¼ surrounding, E ¼ edge, and I ¼ inter-space). Only
on mound locations contained disturbed soil. The combination of these five physical
variables was similar for the on mound locations and was quite different from off
mound locations. See Fig. 3 for statistics.
5
exist. The size of D. pinnata’s seed pools decreased farther away
from parental plants and the mound area. This pattern is typical of
gravity dispersed seeds where the majority of seeds are deposited
near the parent plant.
The remaining plant taxa that we looked at did not show
a similar pattern of seed distribution. Many taxa had low seed
populations with more homogeneous distributions. Of these taxa
there was no obvious mode of dispersal. None of the other seven
plant taxa had seeds small enough or had appendages for wind
dispersal (T. Koontz and H Simpson, personal observation). Since
water is a limiting resource in this arid grassland, it is unlikely that
any of our plant taxa rely on water as their primary mode for
dispersal. While some plants rely on more than one mode of
dispersal (Willson and Traveset, 2000), it is likely that most of the
seeds that we extracted from the soil are dispersed by animals.
The main function of dispersal is to move a seed to an area that
optimizes the survival of the seed and its seedling (Willson and
Traveset, 2000). This leads to many plant species having seed
banks with patchy or heterogeneous distributions on the landscape
(Thompson, 1986). Many of our plant taxa, however, had low seed
populations with homogeneous distributions. Six of the taxa (C.
crassisepala, D. pinnata, Oenothera spp., Phacelia integrifolia, P.
patagonica, and Sphaeralcea spp.) that we assessed are known prey
items of the consumers that exist in this grassland (Andrew Edelman, personal observation; Hope and Parmenter, 2007); and
selective foraging can actually decrease the spatial patchiness of
seed populations (Reichman, 1979). If the seeds we extracted from
the soil were selectively foraged on, this could explain why some
taxa had seed banks with low seed numbers and homogeneous
distributions. Animal dispersal and consumption both probably
contributed to the seed bank sizes and distributions that we found
in this grassland.
Different microhabitats that occupy these small-scale disturbed
areas can also explain the size and distribution of seed populations
on kangaroo rat mounds compared to the adjacent grassland. Both
the base and surrounding mound microhabitats had more
disturbed soil, gravel, and litter than the microhabitats of off
mound locations. One of these physical variables or a combination
of these physical variables could have provided ‘safe sites’ where
these species not only can germinate and establish in this grassland
on kangaroo rat mounds, but also can accumulate seeds. The
increased disturbed soil and gravel could provide depressions and
cracks where dispersed seeds can accumulate. The disturbance and
litter that are common on the mound probably increased nitrogen
levels; and other studies have indicated that kangaroo rat mounds
do have higher nitrogen content than adjacent habitats that they
occur in (Greene and Reynard, 1932; Moorhead et al., 1988; Mun
and Whitford, 1990). This increase in nitrogen can promote the
growth of seedlings that have germinated from captured seeds; so,
species that have dispersed and accumulated seeds on the mound
can persist in a community.
An evaluation of all seed bank populations and the above ground
vegetation on kangaroo rat mounds at different stages of a mound’s
disturbance would help to determine which species are functioning
as early successional species in this arid grassland. Kangaroo rat
mounds are known to have unique plant assemblages compared to
their surrounding environments (Davidson and Lightfoot, 2008;
Moorhead et al., 1988; Moroka et al., 1982). Which of these species
are early successional species is uncertain. For our study we evaluated a subset of the grassland community rather than what is
considered dominant species on kangaroo rat mounds. All of the
plant taxa that we evaluated, except one, had small seed bank
populations, suggesting that they are not early successional species.
‘Snap shots’ are what most seed bank studies report. Understanding the spatial and temporal dynamics of the seed bank
Please cite this article in press as: Koontz, T.L., Simpson, H.L., The composition of seed banks on kangaroo rat (Dipodomys spectabilis) mounds...,
Journal of Arid Environments (2010), doi:10.1016/j.jaridenv.2010.03.008
ARTICLE IN PRESS
6
T.L. Koontz, H.L. Simpson / Journal of Arid Environments xxx (2010) 1e6
community can elucidate how plant communities function. In
desert ecosystems this is especially important, because certain
plants in the desert are only represented in the community by their
seed bank (Kemp, 1989). Our study, even though it was only a ‘snap
shot’, increased the knowledge of how some desert species respond
to small-scale disturbance in a desert grassland community. In
particular, we demonstrated that not only do kangaroo rat mounds
contain unique plant assemblages above ground, but they also
support different plant communities below ground compared to
the surrounding area. More seed bank studies like ours are essential
to understand how consumers and specifically how their disturbances facilitate particular plant populations and communities.
