Journal of Arid Environments 135 (2016) 33e38

Contents lists available at ScienceDirect

Journal of Arid Environments

journal homepage: www.elsevier.com/locate/jaridenv

The importance of plant cover and predation in shaping a desert

community
Moran Segoli a, *, Elli Groner a, b, Yoram Ayal a
a
Mitrani Department of Desert Ecology, BIDR, Ben-Gurion University, Sede Boker Campus 84990, Israel
b
Dead Sea and Arava Science Center, Mitzpe Ramon, Israel

a r t i c l e i n f o a b s t r a c t

Article history: The classical paradigm describe desert communities as being controlled by abiotic conditions and
Received 11 February 2015 therefore it was assumed that predation had a limited role in the determination of desert community
Received in revised form structure and that only one effective trophic level was expected. In the Negev desert, the desert isopod
28 November 2015
(Hemilepistus raeumuri) is common in the less productive habitats but almost absent from the more
Accepted 4 August 2016
Available online 25 August 2016
productive wadis. In the present study we tested whether predation by ectothermic predators has a role
in controlling isopod populations in wadis. We hypothesized that plant cover provides a refuge for these
predators from secondary endothermic predators. We addressed this hypothesis by comparing isopod
Keywords:
Community structure
survivorship with and without predation exclusion, and with and without plant cover removal. Isopods
Desert protected from predation had a sustainable population whereas unprotected isopods had an unsus-
Food-web tainable sink population. Removal of plant cover did not affect the survivorship of predator-protected
Macrodetritivores isopods, however it did increase the survivorship and reduced predation pressure when not protected
Ectothermic predators from predators, for at least four years. Hence, we conclude that predation controls isopod population and
causes an unsustainable sink population in wadis. We have shown that plant cover mediates predation
and that some habitats in the desert have an effective third trophic level, suggesting that energy is not
the major limiting factor in determining the length of food chains.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction
One of the central issues in community ecology is the relative
importance of bottom up (e.g., water) limitation and top-down (e.g.,
consumers) control in shaping terrestrial communities (Hopcraft
et al., 2010). It has been suggested that, in terrestrial commu-
nities, there is a positive correlation between productivity and the
number of effective trophic levels (Fretwell, 1977; Oksanen et al.,
1981). A trophic level was defined as effective if it effectively con-
trols the trophic level below it (Fretwell, 1977) suggesting that the
effect of removal of one trophic level will cascade to lower trophic
levels. Accordingly, Fretwell (1977) recognized two different states
in terrestrial communities, namely, a “green” state, where the
number of trophic levels is odd (i.e., one or three) and consequently
plants are resource limited, and a “brown” state, where the number
of trophic levels is even (i.e., two or four) and plants are herbivore-
* Corresponding author. Current address: Ramat Negev Desert Agro-Research
Center, Negev, Israel.
E-mail address: moran.segoli@gmail.com (M. Segoli).
controlled (Hairston et al., 1960; Oksanen et al., 1981; Oksanen and
Oksanen, 2000; Bond, 2005). In resource-poor hot deserts, only one
effective trophic level is expected (i.e., deserts represent a “green”
state) and plants are water limited and therefore herbivores and
predators are expected to play a minor role in shaping hot desert
communities (Noy-Meir, 1973, 1985; Shmida, 1985; Megias et al.,
2011).
However, an alternative explanation suggests that deserts could
have more trophic levels and that predation can play an important
role in shaping hot desert communities (Polis, 1991; Moore et al.,
2004; Borer et al., 2005; Ayal, 2007). Hot deserts accumulate
detritus which is more predictable and long lasting food source for
primary consumers, especially macrodetritivorous arthropods
(Noy-Meir, 1979). The macrodetritivorous arthropods serve as a
major conduit transferring energy from primary producers to
higher trophic levels (Crawford, 1981; Ayal et al., 2005). The
abundant macrodetritivores can support two additional trophic
levels, namely, primary ectothermic predators (e.g., arachnids and
reptiles) and secondary endothermic predators (i.e., mammals and
birds). Ayal (2007) suggested that the distribution of the primary

