Aust. J . Bot., 1996,44,581-599

Distribution and Floristics of Terricolous Lichens in Soil Crusts in Arid and Semi-arid New South Wales, Australia

D. J. Eldridge

Department of Land and Water Conservation, Graduate School of the Environment, Macquarie University, NSW 2109, Australia.

Abstract

This paper examines the distribution of terricolous lichens at a regional scale across seven landscape types over 60 000 km2 in western New South Wales. Data are also presented on the distribution of lichens within a geomorphic sequence of runoff and runon zones on a red earth soil near Cobar. On a regional scale, 48 taxa from 23 genera were collected from 282 sites in semi-arid and arid eastern Australia. Of these, 74% were crustose or squamulose, and the remainder (26%) were foliose. Six genera (Acarospora, Endocarpon, Catapyrenium, Diploschistes, Peltula and Xanthoparmelia) accounted for 57% of species. Landscape type was a poor predictor of lichen floristics or crust cover. Instead, a core group of species comprising Collema coccophorum, Heppia despreauxii, Endocarpon rogersii, E. simplicatum var. bisporum, E. pallidurn, Psora decipiens, Peltula patellata ssp. australiensis, Catapyrenzum squamulosum and Synalissa symphorea, occurred in all landscape types. Plains wlth red earths had the greatest mean number of species per site (1 1.2) and the greatest mean crust cover (2T7%). Plains of calcareous earths yielded the greatest number of species (38). Across all sites, crust cover was a poor predictor of lichen species richness. However, on landscape types with non-calcareous soils, mean crust cover explained 88% of the variation in mean number of species. Whilst there was no difference in total number of species across a sequence of geomorphic zones, crust cover was significantly greater in the interception zones (79.0%) compared with either the run-on (6.6%) or run-off (24.0%) zones. These distributional data are compared with other published and unpublished studies from similar areas in Australia. The role of terricolous crusts as indicators of ecosystem health, and the influence of land management on crust cover and subsequent landscape stability are discussed.

Introduction Terricolous lichens, particularly squamulose and crustose types, form intimate associations

with other soil-inhabiting biota such as fungi, bryophytes, algae and cyanobacteria to form biological soil crusts (also known as microphytic, microbiotic or cryptogamic crusts; West 1990). They occur over extensive areas of arid and semi-arid rangeland in both the Northern and Southern Hemispheres, in areas which are not extensively sandy, stony or cultivated.

Biological soil crusts have a major impact on physical and ecological processes such as nitrogen fixation, infiltration, erosion, and recruitment and survival of vascular plants. Their resistance to heat and drought enables them to occupy habitats that are unsuitable for flowering plants (Rogers 1977). Lichen photobionts such as the cyanobacterium Nostoc, fix atmospheric nitrogen, thereby increasing the soil nitrogen pool. Some evidence suggests that this nitrogen is made available to developing vascular plant seedlings (Harper and Pendleton 1993). The mutual pressure between lichen thalli and the accumulation of eroded material around the lichens often leads to increases in surface microtopography. This microtopography or surface roughness is often sufficient to capture and retain rainfall. These higher levels of soil moisture and nutrients may eventually lead to the development of small- scale landscape patterning and favourable sites for germination and establishment of vascular plants (Harper and St Clair 1985). Like moss patches, lichen crusts harbour and provide nutrition for soil invertebrates (Gerson and Seaward 1977). Lichens also increase the resistance of soil surfaces to wind and water erosion (Williams et al. 1995; Eldridge and

D. J. Eldridge

Greene 1994~). Fungal hyphae use mucilaginous sheaths and gels to bind microaggregates into macroaggregates (> 0.25 mm diameter) which are moderately stable under the action of raindrops (Greene and Tongway 1989). They also provide a physical barrier on the soil surface which reduces the kinetic energy of falling raindrops and thus their erosive potential.

The only systematic examination of lichens in soil crusts in Australia was undertaken by Rogers in the early 1970s, with studies of the distribution of terricolous lichens in arid and semi-arid landscapes in eastern Australia (Rogers 1970, 1972a, 1972b, 1974). Apart from Rogers' work , studies at a local level have focused on lichens in New South Wales at Lake Mere (Tozer and Eldridge, unpub. data), Mungo National Park (Eldridge 1995; Eldridge and Kinnell 19971, and Cobar and Nymagee (Hynson and Wells, unpub. data) in New South Wales. Lists of soil lichens have been included in vegetation surveys from the mid-north region of South Australia (Hyde 1994), the Nullarbor Plain of Western Australia (Johnson and Baird 1970), and from Wyperfeld National Park in Victoria (Scott 1982). Regional inventories of soil lichens are few, and include species lists for arid Northern Territory (Sammy, unpub, data) and for the Recherche Archipelago in Western Australia (Willis 1951).

Land managers have been slow to embrace the concept that soil crusts are manageable components of the vegetation community, despite the pioneering work by Rogers almost 25 years ago. During the past decade however, largely through the efforts of soil scientists and soil ecologists, there has been increased recognition of the fact that soil crusts and their communities are desirable components of healthy ecosystems (Tongway 1994). Australian research during the past 25 years has focused on the role of crusts in run-off and erosion processes, particularly in the semi-arid woodlands (Eldridge and Greene 1994b). Despite the renewed interest in soil crusts, little is known about the distribution, floristics and ecology of the component lichens, bryophytes, cyanobacteria and fungi.

This paper reports on a survey of the distribution of terricolous lichens in arid and semi- arid New South Wales (NSW), Australia. The aim of this study was to document the extent and floristics of lichens in soil crusts, and their distribution within the main landscape types in western NSW. The work forms part of a larger study investigating the usefulness of lichens and bryophytes as indicators of rangeland health.

Materials and Methods Study Area

The survey area covered approximately 600000 km2 of New South Wales, Australia. It is bounded in the north by the Queensland border (29"00fS), in the west by the South Australian border (141°00'E), in the south by the Murray River and in the east by longitude 147"301E. Grazing of sheep and cattle on native pastures is the principal land-use in the area, but some areas of permanent and opportunistic cropping are carried out along the eastern margins and in the south-west. Much of the survey lies within the Western Division, an administrative division where land tenure is predominantly leasehold, and ownership is vested in the Crown. Under the conditions of the leases, restrictions are placed on certain activities such as clearing and cropping.

Climate

Rainfall in the survey area is spatially and temporally variable, decreasing from 400 mm in the north-east to less than 150 mm in the far north-west (Fig. 1). Rainfall is predominantly summer- dominant in the north-east and winter rainfall increases with distance to the south-west (Fig. 1). Droughts are a regular feature of the climate. Diurnal temperatures are typically hot in summer (> 40") and mild in winter (> lo0), and evaporation increases from south to north and from east to west (Cunningham et al. 1981).

