Journalof Arid Environments (1989) 17, 279-286
The restoration of disturbed arid landscapes with
special reference to mycorrhizal fungi
Edith B. Allen *
Introduction
The accelerated rate of desertification of arid and semi-arid lands worldwide is providing
new incentives for their restoration. Of some 47 million km 2 of these lands, 20% are
classified as severely disturbed (Dregne, 1983) and in need of restoration. Six million ha
additional land are estimated to be desertified yearly (United Nations, 1977).
A landscape-scale approach is essential to dealing with such a large restoration problem.
Different ecosystems in the landscape have different carrying capacities and require
different restoration solutions (Naveh & Lieberman, 1984). Of equal importance are the
interactions between landscape patches (e.g. Risser et al., 1984; Forman & Godron, 1986),
where, for instance, movement of soil or water from one patch to another may be both an
indication of desertification and a deterrent to restoration if both elements are not
considered simultaneously. As an example of the need for a landscape-scale approach to
restoration I will present some research on mycorrhizal fungi in a disturbed semi-arid
landscape.
In keeping with the focus of this workshop, I will consider restoration only of those lands
which are desert or semi-desert. The most viable restoration goal on a large scale is to
return these lands to stable ecosystems which require no further anthropogenic input once
vegetation has been re-established (e.g. as opposed to croplands or pastures), except for
management to avoid future destruction. For the most part the only human use of these
lands is for domestic grazing and shrub harvesting for firewood, or irrigated agriculture on
a smaller scale. Thus the predominant disturbing force is excessive removal of vegetation
followed by erosion (Dregne, 1983). Since continued human use is the major impetus for
restoration, the land use after restoration will be predominantly grazing and wood cutting.
Restoration of natural areas for protection of plant and animal diversity is occurring on a
smaller scale (Janzen, 1988; Jordan et al., 1988), but the two restoration goals need not be
incompatible.
Restoration ecology is an emerging branch of ecology which incorporates applied and
basic research with land management (Allen, 1988a; Gilpin et al., 1988; Jordan et al.,
1988). Much of what we know of restoration is anecdotal. There are projects to restore arid
lands worldwide, but little past emphasis on collecting and analyzing data so that success
of restoration projects can be evaluated and repeated. In the United States, the passage of
the Surface Mine Control and Reclamation Act of 1977 provided an impetus for mine
reclamation in the semi-arid West, and a data base is slowly amassing. Mining is the most
severe of all disturbances, so the concepts and techniques which are devised can be applied
to even the most severe forms of desertification, e.g, where all ofthe plant cover has been
removed and the topsoil has eroded. Not all lands are so severely disturbed, so it is possible
• Systems Ecology Research Group and Department of Biology, San Diego State University, San Diego,
California 92182 U.S.A.
0140-1963/89/050279
+ 08 $03-00/0
© 1989Academic Press Limited
280
E. B. ALLEN
to take advantage of natural succession, coupled with purposeful introductions of
organisms when necessary, to restore these lands (MacMahon, 1983). In spite of these
advances, there are limitations to restoration of arid lands that must still be overcome.
Limitations to restoration of aridlands
While the restoration of any disturbed land presents obstacles to be overcome, there are
special limitations to restoration of arid and semi-arid lands that are driven by the harsh
climate (Table 1). Each of these limiting factors is discussed in terms of its particular
application to arid and semi-arid lands or, where appropriate, to restoration of any lands.
The most important of the limitations to arid land restoration is low and variable
precipitation. When mined land restoration was first proposed for the semi-arid western
United States, a National Academy of Sciences committee held out little hope for
restoration of any lands that received less than about 250 mm precipitation (NAS, 1974).
Such mined lands were termed a 'national sacrifice area' (Box, 1976) because of the
perceived poor prospects for, and high cost of, reclamation. A decade of mandatory
restoration has shown that vegetation can be re-established on these lands at cost, although
the ability to re-establish stability of the restored ecosystem is still an open question (e.g.