http://dx.doi.org/10.1016/j.jaridenv.2016.08.008
0140-1963/© 2016 Elsevier Ltd. All rights reserved.
34 M. Segoli et al. / Journal of Arid Environments 135 (2016) 33e38
ectothermic predators is determined by predation intensity of the
endothermic secondary predators that is affected by plant cover:
primary predators are abundant in more productive habitats where
they are rather protected by plant cover, but rare in less productive
habitats where they are exposed to heavy predation from second-
ary predators (Fig. 1). Therefore, macrodetritivores suffer high
predation pressure in high plant cover habitats by the abundant
primary predators, and rather low predation in low plant cover
habitats where primary predators are rare and their foraging is
limited (Groner and Ayal, 2001; Ayal, 2007). One of the most
abundant primary consumers in deserts are isopods and in the
Negev desert one of the most abundant species is the desert isopod
Hemilepistus raeumuri. This species is less abundant in the lower
areas of the desert water sheds (wadis) than in the adjacent slopes
(Shachak et al., 1976). In the present study we explore Ayal (2007)
theory in experiments that manipulate the distribution of the
desert isopod Hemilepistus raeumuri. Specifically, we tested
whether the low abundance of isopods in more productive wadi
habitats is due to predation by ectothermic predators. This was
tested by (a) protecting isopods in plant covered wadi plots from
predation by ectothermic predators, and (b) by removing the plant
cover and monitoring the densities and predation intensity on
isopods.
2. Methods
2.1. Study area
The study was carried out in a wadi (ephemeral riverbed) in the
Negev Desert Highlands approximately 10 km north of Sede Boqer
Campus of the Ben Gurion University (30º510 N, 34º570 E). The study
site is a hilly limestone area with an average annual precipitation of
92 mm with extreme values of 34 mm and 167 mm. For a detailed
description of the study area and vegetation see Ayal and Merkl
(1994). Isopod burrows were present on slopes adjacent to the
wadi and in low densities in the wadi.
Fig. 1. The suggested features of the trophic control in the Negev Highlands (after Ayal
et al., 2005). The direction of the arrows point on the affected group and their width on
the intensity of the interactions. Secondary predators efficiency is higher in the low
productivity habitat than in the wadi due to plant cover. Hence, in the low productivity
habitat secondary predators efficiency controls primary predators and macro-
detritivores are released from predation and become food limited. In the wadi primary
predators find refuge in the plant cover, maintain high densities and efficiently control
the macrodetritivores.
2.2. Study organism
The desert isopod, Hemilepistus reaumuri, is an annual semi-
social isopod that is common throughout the arid parts of North
Africa and the Middle East (Coenen-Stass, 1984). They live in
monogamous family units in burrows that, at the end of the
growing season, may reach about 50 cm deep. From February to
March the isopods leave their family burrows and the females
search for a new settling site up to several hundred meters from
their original burrow (Shachak and Brand, 1991; Baker, 2004).
Males search for established females, court them, and if accepted
form a couple defending the burrow from any other isopods. The
isopods continue to grow and mature during April. Young hatch in
May and during the first month of their life, stay in the burrow
while their parents provided them with food collected next to the
burrow. In mid-June, the young start to forage independently and
together with the parents continue to dig the burrow all along the
season. The isopods forage for plant litter and soil crust till October
when they cease above ground activity and stay in the family
burrow until next February.
The major predators of the isopod in our study area are different
ectothermic predators, mainly arachnids (Shachak and Brand, 1983;
Baker et al., 1998), centipedes and skinks (Segoli, personal
observations).
2.3. Plant cover manipulations
In the first year (2003), we chose three 750 m2 plots (Fig. 2). We
removed the plant cover (i.e., above ground woody vegetation and
dense herbaceous vegetation) from two of the plots, taking extra
care to minimize damage to the soil crust and detritus. One of the
plots was used for this experiment (Removed). The other plot was
intended to check the effect of artificial woody vegetation and was
not included in this experiment. The third plot was left as a control
Fig. 2. A schematical representation of the experimental design. Intact & Removed
plots are 750 m2, Protected and Unprotected sub-plots are 25 m2. Solid lines represent
fences that reduce movement of predators (see Fig. A1 & A2). Solid circles represent
isopods that were transferred in the first year. Hollow circles represent natural
occurring isopods.
 M. Segoli et al. / Journal of Arid Environments 135 (2016) 33e38 35