Geomorphology

The majority of the survey area lies within the Murray-Darling Basin which is generally a depositional landscape. The Basin is overlain by a mosaic of Quaternary aeolian deposits, man): of

Lichens in Soil Crusts

Q U E E N S L A N D

krn 50 0 50 100. 150 krn

MIN

Fig. 1.. Map of the survey area showing the major towns, rainfall isohyets and the distribution of summer and winter rainfall.

which are calcareous. The most extensive depositional landscape comprises level to gently undulating plains of Quaternary aeolian sediments and localised colluvial materials (Pickard and Norris 1994), on which are superimposed broad sand sheets and linear and subparabolic dunes. Plains of Quaternary alluvium are found on the Riverine Plain in the south-east, and along the floodplains of the major rivers, particularly the Darling, Murrumbidgee and Murray Rivers which drain most of the area. Small areas of erosional landscape including ranges, hills and mesas with rocks of Precambrian and Tertiary age occur in the north of the survey area near White Cliffs, Cobar and Broken Hill.

Landscape Types

The landform, soils and vegetation in the survey area can be classified into seven broad landscape types (Fig. 2). Nomenclature of flowering plants follows Harden (1990-1993).

D. J. Eldridge

Q U E E N S L A N D

V I C T O R I A I

Fig. 2. Distribution of the seven landscape types in the survey area. 1. Ranges and hills; 2. Footslopes and rolling downs; 3. Plains of earthy sands and dunes of neutral red earths; 4. Sandplains and dunefields of calcareous earths; 5. Plains, low ridges and flats of loamy red earths; 6. Relict floodplains, playas and drainage lines; 7. Active floodplains, drainage lines and depressions.

( I ) Ranges and hills of shallow loams

These are scattered throughout the survey area and support sparse mulga (Acacia aneura), and a variable understory of sparse grasses and forbs, and occasionally, perennial shrubs of the family Chenopodiaceae.

Lichens in Soil Crusts

( 2 ) Footslopes and rolling downs of loamy duplex soils

These occur on the margins of the erosional landscapes near White Cliffs, Tibooburra and Broken Hill. The soils, which are often overlain by a mantle of quartz or silcrete gravel, are generally resistant to erosion, but support only sparse grasses. The intervening clay depressions support a shrubland dominated by saltbushes (Atriplex spp.) and bluebushes (Maireana spp.).

(3) Extensive plains of earthy sands and dunes of deep neutral red earths

These landscapes occur in the north-west of the survey area, between Bourke and Tibooburra. The soils are typically sparsely vegetated, and moderately susceptible to wind erosion. They support scattered mulga, moderately dense shrubs in the genera Eremophila, Dodonaea and Senna, and sparse pastures.

(4) Sandplains and dunefields of calcareous earths

This is the predominant landscape in the south-west of New South Wales. It supports a woodland dominated by belah-rosewood (Casuarina cristata-Alectryon oleifolium) and tall shrublands dominated by mallees (Eucalyptus spp.). When pasture cover levels are low these soils are subject to wind and water erosion.

(5) Plains, low ridges andflats of loamy red earths

This landscape type occurs over extensive areas in the east of the survey area between Bourke and Griffith. The woodland vegetation is dominated by bimble box (Eucalyptus populnea), red box (E. intertexta), white cypress pine (Callitris glaucophylla) and mulga, and typically supports a dense understory of perennial shrubs (Dodonaea, Eremophila and Senna spp.) and sparse grasses. These landscapes are often severely degraded through woody shrub encroachment, water sheeting and gullying.

(6) Relictfloodplains, playas and drainage lines

Occurring predominantly on the Riverine Plain, this landscape comprises a mosaic of red and brown duplex soils and grey clays, and supports saltbushes, bluebushes and varied grasses. The duplex soils are susceptible to scalding (Beadle 1948) once the overlying coarser textured topsoil has been removed through erosion or overgrazing.

(7) Activefloodplains, drainage lines and depressions

This landscape supports an open woodland-grassland dominated by river red gum (E. camaldulensis), blackbox (E. largiflorens), coolabah (E. microtheca) and gidgee (Acacia cambagei). It is dominated by relatively stable grey and brown cracking clays, and drains much of the Murray-Darling Basin from the north-east.

Methodology Major roads and tracks, selected to form a regular grid of sites, were traversed in order to

comprehensively survey the area. The sites were selected at intervals of 10-30 km along these routes. Some areas of special interest such as National Parks, long-term grazing exclosures, railway corridors and regeneration areas were included. At each of 282 sites, transects were established perpendicular to, and commencing 100-200 m away from, the edge of the road. Along each transect, 10 x 0.50 m2 quadrats were laid down to form the basis for data and soil crust collection.

Field data were collected from within each quadrat in order to determine relationships between the vascular plants, the soil surface, and soil lichens in the crusts. Quadrat-based measurements of soil surface condition, microtopography, landscape element type, slope, and type and degree of erosion were also made. Concurrently, observations of the cover and relative composition of lichens, bryophytes and cyanobacteria were made from within the same quadrats, along with observations on cover and type of vascular plants, and cover and type of litter. Separate samples of the soil crust were collected from up to four quadrats within each transect, and transported to the laboratory for identification of lichens and bryophytes present in the crusts. Sufficient samples were collected to provide voucher specimens. Additionally, bryophytes and lichens present at the site but absent from the transects were collected separately to ensure that all species at a given site were collected. Only data on cover and floristics of terricolous lichens are discussed in this paper.

D. J. Eldridge

Soil samples were sieved in the laboratory using a 2 mm sieve, and lichens were identified using keys published by Filson (1988, 1992), Filson and Rogers (1979) and McCarthy (1991a), as well as more recent generic revisions. Considerable help was obtained from Australian and international lichenologists. Lichen nomenclature conforms to that of McCarthy (1991b) or more recent monographs. Voucher specimens are lodged with the National Herbarium of New South Wales (NSW).

Identification of Species

In the survey area, some lichens, particularly crustose forms, were infertile and therefore could not be identified. Particular difficulty was experienced in differentiating sterile Lecidea spp. from Diploschistes spp. Specimens whose determinations were in doubt have been omitted from this paper. Two morphological groups of Collema coccophorum have been identified, based on thallus shape. Group A is characterised by a thallus, which has few lobes but abundant cylindrical isidia. Group B has a rosette form with many distinct lobes, which may be erect or prostrate.