Whitford,1988).
The annual variability of precipitation is greater in deserts than in any other biome
(MacMahon, 1980), an observation which has led to the hypothesis of the pulse phenomenon of plant establishment in deserts (Went, 1948; West et al., 1979; Jordan & Nobel,
1981). Once seedlings are established after one or more wet years, the adult plants can
withstand the normal fluctuations in precipitation. While this pulse phenomenon is often
viewed as a limitation, in that restoration will not be successful every year, it is also possible
to take advantage of years of high precipitation to establish vegetation. This may be the
only solution where precipitation is low and sporadic, and irrigation is not possible.
Although wind occurs in many other biomes, it is an especially important factor in
deserts. With the exception of some coastal deserts, low humidity coupled with wind
desiccation is stressful for desert plants (e.g. Smith, 1978), and perhaps more so for
seedlings (Allen & Allen, 1986). Because the surface soils are also dry, wind is a more
forceful erosive factor in deserts than other biomes. Movement of sand dunes is an
outstanding example. Sand dune restoration has received more attention than other arid
lands, not because of any economic benefit but to avoid immediate environmental impacts
(e.g. Gati, 1984).
The loss of topsoil due to wind and water erosion is the most serious consequence of
desertification (Dregne, 1983; Yair, 1987) and at the same time poses the greatest
limitation to restoration. The reversibility of desertification has been questioned (Glantz &
Orlovsky, 1983; Allen, 1988a), primarily because the land is no longer able to support the
Table 1. Limitations to restoration of aridlands
Lowand variable precipitation
Winddesiccation
Erosion and deposition
Lossof topsoil, including organic matter, nutrients, and micro-organisms
Reduction in propagules
Predominance of non-mycotrophic colonizing species
Social, political, and economic barriers
RESTORATION OF DISTURBED ARID LANDSCAPES
281
same level of productivity after topsoil loss. This includes lower soil moisture-holding
capacity because of reduced organic matter, reduced fertility, and loss of propagules of
desired plants and micro-organisms. Restoration is an attempt to reintroduce these
propagules and then to allow succession to take its course. In the case of many plant species
reintroductions are possible, but there are ecological, technological and economic limitations to reintroduction of micro-organisms (e.g. Hayman, 1984; Allen, 1988b).
Since it is not possible at this time to reintroduce most of the beneficial micro-organisms
on a large scale we must rely on natural dispersal. The natural rate of dispersal of
micro-organisms into disturbed lands has been little studied, and is a landscape-scale
problem (Allen, 1987, 1989; Waaland & Allen, 1987).
Among the soil micro-organisms one that has received a great deal of recent attention is
the mycorrhizal fungus. A mycorrhiza is a mutualistic association between a root and
certain species of fungi, whereby the plant has increased growth, nutrient uptake, water
uptake, and drought stress-tolerance, and the fungus receives carbohydrates from the
plant (e.g. Allen & Boosalis, 1983; Harley & Smith, 1983). These fungi are widespread in
arid and semi-arid lands (Trappe, 1981), where the most abundant form is the vesiculararbuscular (VA) mycorrhiza. However, a number of man-caused and natural disturbances
reduce or eliminate the fungus, including drought and erosion (Powell, 1980), tillage
(Allen & Boosalis, 1983), and grazing (Bethlenfalvay & Dakessian, 1984; Wallace, 1987).
Unlike ectomycorrhizal fungi, which can be used to inoculate trees to re-establish forests,
there are no large scale methods for inoculation of VA mycorrhizal fungi.