(Intact). In between the plots we left a buffer zone that was not
manipulated, at least 50 m between the plots. A barrier fence
(Fig. A1) was installed on both sides between the Removed plot and
the buffer zone that reduced the movements of most of the isopod's
predators (i.e., primary ectothermic predators).
In the second year (2004), woody vegetation was again removed
from the two plots and both were used as the Removed plot. We did
not manipulate the Intact plot.
2.5. Population dynamics
Thirty dry pitfall traps were placed in each plot in Boker in 2003.
Each pit fall trap (10 cm in diameter and 10 cm in depth) was
sampled daily for 2e3 weeks during the July and August between
2003 and 2007, except for 2004 that was sampled during
September and October due to technical problems. Collected iso-
pods were counted and released back into the plot.

2.6. Statistical analysis

Comparisons of survival between different treatments were

2.4. Predation manipulations performed using a survival analysis based on the Kaplan-Meier
estimator (Kaplan and Meier, 1958) with Tarone-Ware log-rank
In order to evaluate the ability of H. reaumuri to survive in the test (Tarone and Ware, 1977). Family size and tethered isopod
wadi when protected from predators, we constructed two types of predation was analyzed by two-way ANOVA (Zar, 1999). Data were
sub-plots of 25 m2 in each of the plots in the wadi. The Predator- analyzed using Systat 13 (SPSS Inc., IL, USA).
protected sub-plots were enclosed by a fence (Fig. A2) and all of
the ectothermic predators were removed from the sub-plot. In the
3. Results
Unprotected plots, a similar fence was erected, and removed after
the settlement stage of the isopods, at the end of March.
3.1. Family survivorship
During February of the first year, dispersing isopods were
collected in the Zin Plain, about 10 km south of the experiment
The number of isopod families (i.e., number of active burrows)
wadi, and 75 isopods were transferred into each of the sub-plots.
that settled and survived in the different treatments is displayed in
Burrows were monitored during March and additional isopods
Table 1.
were supplemented, and by the end of March at least 25 couples
The survivorship curve of the isopod families was different be-
settled in each sub-plot.
tween both years (Fig. 3). In the first year of the experiment, plant
In the second year no fence was constructed during the settle-
cover manipulations did not have a significant effect on burrow
ment stage of the isopods. This enabled the local population of
survivorship in the Predator-protected and Unprotected burrows
isopods (mainly from the surrounding slopes) to migrate into and
(Tarone-Ware log-rank test; c2 ¼ 0.249 df ¼ 1 P ¼ 0.62; c2 ¼ 1.491
settle in the plots (Baker et al., 1998). Local isopods were encour-
df ¼ 1 P ¼ 0.22; respectively) and were pooled together. However,
aged to settle within the subplot by drilling 5 cm deep holes and
Predator-protected burrows had a significantly higher survivorship
adding about 4 ml of water into the hole (Baker et al., 1998). A fence
than Unprotected burrows (Tarone-Ware log-rank test; c2 ¼ 7.90
was built in the Predator-protected sub-plot similarly to the pre-
df ¼ 1 P < 0.01).
vious year, after the settlement stage, and ectothermic predators
In the second year of the experiment, in the Unprotected plots
were caught and removed from the Predator-protected sub-plot.
isopod families had higher survivorship in the Removed plant cover
Activity in the burrows was monitored by brushing away the
treatment than in the Intact plant cover treatment (Tarone-Ware
excrement around the entrance to the burrows for a radius of about
log-rank test; c2 ¼ 5.708 df ¼ 1 P ¼ 0.02). Moreover, there was no
5 cm and monitoring the appearance of new faecal pellets near the
effect of plant cover manipulation on the survivorship of isopod
burrow entrance (Shachak, 1980). This was done about every two
families in the Predator-protected sub-plots (Tarone-Ware log-rank
weeks from March to June and in September in both years.
test; c2 ¼ 0.121 df ¼ 1 P ¼ 0.73). In addition, Predator-protected