Results and Discussion Distribution of Species and Growth Forms

Forty-eight terricolous lichens, representing 23 genera from 14 families, were found in the survey area (Table 1). Lichens were found at 236 of the 282 sites, and were associated with all soil and vegetation types. Six genera (Xanthoparmelia, Endocarpon, Catapyrenium, Acarospora, Peltula and Diploschistes) accounted for 28 of the 4 8 species. The most common lichens, in order of occurrence, were Collema coccophorum morphological group B (184 sites), Heppia despreauxii (147), Endocarpon pusillum (134), Collema coccophorum morph. group A (107) and Endocarpon rogersii (95; Table 1). Sixty-three percent of sites contained five lichen species or less, and 7% contained only a single species (Table 2).

The soil crust community was characterised by squamulose (43%) and crustose (31%) growth forms. Foliose growth forms comprised the remainder (26%, Table 1). Although the fruticose lichen Cladonia sp. was collected at 37 sites in the study area (Table I), it was only found as primary thallus squamules. Consequently, consistent with other studies (Rogers 1970; Rogers 1974), it is recorded as squamulose in this study. Although no other typical f ru t icose l ichens were col lected in this s tudy, terr icolous Cladia aggregata and C. corallaizon have been collected from within the study area at Nymagee near Cobar (Wells and Hynson, unpub. data), and from Cocoparra Nature Reserve north of Griffith (Curnow and Lepp, unpub. data). Data on the proportion of squamulose to crustose growth forms (74%) are similar to those reported for other arid and semi-arid regions (57-97%; see Rogers 1974). Apart from the vagrant Chondropsis semiviridis, all species were attached to soil. Only a few of the Xanthoparmelia spp. found in the survey area also were attached to rocks.

Factors Affecting Lichen Distribution

Regional- and landscape-scale distribution

Both mean annual rainfall and diurnal summer temperatures are known to influence the regional distribution of soil crust communities in eastern Australia (Rogers 1972b). High summer temperatures suppress photosynthesis when lichen thalli are hydrated (Rogers 1971). Consequently, crusts are poorly developed in areas where summer rainfall is dominant. Rogers ( 1 9 7 2 ~ ) maintains that either mean annual or seasonal distribution of rainfall can be invoked to adequately explain the distribution of calciphiles, due to the clear relationships between rainfall characteristics, soil pH and presence of free calcium carbonate in the soil.

As lichens are strongly influenced by changes in substrate characteristics (Galun 1963; Rogers and Lange 1971; Downing and Selkirk 1993), distributional patterns are likely to mirror changes in soil and landscape properties such as soil texture, soil depth, water-holding capacity and slope. Fulgensia subbracteata is gypsophilic, and specimens found in arid and

Lichens in Soil Crusts

semi-arid eastern Australia were generally restricted to areas where gypsum crystals occurred on the soil surface. Similar observations are reported by St Clair et al. (1993) for the same species in the semi-arid Intermountain ~ e ~ i o ~ of Western USA. Similarly, in the present study, Diploschistes hensseniae, D, muscorum ssp. bartlettii, D, ocellatus, Neofuscelia pulla and Xanthoparmelia subdistorta were restricted exclusively to calcareous soils. Other species such as the calciphiles Aspicilia calcarea, Eremastrella crystallifera and Toninia sedifolia were more common and occupied a greater surface area on high pH (> 8.5) soils.

Ordination of the presence or absence data of species at the 282 sites of the present study indicated that landscape type was a poor discriminator of lichen floristics. The lowest number of species were found on active floodplains (FLACT; Table 3). From the active floodplains, the total number of species increased through footslopes (FTSLO), ranges (RNGHL) and sandplains (PLSND), relict floodplains (PLREL), to plains with red earths (PLRED) and calcareous (PLCAL) plains (Table 3). Although the ordination did not classify suites of lichens unique to each of the landscape types, clear trends were nevertheless apparent in each of the seven landscape types.

(1) Active floodplains, drainage lines and depressions

Lichens occurring infrequently on the active floodplains (FLACT) dominated by clay soils comprised a 'core' group of species which were broadly distributed on all landscape types (Table 1). These included Collema coccophorum B, Heppia despreauxii, Endocarpon rogersii, E. simplicatum var. bisporum, Psora decipiens, Peltula patellata ssp. australiensis, Catapyrenium squamulosum, Endocarpon pallidum and Synalissa symphorea. Where clays occurred in association with duplex soils, soil conditions were more conducive to lichen growth. Most species, which occurred less than five times on active floodplains included the species comprising the core group and isolated squamae of Endocarpon simplicatum var. simplicatum and small areoles of Lecidea ochroleuca.

(2) Footslopes and rolling downs of loamy duplex soils

Footslopes dominated by desert loams and brown gibber soils supported 17 lichen species. Crust cover was generally low (13.3%, Table 3), and lichens occurred as isolated squamae. Peltula patellata ssp. australiensis was frequent at some sites. Additional species not found in active floodplains included: Collema coccophorum A, Endocarpon pusillum, Eremastrella crystallifera, Catapyrenium pilosellum, C. lacinulatum and Heterodea muelleri. Peltula imbricata was found at one site where grazing had been excluded for more than 50 years.

(3) Ranges and hills of shallow loams

Ranges and hills of shallow loams supported a moderately high number of lichen species (23 species), although the abundance of individuals within a site was relatively low. The dominant lichens were Collema coccophorum B, Endocarpon pusillum and Heppia despreauxii. Although some lichens were found as scattered thalli, crust cover was moderately high (23.9%, Table 3), particularly in areas supporting a dense cover of perennial chenopod shrubs. Heterodea beaugleholei, Caloplaca spp., Cladonia sp., Trapelia coarctata, Diploschistes thunbergianus, Xanthoparmelia bellatula, Buellia subcoronata and Peltula subglebosa were additional species to those found in the previous two landscape types.

(4) Extensive plains of earthy sands and dunes of deep neutral red earths

The plains with sands associated with dunefields and sandplains between Bourke and the South Australian border supported a depauperate cover of ephemeral and perennial vascular plants. As lichens require a stable vascular cover prior to colonising the soil surface (Rogers

Tab

le 1

. D

istr

ibut

ion

of li

chen

spec

ies

amon

g th

e se

ven

land

scap

e ty

pes e

xam

ined

in s

emi-

arid

and

ari

d N

ew S

outh

Wal

es, A

ustr

alia

, wit

h da

ta o

n gr

owth

fo

rm, a

nd th

e nu

mbe

r of

sit

es w

ithi

n w

hich

spe

cies

wer

e fo

und

Gro

wth

for

m:

CR

= c

rust

ose,

SQ

= s

quam

ulos

e, F

O =

folio

se.