Acknowledgements
We thank Sevilleta Long-Term Ecological Research and United
States Fish and Wildlife Service for allowing us to conduct research
at the Sevilleta National Wildlife Refuge and A. Davidson, J. Nighbert,
R. Parmenter, K. Vanderbilt, and K. Wetherill for discussions. J.
Barlow gave advice on extracting seeds from soil. J. Brown, S. Collins,
M. Donovan, and D. Marshall provided helpful comments on earlier
versions of this manuscript. This is Sevilleta publication 511.
References
Alkon, P.U., 1999. Microhabitat to landscape impacts: crested porcupine digs in the
Negev Desert highland. Journal of Arid Environments 41, 183e202.
Boeken, B., Shachak, M., Gutterman, Y., Brand, S., 1995. Patchiness and disturbance:
plant community responses to porcupine diggings in the central Negev. Ecography 18, 410e422.
Brown, J.H., Heske, E.J., 1990. Control of a desertegrassland transition by a keystone
rodent guild. Science 250, 1705e1707.
Bullock, J.M., 2000. Gaps and seedling colonization. In: Fenner, M. (Ed.), Seeds: the
Ecology of Regeneration in Plant Communities, second ed. CABI Publishing,
New York, pp. 375e395.
Davidson, A.D., Lightfoot, D.C., 2006. Keystone rodent interactions: prairie dogs and
kangaroo rats structure the biotic composition of a desertified grassland.
Ecography 29, 755e765.
Davidson, A.D., Lightfoot, D.C., 2008. Burrowing rodents increase landscape
heterogeneity in a desert grassland. Journal of Arid Environments 72,
1133e1145.
Davidson, D.W., Brown, J.H., Inouye, R.S., 1980. Competition and the structure of
granivore communities. BioScience 30, 233e238.
Dostal, P., 2005. Effect of three mound-building ant species on the formation of soil
seed bank in mountain grassland. Flora 200, 148e158.
Fields, M.J., Coffin, D.P., Gosz, J.R., 1999. Burrowing activities of kangaroo rats and
patterns in plant species dominance at a shortgrass steppe-desert grassland
ecotone. Journal of Vegetation Science 10, 123e130.
Guo, Q., 1996. Effects of banner tail kangaroo rat mounds on small-scale plant
community structure. Oecologia 106, 247e256.
Guo, Q., 1998. Microhabitat differentiation in Chihuahuan Desert plant communities. Plant Ecology 139, 71e80.
Guo, Q., Rundel, P.W., Goodall, D.W., 1999. Structure of desert seed banks:
comparisons across four North American desert sites. Journal of Arid Environments 42, 1e14.
Greene, R.A., Reynard, C., 1932. The influence of two burrowing rodents, Dipodomys
spectabilis spectabilis (kangaroo rat) and Neotoma albiqula albigula (pack rat) on
desert soil of Arizona. Ecology 13, 73e80.
Harper, J.L., Clatworthy, J.N., McNaughton, I.H., Sagar, G.R., 1961. The evolution and
ecology of closely related species living in the same area. Evolution 15,
209e227.
Henderson, C.B., Petersen, K.E., Redak, R.A., 1988. Spatial and temporal pattern in
the seed bank and vegetation of a desert grassland community. Journal of
Ecology 76, 717e728.
Heske, E.J., Brown, J.H., Guo, Q., 1993. Effects of kangaroo rat exclusion on vegetation
structure and plant species diversity in the Chihuahuan Desert. Oecologia 95,
520e524.
Hobbs, R.J., Yates, S., Mooney, H.A., 2007. Long-term data reveal complex dynamics
in grassland in relation to climate and disturbance. Ecological Monographs 77,
545e568.
Hope, A.G., Parmenter, R.R., 2007. Food habits of rodents inhabiting arid and semiarid ecosystems of central New Mexico. Special Publication of the Museum of
Southwestern Biology 9, 1e75.
Kemp, P.R., 1989. Seed banks and vegetation processes in deserts. In: Leck, M.A.,
Parker, V.T., Simpson, R.L. (Eds.), Ecology of Soil Seed Banks. Academic Press Inc.,
San Diego, pp. 257e281.
King, T.J., 1977. The plant ecology of ant-hills in calcareous grasslands. III. Factors
affecting population sizes of selected species. Ecology 65, 279e315.
Kinucan, R.J., Smeins, F.E., 1992. Soil seed bank of a semiarid Texas grassland under
three long-term (36 years) grazing regimes. American Midland Naturalist 128,
11e21.