At the end of the growing season (SeptembereOctober) we
isopod families had a significantly higher survivorship than Un-
collected all the remaining isopods from the Predator-protected
protected isopod families in the Intact and Removed treatment
and Unprotected sub-plots. In addition, we collected isopods that
(Tarone-Ware log-rank test; c2 ¼ 7.63 df ¼ 1 P < 0.01; c2 ¼ 6.028
settled naturally (i.e., isopods that were not transferred and were
df ¼ 1 P ¼ 0.01; respectively).
not encouraged to settle) from the natural isopod population bur-
rows that were inside the plots but outside of the sub-plot. Settled
isopods were defined as isopods that lived in burrows at the end of 3.2. Family size
March and survived isopods were defined as live isopods in
SeptembereOctober. Family size (number of isopods per burrow) was measured at
We used tethered isopods in order to test the efficiency of the the end of the first year (Fig. 4). Plant manipulation did not have a
Predator-protected sub-plots and to receive a direct measurement significant effect on the family size (two-way ANOVA; F(2,
of predation. During three nights of the new moon phase of July of 13) ¼ 26.104, P ¼ 0.62) while protection from predation had a
the second year we stuck 5 metal pins chosen haphazardly in
Predator-protected and Unprotected sub-plots in the Intact and
Table 1
Removed plots. To each metal pin, we tied a 20 cm piece of sewing The number of settled and surviving isopod families.
thread that was glued to the isopod's pereion. Tethered isopods
Year Plot Sub-plot Settled Survived
were left in the field from about two hours prior to sunset until
about three hours after sunrise. We recorded the number of sur- First Removed Protected 30 7
Unprotected 44 4
viving isopods and the number of isopods that suffered predation,
Intact Protected 31 4
while dead non-preyed isopods were excluded from the analysis. Unprotected 42 3
To measure predation impact we recorded two indices that es- Second Removed Protected 107 44
timate the success: number of active burrows and family size. Unprotected 127 27
Active burrows represent the number of successful families while Intact Protected 35 13
Unprotected 48 4
family size represents the average fitness of each family.
36 M. Segoli et al. / Journal of Arid Environments 135 (2016) 33e38

Fig. 5. Ratio between the density of the isopods at the end of the experiment to the
density at the beginning of the experiment, in Boker. Line represents the minimum
ratio needed for a sustainable population. Data are represented as means ± SE.

Fig. 3. The percent of surviving isopod families (active burrows) for the (a) first and (b)
second year.

Fig. 6. Percent of tethered isopods that were preyed upon in the different manipula-
tions. Data are represented as means ± SE.

Tethered isopods in the Removed treatment had a non significant

trend of lower predation than in the Intact (Fisher's least significant
Fig. 4. Average family size (number of isopods per burrow) at the termination of the
difference; P ¼ 0.15).
experiment in the first year. Natural isopods represent isopods that settled naturally
outside the sub-plots but within the plots. Data are represented as means ± SE.

3.5. Population dynamics

significant effect (one-way ANOVA; F(2, 13) ¼ 3.956, P ¼ 0.04).
Predator-protected burrows had larger families than Unprotected The density of natural occurring isopods in summer was similar
or natural settled isopods outside of the sub-plots that did not differ during the first year (2003) between the different plant manipu-
between them (P ¼ 0.02; P ¼ 0.05; P ¼ 0.64; respectively). lations (Fig. 7). From 2004 there was a linear increase in the ratio

3.3. Replacement rate

Replacement rate (ratio of final to initial isopod population) was

calculated in the first year (Fig. 5). The Unprotected isopods had
survivorship too low to replace the initial number of settled iso-
pods, whereas those protected from predators had a higher number
of surviving individuals at the end of the year compared to the
initial number of settled isopods.