Lan

dsca

pe t

ype:

FL

AC

T =

act

ive

floo

dpla

ins,

FT

SLO

= f

oots

lope

s, R

NG

HL

= r

ange

s an

d hi

lls,

PLSN

D =

pla

ins

with

san

ds, P

LR

ED

= p

lain

s w

ith r

ed e

arth

s, F

LR

EL

= re

lict f

lood

plai

ns, P

LC

AL

= p

lain

s w

ith c

alca

reou

s ea

rths

. All

spec

ies,

apa

rt fr

om th

e va

gran

t C

hond

rops

is se

miv

irid

is a

re a

ttach

ed. T

he m

ost c

omm

only

occ

urri

ng s

peci

es a

re p

rese

nted

fir

st; +

=pre

sent

, - =

abs

ent

Lan

dsca

pe ty

pe

Gro

wth

N

o. o

f FL

AC

T

FTSL

O

RN

GH

L

PLSN

D

PLR

ED

F

UE

L

PLC

AL

fo

rm

site

s

Col

lem

a co

ccop

horu

m T

uck.

(m

orph

olog

ical

gro

up B

) H

eppi

a de

spre

auxi

i (M

ont.)

Tuc

k E

ndoc

arpo

n ro

gers

ii M

cCar

thy

End

ocar

pon

sim

plic

atum

(N

yl.)

Nyl

. va

r.

bisp

orum

McC

arth

y P

sora

dec

ipie

ns (

Hed

wig

) Hof

fm.

Pel

tula

pat

ella

ta (

Bag

l.) S

win

scow

& K

rog

ssp.

au

stra

liens

is (

Mue

lLA

rg.)

Biid

el

Cat

apyr

eniu

m s

quam

ulos

um (

Ach

.) B

reus

s E

ndoc

arpo

n pa

llidu

m A

ch.

Syna

lissa

sym

phor

ea (A

ch.)

Nyl

. k

cid

ea o

chro

leuc

a Pe

rs.

End

ocar

pon

sim

plic

atum

(N

yl.)

Nyl

var

. sim

plic

atum

E

ndoc

arpo

n he

lmsi

anum

Mu

ell

kg

C

olle

ma

cocc

opho

rum

Tuc

k. (

mor

phol

ogic

al g

roup

A)

End

ocar

pon

pusi

llum

Hed

wig

C

atap

yren

ium

pilo

sellu

rn B

reus

s E

rem

astr

ella

cry

stal

lifer

a (T

aylo

r) G

.Sch

neid

er

Cat

apyr

eniu

m la

cinu

latu

m (

Ach

.) B

reus

s H

eter

odea

mue

lleri

(H

ampe

) Nyl

. P

eltu

la i

mbr

icat

a R

.Fils

on

Het

erod

ea b

eaug

leho

lei

R.F

ilson

Cal

opla

ca s

pp.

Cla

doni

a sp

p.

Dip

losc

hist

es th

unbe

rgia

nus

(Ach

.) L

umbs

ch &

Vez

da

Tra

pelia

coa

rcta

ta (S

m.)

M.C

hois

y X

anth

opar

mel

ia b

ella

tula

(Kur

okaw

a &

Fils

on)

Elix

&Jo

hnst

on

Bue

llia

subc

oron

ata

(Mue

llA

rg.)

Mal

me

Pel

tula

sub

gleb

osa

(Mue

lLA

rg.)

R.F

ilson

F

ulge

nsia

sub

brac

teat

a (N

yl.)

Poel

t P

arap

orpi

dia

glau

ca (

T. T

aylo

r) R

ambo

ld

Aca

rosp

ora

nodu

losa

(D

ufou

r) H

ue

Ton

inia

sed

ifolia

(Sc

op.)

Tim

dal

Aca

rosp

ora

schl

eich

eri (

Ach

.) M

assa

l. X

anth

opar

mel

ia p

seud

oam

phix

anth

a (E

lix)

Elix

& J

ohns

ton

Xan

thop

amel

ia r

epta

ns (K

urok

awa)

Elix

& J

ohns

ton

Cho

ndro

psis

sem

ivir

idis

(F.

Mul

ler e

x N

yl.)

Nyl

. A

spic

ilia

calc

area

(L.)

Mud

d A

spic

ilia

cont

orta

(Hof

f.)

Kre

mpe

lh.

Xan

thop

amel

iafl

aves

cent

irea

gens

(G

yeln

ik) G

allo

way

A

caro

spor

a re

agen

s Z

ahlb

r.

Xan

thop

amel

ia a

mph

ixan

tha

(Mue

lLA

rg.)

Hal

e X

anth

opar

mel

ia v

ersi

colo

r H

ale

Xan

thop

arm

elia

terr

estr

is (K

urok

awa &

Fils

on)

Elix

& J

ohns

ton

Xan

thop

arm

elia

con

stip

ata

(Kur

okaw

a &

Fils

on)

Elix

&Jo

hnst

on

Xan

thop

arm

elia

sub

dist

orta

(K

urok

awa)

Hal

e D

iplo

schi

stes

hen

ssen

iae

Lum

bsch

& E

lix

Dip

losc

hist

es m

usco

rum

(Sco

p.)

R.S

ant.

ssp.

ba

rtle

rtii

Lum

bsch

D

iplo

schi

stes

oce

llatu

s (V

ill.)

Nor

man

N

eofu

scel

ia p

ulla

(A

ch.)

Ess

linge

r

Table 2. Frequency distribution of number of lichen species in the 282 sites in the survey area

D. J. Eldridge

No, of taxa No. of sites Percentage per site of sites

Table 3. Mean crust cover, total number of species and average number of species per site for each of the seven landscape types examined in semi-arid and arid New South Wales

Landscape type No. of Crust cover (%) Total no. Average no. sites of taxa taxa per site

Mean (s.e.m.)

Active floodplains (FLACT) 15 7.3 (7.24) 12 5.0 Footslopes (FTSLO) 39 13.3 (2.49) 16 5.7 Ranges and hills (RNGHL) 17 23.9 (3.90) 23 8.4 Plains with sands (PLSND) 54 8.6 (1.77) 28 5.5 Plains with red earths (PLRED) 43 27.7 (2.85) 34 11.2 Relict floodplains (PLREL) 47 14.6 (2.60) 3 1 2.8 Plains of calcareous earths (PLCAL) 60 20.9 (2.85) 38 8.8

1974), the actively eroding dunes and sandplains supported few lichens and few if any bryophytes (Doody, pers. comm.). Twenty-eight species were associated with these landscapes, the more commonly occurring lichens included Collema coccophorum B, Heppia despreauxii, Endocalpon pusillum, E, rogersii and occasionally E. simplicatum var. bisporum, Lecidea ochroleuca and Peltula patellata ssp, australiensis. Additional lichens included Fulgensia subbracteata, Paraporpidia glauca, Acarospora nodulosa, Toninia sedifolia, Acarospora schleicheri and Xanthoparmelia pseudoamphixantha. Crust cover was low (8.6%, Table 3).