The loss of mycorrhizae in arid lands is a particular problem because a large proportion
of the initial colonizing species are nonmycotrophic, e.g. do not form a mycorrhizal
association (Allen & Allen, 1989). Although mycorrhizae are important to the growth and
survival oflater seral species, annuals in such families as the Chenopodiaceae, Brassicaceae,
Amaranthaceae, and Zygophyllaceae are non-mycotrophic and appear to be well adapted
to growth in desertified soils. These annuals persisted for 6-15 years in arid and semi-arid
mine soils in Wyoming which had no initial mycorrhizal inoculum (Allen & Allen, 1980,
1988; and unpublished). Even though dispersal of mycorrhizal spores can occur within a
few years on these sites (Waaland & Allen, 1987), the continued presence of nonmycotrophic annuals precludes mycorrhizal establishment. Thus even at the small scale of
a micro-organism the phrase 'desertification begets desertification' may be true. Without
the introduction of propagules of mycorrhizal plant species, and in some cases cultural
manipulation to assist in their establishment, the mycorrhizal component of the ecosystem
cannot re-establish.
The social, political, and economic limitations to arid and semi-arid land restoration
have been dealt with in other publications (e.g. Mabbut, 1984; Tolba, 1987). They will not
be further discussed here, but it is important to keep them in mind while solving ecological
problems. The economic return from arid lands is low, so many governments are not
willing to invest in their restoration. However, ecological solutions may be economically
viable solutions, even ifthey are not as rapid as the more common agronomic practices (e.g.
seed drilling, irrigation, fertilization). An ecological approach might include taking
advantage of the pulse phenomenon of seedling establishment and natural succession. It
might also involve natural dispersal of mycorrhizal fungi combined with an understanding
of their effects on various planted species and in different environments. The next section
of this paper is a report on an ecological, landscape scale approach to restoration which
incorporates mycorrhizal fungi.
VA mycorrhizal fungi and succession in a disturbed landscape
Restoration research was initiated in 1982 on a recontoured coal mine in SW Wyoming.
The undisturbed native area is an Artemisia steppe in hilly terrain at 2000-2200 m elevation
which receives 230 mm precipitation, 60% as snow. The landscape is heterogeneous,
E.B.ALLEN
282
varying from patches of taller woody vegetation on east facing convex slopes to 'cushion'
plants on wind-scoured ridge tops. Mining tends to homogenize the landscape, creating
slopes of equivalent 1: 5 pitch, equal depths of replaced topsoil, and similar mixtures of
plant species that are seeded everywhere. The homogeneity of the mine soil nutrients and
micro-organisms, compared to the heterogeneous native soils, has been characterized
(Allen & MacMahon, 1985).
The remaining elements of heterogeneity on the mined site are due to microclimatic
effects such as ridge tops vs. slopes and replacement of sub-soils rather than topsoils in
some areas. The sub-soils are lower in nutrients, have virtually no organic matter, and are
higher in clay than the topsoils (Waaland & Allen, 1987). Because the topsoils were stored
for seven years prior to their replacement, neither they nor the sub-soils had enough viable
mycorrhizal spores to cause greater than 1% of the root length of planted grasses and
shrubs to become infected in the first growing season. The plants in undisturbed areas may
have up to 75% infection (Waaland & Allen, 1987).
Because mycorrhizae improved the growth of several of the species which were planted,
and increased their ability to compete with colonizing non-mycotrophic annuals (Allen &
Allen, 1984), we hypothesized that the addition of mycorrhizal inoculum should increase
the rate of mycotrophic plant establishment and succession on this site. In order to test the
hypothesis, we inoculated five different sites on the mine with a thin layer (1-2 cm) of
topsoil from the native area containing a high density of mycorrhizal spores, and removed
non-mycotrophic annuals by hand in a 2 x 2 factorial experiment (with and without
mycorrhizae, with and without annuals). Further details of the experimental design are
published elsewhere (Allen & Allen, 1986; 1988). The five sites included two on stockpiled
topsoil that were located midslope, and three on sub-soil that were either on the
wind-scoured upper portion of the slope, on the lower portion of the slope, or midslope.
Three of the sites were initiated (seeded and inoculated) in 1982, two in 1984 (Fig. 1).