3.4. Direct measurement of predation

Unprotected tethered isopods were preyed upon significantly

more than Predator-protected isopods, which were not preyed Fig. 7. Density (histograms) and ratio (line) of natural occurring isopods between
upon at all (Fig. 6; two-way ANOVA; F(1, 6) ¼ 7.00, P ¼ 0.04). Removed (grey) and Intact (black) treatments.
 M. Segoli et al. / Journal of Arid Environments 135 (2016) 33e38
between the isopod densities in the Removed and Intact treatments
(R2 ¼ 0.97, P < 0.01). The increase in the ratio was due to an increase
in the density of isopods in the Removed treatment compared to a
constant density in the Intact treatment. The lower densities in
2004 are probably due to sampling two months later than the
sampling dates in the other years.
4. Discussion
Ecologists have long debated on the relationship between pro-
ductivity and number of trophic levels, and consequently on the
role of predation in the determination of desert communities
(Fretwell, 1977; Oksanen et al., 1981; Polis, 1991; Ayal, 2007; Holt
and Barfield, 2013). In the present study we tested whether pre-
dation has a role in controlling isopod populations in wadis and if
this predation is mediated by plant cover. Top-down control (i.e.,
predation) as opposed to bottom-up limitation (i.e., resources)
supports the notion that an effective third trophic level exists (Ayal,
2007).
Isopod survivorship in the Predator-protected plots was higher
than in the Unprotected plots (Figs. 1b and 3). In the first year,
removal of plant cover did not affect survivorship of isopod fam-
ilies, as opposed to the second year where removal of plant cover
increased survivorship of isopod families in the Unprotected plots
(Fig. 3). This difference is probably due to a time lag in the effect of
predation of ectothermic predators by endothermic predators, as
suggested by the population dynamics (Fig. 7). Additionally,
removal of plant cover did increased isopod family size and reduced
predation intensity in the Unprotected treatment, but did not affect
the survivorship of Predator-protected isopod burrows (Figs. 1b and
2). The effect of reduced plant cover continued to increase natural
occurring isopod densities for at least four years, while the den-
sities in the Intact were constant (Fig. 7). Replacement rate showed
that the Unprotected isopods had survivorship too low to replace
the initial number of settled isopods, suggesting that the popula-
tion is not sustainable (Fig. 5). However, isopods protected from
predators had a higher number of surviving individuals at the end
of the year compared to the initial number of settled isopods
(Fig. 5). This suggests that the isopod population in the wadi is a
sink population since predation causes an unsustainable isopod
population in the wadi that cannot persist without immigration
from the adjacent slopes (Schreiber and Kelton, 2005). Since
removing the plant cover increases isopods densities, our results
partly support Ayal (2007) hypotheses that the low abundance of
isopods in the wadi is facilitated by predation and that this pre-
dation is mediated by plant cover.
We have shown that some habitats in the desert have an
effective third trophic level. In the Negev highland, the third trophic
level is made of insectivores that feed on the macrodetritivores.
This predation is mediated by shrubs that protect the ectotherm
small predators from their endotherm large predators. The absence
of isopods from the vegetated wadies due to predation by scorpions
is an evidence of predation shaping the desert foodweb. This con-