(5) Plains, low ridges andflats of loamy red earths

Plains with red earths supported a suite of 33 lichen species and the highest mean crust cover of 27.7% (Table 3). Common species included Collema coccophorum B, Cladonia sp., Heppia despreauxii, Endocarpon pusillum, E, simplicatum var, bisporum and Peltula patellata ssp. australiensis, Psora decipiens and Lecidea ochroleuca. Less commonly occurring species included Diploschistes thunbergianus, Endocarpon rogersii, Xanthoparmelia spp. and Heterodea beaugleholei. Other species are given in Table 1. On the eastern margin of the survey area east of Cobar, Endocarpon simplicatum var. bisporum was a common component of soil crusts. The distribution of this lichen is thought to be restricted to areas exceeding 350400 mm rainfall (McCarthy 1991b).

(6) Relictfloodplains, playas and drainage lines

Thirty-one lichen species occurred on relict floodplains dominated by duplex soils. Some of these sites were located within the Willandra Lakes system north-east of Mildura, a

Lichens in Soil Crusts

system of relict lakebeds and paleochannels supporting open grasslands and chenopod shrublands. Common species included Collema coccophorum B, Endocarpon pusillum, Heppia despreauxii and Catapyrenium squamulosum. There were many species with less than six occurrences in this landscape type. On low chenopod shrub steppe on the Riverine Plain, sparse Xanthoparmelia spp. were found in sheltered rnicrosites such as under shrubs, particularly where the soils were slightly alkaline (pH > 8.0).

(7) Sandplains and dunefields of calcareous earths

Plains of calcareous earths supported a rich suite of lichens (and associated bryophytes, Eldridge and Tozer 19966). Despite a mean crust level of 20.9% (Table 3), crust coverage on some quadrats in this landscape type often exceeded 80%. The most common lichens included Collema coccophorum A and B , Endocarpon pusillum, Heppia despreauxii, Endocarpon rogersii and Psora decipiens. Less commonly occurring species included Catapyrenium pilosellum, C. squamulosum, Endocarpon simplicatum var. bisporum, Eremastrella crystallifera, Lecidea ochroleuca and Synalissa symphorea. Xanthoparmelia spp. were locally abundant in belah-rosewood woodlands and Aspicilia calcarea was found almost exclusively on calcareous soils.

Microsite-scale distribution

Apart from differences in distribution at a regional and landscape scale, terricolous lichens are strongly influenced by changes at micro-habitat scales (Scott 1982). Large-scale patterning or heterogeneity is a regular feature of many semi-arid landscapes, and in the semi-arid woodlands in eastern Australia, it operates at spatial scales of 50-100 m (Tongway and Ludwig 1990). The operation of these landscapes is controlled by topography, with small changes in slope generating toposequences of run-on and run-off. Water lost from smooth, poorly draining bare slopes (run-off zones) accumulates in vegetated flats and depressions (run-on zones) lower down the slope. Organic matter and soil nutrients accumulate in these run-on zones through a combination of biotic recycling and abiotic processes, particularly erosion (Tongway and Ludwig 1994). In most systems, an intermediate interception zone occurs at the upslope edge of the run-on zone. This has characteristics of both run-on and run-off zones and supports a variable cover of perennial grasses. Vascular plant growth and biological activity is greatest in the interception and run- on zones.

The distribution of lichen- and bryophyte-dominated soil crusts was examined in a patterned Callitris glaucophylla woodland that exhibited clearly defined run-off, interception and run-on zones. Data on cover and floristics of soil crust biota were collected from 10 x 0.5 m2 quadrats in each of the three zones. Crust cover was significantly greater in the interception zone (mean 2 s.e.m. 79.0 + 3.48%) compared with either the run-off zones (24.0 +. 4.09%) or the densely timbered run-on zones (6.6 rt 3.37%; F2,,, = 106.57, P < 0.001). Low cover levels in the run-on zones were attributed to the buildup of litter from the Callitris trees, and individual lichens were restricted to areas of bare soil around the tree trunks. Whilst there were no significant differences in the number of lichen species between the run- off (14), interception (19) and run-on (15) zones (P = 0.1 18; Table 4), five species, Buellia subcoronata, Heterodea beaugleholei, Psora decipiens and Xanthoparmelia sp. and Catapyrenium pilosellum, were restricted exclusively to the interception zones (though the latter was collected in a single quadrat only).

At a finer scale, some lichens prefer certain microsites over others. These microsites, patterned by local changes in slope, tree, shrub and pasture cover, soil surface micro-relief, and erosion, varied at scales of centimetres within the boundaries of the 0.5 m2 quadrats. By contrast to Heterodea spp., which prefer shaded microsites with high cover of perennial grasses, moderately dense shrubs (e.g. Dodonaea spp., Senna spp.) and sparse litter,

D. J. Eldridge

Table 4. Relative abundance of lichen species from three geomorphic zones in a patterned semi- arid woodland Values indicate number of quadrats (maximum of 10) containing a given species

Lichens Run-off zone Interception zone Run-on zone

Buellia subcoronata Catapyrenium squamulosum Catapyrenium lacinulatum Catapyrenium pilosellum Cladonia sp. Collema coccophorum A Collema coccophorum B Endocarpon pallidum Endocarpon pusillum Endocarpon robustum Endocarpon rogersii Endocarpon simplicatum var. bisporum Eremastrella crystallifera Heppia despreauxii Heterodea beaugleholei Heterodea muelleri Lecidea ochroleuca Paraporpidia glauca Peltula sp. Psora decipiens Trapelia coarctata Xanthoparmelia sp.

Total no. of species

squamules of Cladonia sp. tended to occur in open microsites where plant cover and biomass were sparse, or on sites with steeper (> 5%) slopes (Eldridge and Gittins, unpub. data). Open sites were also favoured by the crustose lichen Lecidea ochroleuca, and to a lesser extent the ubiquitous Peltula patellata ssp. australiensis. Endocarpon pal l idum was generally associated with sites of low slope and surface micro-relief, generally as individual squamae unassociated with extensive crusts (Eldridge and Gittins, unpub. data).