Leck, M.A., Graveline, K.J., 1979. The seed bank of a freshwater tidal marsh. American Journal of Botany 66, 1006e1015.
Lopez-Marino, A., Luis-Calabuig, E., Fillat, F., Bermudez, F.F., 2000. Floristic
composition of established vegetation and the soil seed bank in pasture
communities under different traditional management regimes. Agriculture,
Ecosystems and Environment 78, 273e282.
Major, J., Pyott, W.T., 1966. Buried, viable seeds in two California bunchgrass sites
and their bearing on the definition of a flora. Vegetatio 13, 253e282.
Malone, C.R., 1967. A rapid method for enumeration of viable seeds in soil. Weeds
15, 381e382.
Meissner, R.A., Facelli, J.M., 1999. Effects of sheep exclusion on the soil seed bank
and annual vegetation in chenopod shrublands of South Australia. Journal of
Arid Environments 42, 117e128.
Milberg, P., Hansson, M.L., 1993. Soil seed bank and species turnover in a limestone
grassland. Journal of Vegetation Science 4, 35e42.
Moore, P.D., 1980. Soil seed banks. Nature 284, 123e124.
Moorhead, D.L., Fisher, F.M., Whitford, W.G., 1988. Cover of spring annuals on
nitrogen-rich kangaroo rat mounds in Chihuahuan Desert grassland. American
Midland Naturalist 120, 443e447.
Moroka, N., Beck, R.F., Pieper, R.D., 1982. Impact of burrowing activity of the bannertail kangaroo rat on southern New Mexico desert rangelands. Journal of Range
Management 35, 707e710.
Mun, H., Whitford, W.G., 1990. Factors affecting annual plants assemblages on
banner-tailed kangaroo rat mounds. Journal of Arid Environments 18, 165e173.
Peart, D.R., 1989. Species interactions in a successional grassland I. Seed rain and
seedling establishment. Journal of Ecology 77, 236e251.
Platt, W.J., 1975. The colonization and formation of equilibrium plant species
associations on badger disturbances in a tall-grass prairie. Ecological Monographs 45, 285e305.
Rabinowitz, D., 1981. Buried viable seeds in a North American tall-grass prairie: the
resemblance of their abundance and composition to dispersing seeds. Oikos 36,
191e195.
Reichman, O.J., 1979. Desert granivore foraging and its impact on seed densities and
distributions. Ecology 60, 1086e1092.
Reichman, O.J., Wicklow, D.T., Rebar, C., 1985. Ecological and mycological characteristics of caches in the mounds of Dipodomys spectabilis. Journal of
Mammalogy 66, 643e651.
Roberts, H.A., Ricketts, M.E., 1979. Quantitative relationships between the weed
flora after cultivation and the seed population in the soil. Weed Research 19,
269e275.
Samson, D.A., Philippi, T.E., Davidson, D.W., 1992. Granivory and competition as
determinants of annual plant diversity in the Chihuahuan Desert. Oikos 65,
61e80.
Schroder, G.D., Geluso, K.N., 1975. Spatial distribution of Dipodomys spectabilis
mounds. Journal of Mammalogy 56, 363e368.
Schutz, M., Kretz, C., Dekoninck, L., Iravani, M., Risch, A.C., 2008. Impact of Formica
exsecta Nyl. on seed bank and vegetation patterns in a subalpine grassland
ecosystem. Journal of Applied Entomology 132, 295e305.
Thompson, K., 1986. Small-scale heterogeneity in the seed bank of an acidic
grassland. Journal of Ecology 74, 733e738.
Thompson, K., Bakker, J.P., Bekker, R.M., Hodgson, J.G., 1998. Ecological correlates of
seed persistence in soil in the north-west European flora. Journal of Ecology 86,
163e169.
Thompson, K., 2000. The functional ecology of soil seed banks. In: Fenner, M. (Ed.),
Seeds: the Ecology of Regeneration in Plant Communities, second ed. CABI
Publishing, New York, pp. 215e235.
Thompson, K., Grime, J.P., 1979. Seasonal variation in the seed banks of herbaceous
species in ten contrasting habitats. Journal of Ecology 67, 893e921.
Willson, M.F., Traveset, A., 2000. The ecology of seed dispersal. In: Fenner, M. (Ed.),
Seeds: the Ecology of Regeneration in Plant Communities, second ed. CABI
Publishing, New York, pp. 85e110.
Please cite this article in press as: Koontz, T.L., Simpson, H.L., The composition of seed banks on kangaroo rat (Dipodomys spectabilis) mounds...,
Journal of Arid Environments (2010), doi:10.1016/j.jaridenv.2010.03.008