tradicts Oksanen et al. (1981) hypothesis and suggests that energy
is not the limiting factor in determining the length of food chains
(Holt and Barfield, 2013). Ayal (2007) suggested that the desert food
webs have different properties that enable a high number of tro-
phic levels with low productivity. Deserts have low microbial
decomposition and most of the plant biomass dries up and enters
the above ground food web through consumption by macro-
detritivores (Ayal et al., 2005). Macrodetritivores are donor-
controlled and therefore they have no negative effects on plant
productivity. Most of the macrodetritivores in the desert are small
and similar in size and therefore allow longer food chains than food
webs that are based on larger animals (Hutchinson, 1959; Hairston
37
and Hairston, 1993; Ayal and Groner, 2009). In addition, ecto-
thermic food webs are very efficient physiologically and enable
longer food chains than endothermic food webs (Yodzis, 1984).
This suggests that desert communities are not as simple as
suggested in the past (Oksanen et al., 1981) and the mechanisms
controlling desert communities should be revised (Polis, 1991). In
addition, this study emphasizes the importance of plants in the
desert not only as a food resource but also as a structure mediating
predation intensity and can determine community structure.
Acknowledgements
We thank Sonya Rozin for laboratory assistance, Ittai Renan for
field assistance, and Moshe Shachak for scientific contributions.
This study was supported by grant # 505/02 to YA by the Israel
Science Foundation (ISF) managed by Israel Academy of Sciences.
This is publication No. 889 of the Marco and Louise Mitrani
Department of Desert Ecology.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jaridenv.2016.08.008.
References
Ayal, Y., 2007. Trophic structure and the role of predation in shaping hot desert
communities. J. Arid. Environ. 68, 171e187.
Ayal, Y., Groner, E., 2009. Primary consumer body size and food-chain length in
terrestrial communities. Isr. J. Ecol. Evol. 55, 329e343.
Ayal, Y., Merkl, O., 1994. Spatial and temporal distribution of tenebrionid species
(Coleoptera) in the Negev Highlands. Isr. J. Arid. Environ. 27, 347e361.
Ayal, Y., Polis, G.A., Lubin, Y., Goldberg, D.E., 2005. How can high animal diversity be
supported in low productivity deserts? the role of macrodetritivory and habitat
physiognomy. In: Shachak, M., Gosz, J.R., Pickett, S.T.A., Perevolotsky, A. (Eds.),
Biodiversity in Drylands: toward a Unified Framework. Oxford University Press,
New York, pp. 15e29.
Baker, M.B., 2004. Sex biased state dependence in natal dispersal in desert isoponds,
Hemilepistus reaumuri. J. Insect Behav. 17, 579e598.
Baker, M.B., Shachak, M., Brand, S., 1998. Settling behavior of the desert isopod
Hemilepistus reaumuri in response to variation in soil moisture and other
environmental cues. Israel J. Zool. 44, 345e354.
Bond, W.J., 2005. Large parts of the world are brown or black: a different view on
the ‘Green World’ hypothesis. J. Veg. Sci. 16, 261e266.
Borer, E.T., Seabloom, E.W., Shurin, J.B., Anderson, K.E., Blanchette, C.A., Broitman, B.,
Cooper, S.D., Halpern, B.S., 2005. What determines the strength of a trophic
cascade? Ecology 86, 528e537.
Coenen-Stass, D., 1984. Observations on the distribution of the desert woodlouse
Hemilepistus reaumuri (crustacea, isopoda, oniscoidea) in North Africa. In:
Symposium of the Zoological Society of London, vol. 53, pp. 369e380.
Crawford, C.S., 1981. Biology of Desert Invertebrates. Springer-Verlag, Berlin; New
York.
Fretwell, S.D., 1977. Regulation of plant communities by food-chains exploiting