The data after three years of growth are presented here to compare results for sites
established in different years (Fig. 1). Only the results for the planted Agropyron species
(A. dasystachyum, A. smithii, A. trachycaulum) are reported, as they were by far the most
abundant species. Seeded shrubs (Artemisia tridentata and Atriplex canecsens) established
sparsely, with only 1% cover on any of the sites. Colonizing annuals persisted into the third
year in only two of the sites, the subsoil upper where the non-mycotrophic species Salsola
Subsoil
Upper
1982
8
c
if!.
e
•
0
Low ...
Q)
>
'5
4i
a:: ;:'iii
~
Topsoil
Lower
1982
Midslope
1984
1982
I :1 ~1 J :1
• H.hA
~
Midslope
1984
Hight:.
a
Low:1
...
t:.
a
:1
0
t:.
0
~I
;r
0
r:P
:1
Figure 1. Third year summary of data from four treatment effectson % cover and density of planted
Agropyron species. Sites were initiated in different years (1982, 1984), on topsoil vs. sub-soil, and on
different heights on slopes. For comparison, the relative ranks of the treatments are presented. Total
percent cover and density were always greater on topsoil than sub-soil sites. Actual data are
published elsewhere (Allen & Allen, 1986, 1988).0, mycorrhizal, with annuals;., mycorrhizal, no
annuals; !:::., non-mycorrhizal, with annuals; A, non-mycorrhizal, no annuals.
RESTORAnON OF DISTURBED ARID LANDSCAPES
283
Table 2. Summary of ranked data for the effects of mycorrhizal inoculation and
annualweed removal treatments on % coverand density ofplanted Agropyron species
(see Fig. I). M = mycorrhizal inoculum added, NM = no mycorrhizal inoculum
added, A = with annual weeds, NA = no annual weeds. X = numberof times a
treatment had highest or lowestvalues, as indicated. Data arefrom the thirdgrowing
season at eachsite
Treatments with
highest % cover
A NMNA2X
Subsoil midslope
Subsoil lower
t::" NMA 2X
Subsoil upper
Topsoil 1982
lowest % cover
A NMNA IX
Subsoil upper
OMA4X
highest density
A NMNA IX
Subsoil midslope
t::" NMA 3X
Subsoil upper
Subsoillower
Topsoil 1982
• MNA IX
Topsoil 1984
lowest density
A NMNA 2X
Subsoil upper
Topsoil 1982
OMA2X
Subsoil midslope
Topsoil 1984
• MNA IX
Subsoil lower
• MNA IX
Topsoil 1984
kali was still abundant (Allen & Allen, 1988) and the 1984 topsoil site where Bromus
rectorum still formed 15%of the ground cover (unpublished data).
The most notable result from these five sites is that no two sites had the same outcome
with respect to Agropyron density and cover (Fig. 1). The order of abundance of the grasses
in the four treatments was different on each site. The predictions that mycorrhizal
inoculation would increase growth of the planted grasses, and that removal of nonmycotrophic annuals would improve grass establishment, were in general not borne out
(Table 2). If mycorrhizae and annual plant removal had been the most important factor in
causing increased grass establishment, the MNA (mycorrhizal, no annuals) treatment
would have had the greatest grass abundance on all sites. However, this treatment had the
greatest benefit on only one site, the topsoil site established in 1984. In fact, in some cases
a treatment which caused the greatest percent cover and density on one site caused the
lowest cover and density on another site (see, for instance, the non-mycorrhizal, no annual
treatment-NMNA-and the mycorrhizal, no annual treatment-MNA-in Table 2).
Sorting out the causes of these different results requires an examination of individual
species interactions and characteristics of the sites, or in other words, a reductionistic
coupled with a landscape-scale approach. The fact that mycorrhizal inoculum did not
increase grass growth to all of the field treatments, and in fact decreased their growth in
some cases, was contrary to our preliminary greenhouse experiments (Allen & Allen,
1984). The relatively high nutrient soils coupled with high precipitation may in part be
responsible, as mycorrhizae may act as a carbon drain under these conditions. In spite of
their reduced growth, the grasses had other physiological benefits, including decreased
stomatal resistance to water vapor, decreased leaf mortality, and delayed phenology (Allen
& Allen, 1986).