The lichen species referred to above as the 'core group' had widespread distributions which were disassociated with any of the measured environmental gradients. These species, which included Collema coccophorum B, Heppia despreauxii, Synalissa symphorea and Peltula patellata ssp. australiensis often occurred as individual squamae in landscapes where crusts were poorly developed. Rogers (1974) termed these species 'pioneers', as they tended to occur early in the successional development on soil surfaces, and did not appear to be adversely affected by grazing (Rogers and Lange 1971). For example, both morphological groups of Collema coccophorum were often found partially buried in the soil surface, suggesting that they are capable of occupying less stable microsites.

Crust Cover and Floristics

Crust cover is defined as the proportion of the soil surface occupied by biological soil crusts. Whilst crusts also included bryophytes, cyanobacteria, fungi and bacteria (Eldridge and Tozer 1996b), lichens comprised the bulk of the visible biota inhabiting the crusts. Analysis of individual site data ( n = 282) indicated no clear relationship between crust cover

Lichens in Soil Crusts

and number of species found at a site (P > 0.05). Even when data for the seven landscape types were analysed separately, there were still no significant relationships (P ranged from 0.08 (for plains of calcareous earths) to 0.91 (for active floodplains))

Whilst data in Table 3 suggest a linear relationship between mean crust cover and mean number of species (n = 7), the relationship was not significant (F1 = 2.44, P = 0.179) when all landscape types are considered. However, using data for only those landscapes with non- calcareous soils (ranges and hills, footslopes, active floodplains and plains with red earths), the relationship between cover and floristics was significant (F1,2 = 23.17, P = 0.041), and mean crust cover explained 88% of the variation in mean number of species.

Intuitively, it would be expected that greater cover would equate with a more stable surface, and thus greater biodiversity. However, considering all sites and landscapes, generally this was not the case in the present study. Stable sites, in terms of maximum crust cover and therefore resistance to erosion, were often dominated by fewer species. The weak relationship probably relates to different histories of disturbance, particularly grazing, and recovering sites would probably have been dominated by few species, whereas sites in good condition may have supported many more species despite similar cover levels between the two.

Comparisons with Other Studies

A comprehensive list of lichens reported from other semi-arid and arid regions in Australia is given in Table 5. Although differences in the scale of the studies (i.e. regional cf. landform scale), taxonomic changes and problems in identifying some species to species level make direct comparisons between these various studies difficult, a few generalisations can be made. Only one lichen, Psora decipiens, was recorded in all studies, possibly because of its extensive distribution on calcareous soils throughout semi-arid and arid Australia, but also because its pink squamules are distinctive in the field. With the exception of the Northern Territory lichens (Sammy, unpub, data), the other studies lie within the Murray Basin, where the contemporaneous material overlies Tertiary marine deposits rich in calcareous sediments (Northcote 1980). Other species found in 7 or more of the 10 studies included Collema coccophorurn, Cladonia sp., Endocarpon pusillurn, Erernastrella crystallifera and Peltula patellata ssp. australiensis (P. australiensis or P. patellata).

The greatest number of species (48) was recorded in the present study, followed by the Cobar-Nymagee study (Wells and Hynson, unpub, data) (39) and lichens found in the arid regions of the Rogers and Lange (1972) study in NSW and SA (28). The Cobar-Nymagee and the Rogers and Lange (1972) studies recorded 28 (57%) and 22 (45%) lichens in common with the present study respectively.

Roles of Soils Crusts in Rangeland Health and Biodiversity

Soil crust lichens are susceptible to trampling and fire (Eldridge and Greene 1994b), but during droughts under low to moderate stocking levels, they are often the only form of biological covering on the soil surface. Whilst traditional rangeland assessment has relied on measurements of the vascular plant community alone, there has been an increasing emphasis on the use of soil surface condition as an indicator of ecosystem health. Methodologies have been developed that link soil crusts with indices of rangeland health (Tongway and Smith 1989; Tongway 1994). Integral components of the soil surface condition assessment are cover and composition of mosses, lichens and other non-vascular plants (cryptogams) on the surface.

Preliminary work from the semi-arid woodlands of eastern Australia (Eldridge et al. 1995b; Eldridge and Koen, unpub. data) indicates that some lichens are useful indicators of landscape health. The foliose lichens Xanthoparmelia spp., Chondropsis semiviridis, Heterodea spp, and the squamulose Cladonia sp. were generally associated with landscapes in excellent condition. Many of the squamulose lichens such as Endocarpon pusillum,

D. J. Eldridge

Table 5. Terricolous lichens from arid and semi-arid regions of Australia Data drawn from various published and unpublished reports. 1, Tozer and Eldridge (unpub. data), Lake Mere, New South Wales (NSW); 2, Eldridge (1995), Mungo National Park, NSW; 3, Eldridge and Kinnell (1997), Mungo National Park, NSW; 4, Sammy (unpub. data), arid Northern Territory; 5, Scott (1982), Wyperfeld National Park, Victoria; 6, Rogers (1974), Koonamore Reserve, South Australia (SA); 7, Rogers and Lange (1972), arid NSW and SA; 8, Hyde (1994), mid-North SA; 9, Wells and Hynson (unpub. data), Cobar and Nymagee, NSW. 10, Eldridge (current study). + = present; - = absent

Study

Taxa

Acarospora nodulosa Acarospora reagens Acarospora schleicheri Acarospora smaragdula Acarospora sp. Aspicilia calcarea Aspicilia contorta Aspicilia sp. Buellia epigaea Buellia subcoronata Buellia sp. Caloplaca cinnabarina Caloplaca subpyracella Caloplaca sp. Catapyrenium lacinulatumA Catapyrenium pilosellum Catapyrenium squamulosum Catapyrenium sp. Chondropsis semiviridis Cladia aggregata Cladia corallaizon Cladia sp. Cladonia sp. Collema coccophorum A Collema coccophorum B Diploschistes hensseniae Diploschistes muscorum ssp. bartlettii Diploschistes ocellatus Diploschistes scruposus Diploschistes thunbergianus Diploschistes sp. Endocarpon helmsianum Endocarpon pallidum Endocarpon pusillum Endocarpon robustum Endocarpon rogersii Endocarpon simplicatum

var. simplicatum var. bisporum

Endocarpon sp.B Eremastrella crystallifera Fulgensia subbracteata Heppia despreauxii Heterodea beaugleholia Heterodea muelleri

Lichens in Soil Crusts

Table 5. (continued)