them. Perspect. Biol. Med. 20, 169e185.
Groner, E., Ayal, Y., 2001. The interaction between bird predation and plant cover in
determining habitat occupancy of darkling beetles. Oikos 93, 22e31.
Hairston, N.G., Hairston, N.G., 1993. Cause-effect relationships in energy-flow, tro-
phic structure, and interspecific interactions. Am. Nat. 142, 379e411.
Hairston, N.G., Smith, F.E., Slobodkin, L.B., 1960. Community structure, population
control, and competition. Am. Nat. 94, 421e425.
Holt, R.D., Barfield, M., 2013. Direct plant-predator interactions as determinants of
food chain dynamics. J. Theor. Biol. 339, 47e57.
Hopcraft, J.G.C., Olff, H., Sinclair, A.R.E., 2010. Herbivores, resources and risks:
alternating regulation along primary environmental gradients in savannas.
Trends Ecol. Evol. 25, 119e128.
Hutchinson, G.E., 1959. Homage to Santa-Rosalia or why are there so many kinds of
animals? Am. Nat. 93, 145e159.
Kaplan, E.L., Meier, P., 1958. Nonparametric-estimation from incomplete observa-
tions. J. Am. Stat. Assoc. 53, 457e481.
Megias, A.G., Sanchez-Pinero, F., Hodar, J.A., 2011. Trophic interactions in an arid
ecosystem: from decomposers to top-predators. J. Arid. Environ. 75, 1333e1341.
Moore, J.C., Berlow, E.L., Coleman, D.C., de Ruiter, P.C., Dong, Q., Hastings, A.,
Johnson, N.C., McCann, K.S., Melville, K., Morin, P.J., Nadelhoffer, K.,
Rosemond, A.D., Post, D.M., Sabo, J.L., Scow, K.M., Vanni, M.J., Wall, D.H., 2004.
Detritus, trophic dynamics and biodiversity. Ecol. Lett. 7, 584e600.
Noy-Meir, I., 1973. Desert ecosystems: environment and producers. Annu. Rev. Ecol.
Syst. 4, 25e51.
38 M. Segoli et al. / Journal of Arid Environments 135 (2016) 33e38
Noy-Meir, I., 1979. Structure and function of desert ecosystems. Isr. J. Bot. 28, 1e19.
Noy-Meir, I., 1985. Desert ecosystem structure and function. In: Evenari, M., Noy-
Meir, I., Goodall, D.W. (Eds.), Ecosystems of the World. Elsevier, Amsterdam,
pp. 93e104.
Oksanen, L., Fretwell, S.D., Arruda, J., Niemela, P., 1981. Exploitation ecosystems in
gradients of primary productivity. Am. Nat. 118, 240e261.
Oksanen, L., Oksanen, T., 2000. The logic and realism of the hypothesis of exploi-
tation ecosystems. Am. Nat. 155, 703e723.
Polis, G.A., 1991. Complex trophic interactions in deserts e an empirical critique of
food-web theory. Am. Nat. 138, 123e155.
Schreiber, S.J., Kelton, M., 2005. Sink habitats can alter ecological outcomes for
competing species. J. Animal Ecol. 74, 995e1004.
Shachak, M., 1980. Energy allocation and life-history strategy of the desert isopod
H. Reaumuri. Oecologia 45, 404e413.
Shachak, M., Brand, S., 1983. The relationship between sit and wait foraging strategy
and dispersal in the desert scorpion, Scorpio maurus palmatus. Oecologia 60,
371e377.
Shachak, M., Brand, S., 1991. Relations among spatiotemporal heterogeneity, pop-
ulation abundance, and variability in a desert. In: Kolasa, J., Pickett, S.T.,
Allen, T.F.H. (Eds.), Ecological Heterogeneity. Springer-Verlag, New York,
pp. 202e223.
Shachak, M., Chapman, E.A., Steinberger, Y., 1976. Feeding, energy-flow and soil
turnover in desert isopod, Hemilepistus reanumuri. Oecologia 24, 57e69.
Shmida, A., 1985. Biogeography of the desert flora. In: Evenary, M.E.A. (Ed.), Hot
Deserts and Arid Shrublands. Elsevier, Amsterdam, pp. 23e77.
Tarone, R.E., Ware, J., 1977. Distribution-free tests for equality of survival distribu-
tions. Biometrika 64, 156e160.
Yodzis, P., 1984. Energy-flow and the vertical structure of real ecosystems. Oecologia
65, 87e88.
Zar, J.H., 1999. Biostatistical Analysis. Prentice Hall, Upper Saddle River.