One very surprising result was the number of times the treatments with annuals had the
greatest grass cover and density (see NMA, Table 2). This implies that the annuals were
facilitating grass establishment, rather than competing with the grasses. In fact, removal
of Salsola kali did cause a reduction in grass establishment, but only on the harshest sites
where the soil was poor, plant establishment was low, and wind was important (Allen &
Allen, 1988). The removal of S. kali reduced the amount of snow trapped, especially on
the upper site, and reduced moisture input. In addition to increasing moisture, the
284
E.B.ALLEN
standing litter of S. kali, which persisted during the spring germination period, enhanced
seedling establishment (Allen & Allen, 1988).
Another surprising result was that inoculation with mycorrhizal fungi directly reduced
the growth and density of S. kali and Atriplex rosea, both non-mycotrophic species (Allen,
1984; Allen & Allen, 1984, 1988). This apparently 'pathogenic' effect of mycorrhizal fungi
on non-mycotrophic species may help explain why non-mycotrophic species are so
abundant in severely disturbed sites where inoculum density is low, and at the same time
inabundant in soils of high inoculum (Pendleton & Smith, 1983; Allen & Allen, 1988). The
loss of the S. kali 'nurse crop', due to either mycorrhizal inoculation or experimental
removal, reduced grass establishment. However, S. kali facilitated grass establishment on
only some of the five sites (two sites if grass cover is considered, three if grass density is
considered). The important point here is that the harshness of the site, which was
determined by its position in the landscape and whether topsoil was used, determined
whether the annuals acted as facilitators or as competitors. A high initial density of
mycorrhizal inoculum may either hinder or promote the early stages of succession, and an
understanding of the heterogeneity of the landscape is critical before making decisions
such as whether to inoculate, or whether to rely on natural dispersal of the fungus.
The natural dispersal of spores of VA mycorrhizal fungi in mined sites appears to be
great enough to begin detecting infection in the roots within one to three years (e.g.
Waaland & Allen, 1987). These sites are characterized by small patches of disturbance,
usually no larger than 100 ha, with nearby undisturbed native areas which serve as sources
of spores. Spores from these native areas are moved primarily by wind (via soil erosion) and
animals which ingest the spores, although wind is most important in arid regions (Warner
et al., 1987). In a much larger disturbance, the Mt. St Helens volcano in the state of
Washington, U.S.A., plant species which had established c. 5 km from the undisturbed
forest edge still had no VA mycorrhizal inoculum after 7 years (Allen, 1987; and
unpublished). The scale of desertification, which includes the erosion of topsoil and loss of
inoculum, is larger still. Whether islands of vegetation which are re-established in a sea of
desertified soils will become naturally inoculated must still be examined.
In a lO-year anniversary review on the progress since the 1977 United Nations
Conference on Desertification, the authors agreed that the lack of political, social, and
economic support are the greatest limitations to arid land restoration (see Desertification
Control Bulletin No. 15, 1987). In keeping with the theme of the workshop, 'What is
Special about Desert Ecology?' I assert that destruction and the need for restoration are not
unique to deserts. However, more productive lands can be economically restored while
deserts continue to lie barren. Continued scientific advancement in landscape-scale
approaches to restoration, as exemplified in the discussion above, may eventually lead to
more rational restoration plans which are compatible with the low economic returns from
arid lands.
I thank Michael F. Allen for reviewing the manuscript, and the Women and Gender Research
Institute, Utah StateUniversity, for partialtravelsupport. This synthesis wassupportedby USDA
grant nos. 83-CRCR-I-1229, 83-CRSR-2-2274, and 85-CRSR-2-2719 through the Ecology Center,
Utah State University.
References
Allen, E. B. (1984). VAmycorrhizae and colonizing annuals:implications forgrowth, competition,
and succession. In: Williams, S. E. & Allen,M. F. (Eds), VA Mycorrhizae andReclamation ofArid
andSemiarid Lands. pp. 41-51. University ofWyoming Agricultural ExperimentStationScience
Report No. SA1261. 92 pp.