Study

Taxa 1 2 3 4 5 6 7 8 9 1 0

Lecidea ochroleuca Lecidea sp. Neofuscelia pulla Paraporpidia glauca Peltula imbricata Peltula patellata var australiensisC Peltula subglebosa Peltula sp. Psora crenata Psora decipiens Synalissa symphorea Siphula coriacea Tofiinia sedifolia Trapelia coarctata Xanthoparmelia alternata Xanthoparmelia amphixantha Xanthoparmelia bellatula Xanthoparmelia constipata Xanthoparmelia convoluta Xanthoparmelia eilifii Xanthoparmeliaflavescentireagens Xanthoparmelia pseudoamphixantha Xanthoparmelia reptans Xanthoparmelia subdistorta Xanthoparmelia terrestris Xanthoparmelia versicolor Xanthoparmelia spp.D

*includes species assigned to Catapyrenium lachneum and Dermatocarpon lachneum. B~ncludes species assigned to Parmelia spp. sens. lat. C~ncludes species assigned to Heppia polyspora by Rogers (1974). D~robably includes Catapyrenium spp.

E. rogersii and Collema coccophorum had no preference for sites within a given condition class. Xanthoparmelia spp., particularly, were only found in sites with a conservative grazing history. Their role as indicator species is not surprising given that they are extremely long- lived, slow growing and susceptible to trampling. Low levels of trampling probably aids dispersal of vegetative propagules, however the thalli are likely to disintegrate under moderate to high levels of trampling (Eldridge and Tozer 1996~) .

There are few data on the conservation status of Australian soil crust lichens. Most data relate to the loss of biodiversity in more mesic systems such as old growth forests or coastal rainforests where habitats are being altered by logging and urban encroachment. Because soil lichens are small, often infertile, and strongly weathered due to the extremes of climate, some genera are difficult to identify to species level. Like the bryophytes, their presence or absence in certain environments may merely be an artefact of the collection procedure, rather than a true reflection of intrinsic habitat preference (Eldridge and Tozer 199617). The value of this survey however is that it provides a benchmark upon which future assessments of lichen biodiversity can be made.

D. J. Eldridge

A number of species including Aspicilia contorta, Diploschistes muscorum ssp. bartlettii, D: hensseniae, Peltula imbricata and Acarospora reagens were found at only a few locations in the survey area. Although more intensive collections may uncover additional specimens, data from the present study suggest that more information is needed about their habitat requirements if their conservation status is to be assessed. For example, Heterodea muelleri was found at only five locations in arid NSW and is probably at the edge of its distribution. It is known to be more abundant in the temperate areas of eastern Australia (Rogers 1970), where its conservation status is probably secure.

Two of the above species were generally restricted to lightly grazed or ungrazed sites. Peltula imbricata was collected at only four locations, including the regeneration area around the mining town of Broken Hill where it is locally abundant. Fertile specimens were common in this area which has been largely ungrazed by domestic livestock for more than 60 years. Similarly, Acarospora reagens was a common component of the soil crust, occurring with Catapyrenium squamulosum and Collema coccophorum on non-calcareous clays at only one site west of Condobolin. The site, a Soil Conservation Service Experimental Area, was fenced in the early 1950s to exclude domestic grazing animals, and represents a remnant of Atriplex vesicaria shrubland which was formerly common across the district. A single specimen of Xanthoparmelia constipata was collected from Cobar Regeneration Area, a formerly degraded area around the township of Cobar which has been allowed to regenerate in the absence of domestic livestock. More intensive examination of these exclosed areas will undoubtedly reveal other new and interesting records for the region.

Soil Crusts and Land Management

Soil surfaces with crusts comprising an extensive cover and rich diversity of lichens, mosses and other microbiota are generally associated with enhanced soil structure, aggregate stability and infiltration, and lower rates of sediment removal compared with sites with low cover and composition of crusts (Greene and Tongway 1989; Eldridge 1993a, 1993b; Eldridge and Greene 1994b).

Superimposed, upon landscape and microsite distributions, is the overriding influence of land management, particularly grazing, fire and land clearing. These activities have markedly reduced crust cover and composition in many areas (Eldridge and Greene 1994b). Whilst crusts may have coexisted under grazing by soft-footed native animals such as kangaroos, the accepted view is that a century of European pastoralism, as well as human and vehicular damage, has resulted in the their widespread destruction. This is also associated with an increasing cover of woody shrubs, increases in erosion and a loss of pastoral productivity (Harrington et al. 1984; Harper and Marble 1989; West 1990).

The occurrence of poorly developed lichen crusts on active floodplains with clay soils probably relates to their inability to persist under frequent flooding or to survive on seasonally cracking soils. On relict floodplains however, changes in species composition of the vascular plant community may lead to reductions in the cover and composition of lichen crusts. For example, conversion of a stable shrubland dominated by Atriplex vesicaria and Maireana pyramidata to an annual grassland dominated by species such as Hordeum leporinum and Medicago polymorpha leads to higher levels of biomass and cover, and reductions in the area of bare ground during average seasons. Consequently, increased competition for light and moisture may lead to reduced lichen cover. This has been demonstrated in mallee (Eucalyptus spp.) shrublands where buildup of leaf litter on the soil from the Eucalyptus spp. eventually restricts the growth of lichens and mosses, probably by occupying suitable microsites for establishment (Eldridge and Bradstock 1994). Conversion of shrublands to grasslands through inappropriate grazing management may increase the frequency of wildfires in areas where they were once uncommon. As lichens are known to be susceptible to fire (Greene et al. 1990; Eldridge and Bradstock 1994), increased fire frequency may result in substantial damage to the lichen community.

Lichens in Soil Crusts

Overgrazing of soils where crusts are a dominant component is likely to lead to a reduction in species diversity and changes in soil hydrology. Rogers and Lange (1971) showed that stocking pressure caused a differential destruction of lichen species at distances out from the watering point. Unlike vascular plants, all lichens were affected to a certain degree, and their frequency decreased as stocking intensity increased. Similarly, Andrew and Lange (1986) showed that after a new water point was constructed in a semi-arid chenopod shrubland, the frequency of crustose lichen increased markedly with distance from the water point. They attributed this to sheep grazing, and showed that soil bulk density increased by up to 20%, but dust fallout increased only marginally (Andrew and Lange 1986).