Allen, E. B. (Ed) (l988a). The Reconstruction of Disturbed Arid Ecosystems. Boulder, Colorado:
Westview Press.
Allen, E. B. & Allen, M. F. (1980). Natural re-establishment of vesicular-arbuscular mycorrhizae
following stripminereclamation in Wyoming. Journalof AppliedEcology, 17: 139-148.
RESTORATION OF DISTURBED ARID LANDSCAPES
285
Allen, E. B. & Allen, M. F. 1984). Competition between plants of different successional stages:
mycorrhizae as regulators. Canadian Journalof Botany, 62: 2625-2629.
Allen, E. B. & Allen, M. R. (1986). Water relations of xeric grasses in the field: interactions of
mycorrhizas and competition. New Phytologist, 104: 559-571.
Allen, E. B. & Allen, M. R. (1988). Facilitation of succession by the nonmycotrophic colonizer
Salsola kali (Chenopodiaceae) on a harsh site: effects of mycorrhizal fungi. American Journal of
Botany, 75: 257-266.
Allen, E. B. & Allen, M. F. (1989). The mediation of competition by mycorrhizal fungi in
successional and patchy environments. In: Grace, J. & Tilman, D. (Eds), Perspectives in Plant
Competition. New York: Academic Press. (in press).
Allen, M. F. (1987). Re-establishment of mycorrhizas on Mt. St. Helens: migration vectors.
Transections of theBritishMycological Society, 88: 413-417.
Allen, M. F. (1988b). Reestablishment of VA mycorrhizae following severe disturbance: comparative patch dynamics of a shrub desert and a subalpine volcano. Proceedings of theRoyal Society of
Edinburgh. (Section B) 94: 63-71.
Allen, M. F. & Boosalis, M. G. (1983). Effects oftwo species of VA mycorrhizal fungi on drought
tolerance of winter wheat. New Phytologist, 93: 67-76.
Allen, M. F. & MacMahon, J. A. (1985). Impact of disturbance on cold desert fungi: comparative
microscale dispersion patterns. Pedobiologia, 28: 215-224.
Bethlenfalvay, G. J. & Dakessian, S. (1984). Grazing effect on mycorrhizal colonization and floristic
composition of the vegetation on a semiarid range in northern Nevada. Journal of Range
Management, 37: 312-316.
Box, T. W. (1976). The significance and responsibility of rehabilitating drastically disturbed land.
In: Schaller, F. W. & Sutton, P. (Eds), pp. 1-10. Reclamation of Drastically Disturbed Lands.
Madison, Wisconsin: American Society of Agronomy, Crop Science Society of America, Soil
Science Society of America. 742 pp.
Dregne, H. E. (1983). Desertification of Arid Lands. Advances in Desert and Arid Land Technology.
Vol. 3. Chur, Switzerland: Harwood Academic. 242 pp.
Forman, R. T. T. & Godron, M. (1986). Landscape Ecology. New York: Wiley. 619 pp.
Gati, F. (1984). Controlling sand dune encroachment in Iraq. Desertification Control Bulletin, 11:
22-26.
Gilpin, M., Aber, J. & Jordan, W. (1988). Restoration Ecology. Cambridge: Cambridge University
Press. 342 pp.
Glantz, M. H. & Orlovsky, N. (1983). Desertification: a review of the concept. Desertification
Control, 9: 15-22.
Hayman, D. S. (1984). Methods for evaluating and manipulating vesicular-arbuscular mycorrhizas.
In: Grainger, J. M. & Lynch, J. M. (Eds). pp. 95-117. Microbiological Methods forEnvironmental
Biotechnology. New York: Academic Press.
Harley, J. L. & Smith, S. E. (1983). Mycorrhizal Symbiosis. New York: Academic Press. 483 pp.
Janzen, D. H. (1988). Tropical ecological and biocultural restoration. Science, 239: 243-244.
Jordan, P. W. & Nobel, P. S. (1981). Seedling establishment of Ferocactus acanthodes in relation to
drought. Ecology, 62: 901-906.
Jordan, W. R., Peters, R. L. & Allen, E. B. (1988). Ecological restoration as a strategy for
conserving biological diversity. Environmental Management, 12: 55-72.
Mabbutt, .J. A. (1984). A new global assessment of the status and trends of desertification.
Environmental Conservation, 11: 103-118.
MacMahon, J. A. (1980). Ecosystems overtime: succession and othertypes of change. Pages 27-57.
In: Waring, R. H. (Ed.), Forests: Fresh Perspectives from Ecosystem Analysis. Proceedings of the
40th Annual Biology Colloquium (1979), Oregon State University Press, Corvallis, Oregon.
MacMahon, J. A. (1983). Nothing succeeds like succession: ecology and the human lot. 67th Faculty
Honor Lecture. Logan, Utah. Utah State University Press. 32 pp.
National Academy of Sciences. (1974). Rehabilitation of Western Coal Lands. Cambridge, Massachusetts: J. B. Lippencott. 198 pp.
Naveh, Z. & Lieberman, A. S. (1984). Landscape Ecology. New York: Springer-Verlag. 356 pp.
Pendleton, R. L. & Smith, B. N. (1983). Vesicular-arbuscular mycorrhizae of weedy and colonizer
plant species at disturbed sites in Utah. Oecologia, 59: 296-301.
Powell, C. L. (1980). Mycorrhizal infectivity of eroded soils. Soil Biology and Biochemistry, 12:
247-250.
286
E. B. ALLEN
Risser, P. G., Karr, J. R. & Forman, R. T. T. 1984. Landscape Ecology. Directions and
Approaches. Illinois NaturalHistory SurveySpecialPublication Number 2. 1800 pp.
Smith, W. K. (1978). Temperatures of desert plants: another perspective on the adaptability of leaf
size. Science, 20: 614-616.
Tolba, M. K. (1987). Ten years after the United Nations Conference on Desertification (UNCOD).
Land Use Policy, 4: 363-370.
Trappe, J. M. (1981). Mycorrhizaeand productivity of arid and semi-arid rangelands. In: Manassah
J. T. & Briskey, E. J. (Eds), Advances in FoodProducing Systems for Arid andSemi-AridLands.
pp. 581-599. New York: Academic Press. 753 pp.
United Nations. (1977). Desertification: Its Causes and Consequences. Secretariat of the U.N.
Conference on Desertification, Nairobi (Ed.), Oxford: Pergamon Press. pp. 448.
Waaland, M. E. & Allen, E. B. (1987). Relationships between VA mycorrhizal fungi and plant cover
following surface mining in Wyoming. JournalofRangeManagement, 40: 271-276.
Wallace, L. L. (1987). Mycorrhizae in grasslands: interactions of ungulates, fungi and drought . New
Phytologist, 105: 619--632.
Warner, N. J., Allen, M. F. & MacMahon, J. A. (1987). Dispersal agents of VA mycorrhizal fungi
in a disturbed arid ecosystem. Mycologia, 79: 721-730.
Went, F. W. (1948). Ecology of desert plants. I. Observations on germination in the Joshua Tree
National Monument, California Ecology, 29: 242-253.
West, N. E., Rhea, K. H. & Harniss, R. O. (1979). Plant demographic studies in sagebrush grass
communities of southeastern Idaho. Ecology, 60: 376-388.
Whitford, W. G. (1988). Decomposition and nutrient cycling in disturbed arid ecosystems. In:
Allen, E. B. (Ed.), The Reconstruction of Disturbed Arid Ecosystems. pp. 136-161. Boulder,
Colorado: Westview Press. 268 pp.
Yair, A. (1987). Environmental effects ofloess penetration into the northern Negev Desert.Journal
of Arid Environments, 13: 9-24.