Land clearance for dryland farming destroys stable soil crusts, and probably leads to reduced vascular plant and invertebrate biodiversity. Furthermore, as the lichen components of these crusts are slow-growing, regeneration after cropping is likely to be slow or non- existent. Generally, lichens require a permanent vegetation cover before actively stabilising degraded surfaces. Rogers (1970) reported that lichen crusts at Koonamore Regeneration Area in the South Australian pastoral zone were still markedly reduced even after 44 years of exclosure. In the present study, sparse crusts on abandoned cropping sites were dominated by Collema coccophorum or various cyanobacteria.

Low risk stocking is recommended as an appropriate strategy for maintaining vital soil processes and minimising damage to soil crusts and the associated surface (Eldridge et al. 1995~) . Land managers should aim to reduce grazing intensity by dispersing stock over large areas of rangeland rather than concentrating them in small areas. This is often inconsistent with practices such as feeding stock in small holding paddocks during droughts, which will increase the risk of damage to the soil surface and lichen crust, particularly during summer when the soil is dry and susceptible to destruction (Harper and Marble 1989).

The positive role of lichen crusts in infiltration, soil stability, soil nutrition and vascular plant regeneration, reinforces the popular belief that crusts are indicators of a healthy land condition. Their importance in ecological processes in semi-arid rangelands means that they should be included in Environmental Impact Assessment methodologies. Protocols to assess land condition have already been developed (Tongway 1994), and cover and composition of biological soil crusts, and particularly lichens, are a major component of this system.

Acknowledgments I am grateful to Merrin Tozer for her contribution during all aspects of this study, from

field collection, identification, cataloguing and preparation of voucher specimens, to editing and commenting on earlier drafts. Many people assisted with lichen identification including Alan Archer, Othmar Breuss, Burkard Biidel, Jack Elix, Patrick McCarthy, Thorsten Lumbsch, Bruce Reardon, Rod Rogers and Roger Rosentreter. Melissa Dryden and Sabine Slangen assisted with field work, and Patrick McCarthy, Rod Rogers and Rex Filson provided useful comments on an earlier draft. The financial and logistical support of the Land and Water Resources Research and Development Corporation and the Department of Conservation and Land Management (Soil Conservation Service) is gratefully acknowledged. Finally, Rod Rogers and Neil West fueled my interest in lichens in the early 1980s with their pioneering work on soil crusts. They continued to be enthusiastic supporters of research aimed at understanding more about the ecology and management of these soil crust organisms.

References Andrew, M. H., and Lange, R. T. (1986). Developments of a new piosphere in arid chenopod shrubland

grazed by sheep. I. Changes to the soil surface. Australian Journal of Ecology 11,395410. Beadle, N. C. W. (1948). 'The Vegetation and Pastures of Western New South Wales with Special

Reference to Soil Erosion.' (Government Printer: Sydney.)

D. J. Eldridge

Cunningham, G. M, Mulham, W. E., Milthorpe, P. L., and Leigh, J. H. (1981). 'Plants of Western New South Wales.' (Government Printer: Sydney.)

Downing, A. J., and Selkirk, P. M. (1993). Bryophytes on the calcareous soils of Mungo National Park, and arid area of southern central Australia. Great Basin Naturalist 53, 13-23.

Eldridge, D. J. (1993~). Cryptogams, vascular plants and soil hydrological relations: some preliminary results from the semi-arid woodlands of eastern Australia. Great Basin Naturalist 53,48-58.

Eldridge, D. J. (1993b). Cryptogam cover and soil surface condition: effects on hydrology on a semi- arid woodland soil. Arid Soil Research and Rehabilitation 7, 203-217.

Eldridge, D. J. (1995). Dispersal of microphytes by water erosion in an Australian semi-arid woodland. Lichenologist 28, 97-100.

Eldridge, D. J., and Bradstock, R. A. (1994) The effect of time since fire on the cover and composition of cryptogamic soil crusts on a eucalypt shrubland soil. Cunninghamia 3,521-527.

Eldridge, D. J., and Greene, R. S. B. (1994~). Assessment of sediment yield from a semi-arid red earth with varying cover of cryptogams. Journal of Arid Environments 26,221-232.

Eldridge, D. J., and Greene, R. S. B. (1994b). Microbiotic soil crusts: a review of their roles in soil and ecological processes in the rangelands of Australia. Australian Journal of Soil Research 32, 389-415.

Eldridge, D. J., and Kinnell, P. I. A. (1997). Quantitative assessment of erosion rates from microphyte- dominated calcareous soils under rain-impacted flow. Australian Journal of Soil Research (in press).

Eldridge, D. J., and Tozer, M. E. (1996~). Xanthoparmeliae: key indicators of rangeland health in semi- arid New South Wales. Australian Lichenological Newsletter 38, 19-20.

Eldridge, D. J., and Tozer, M. E. (19963). Distribution and floristics of bryophytes in soil crusts in semi-arid and arid eastern Australia. Australian Journal of Botany 44, 223-247.

Eldridge, D. J., Chartres, C. J., Greene, R. S. B., and Mott, J. J. (1995~) . Management of sealing, crusting and hardsetting soils under rangeland conditions. In 'Sealing, Crusting and Hardsetting Soils: Productivity and Conservation'. (Eds H. B. So, G. D. Smith, S. R. Raine, B. M. Schafer, B. M. and R. J. Loch.) pp. 381400. (Australian Society of Soil Science, Qld. Branch: Brisbane.)

Eldridge, D. J., Tozer, M. E., and Koen, T. B. (19956). Composition of bryophytes in arid soil crusts: what can they tell us about rangeland health? In 'Proceedings of the Fifth International Rangelands Congress, Salt Lake City, Utah'. (Ed. N. E. West.) pp. 130-131. (Society for Range Management: Denver, Colorado.)

Filson, R. B. (1988). The lichen genera Heppia and Peltula in Australia. Muelleria 6,495-517. Filson, R. B. (1992). Heterodeaceae. In 'Flora of Australia'. Vol. 54. (Ed. A. S. George.) pp. 198-200.

(Australian Government Publishing Service: Canberra.) Filson, R. B., and Rogers, R. W. (1979). 'Lichens of South Australia.' (Government Printer, South

Australia: Adelaide.) Galun, M. (1963). Autecological and synecological observations on lichens of the Negev, Israel. Israel

Journal of Botany 12, 179-187. Gerson, U., and Seaward, M. R. D. (1977). Lichen-invertebrate associations. In 'Lichen Ecology'.

(Ed. M. R. D. Stewart.) pp. 69-120. (Academic Press: London.) Greene, R. S. B., and Tongway, D. J. (1989). The significance of (surface) physical and chemical

properties in determining soil surface condition of red-earths in rangelands. Australian Journal of Soil Research 27, 213-225.

Greene, R. S. B., Chartres, C. J., and Hodgkinson, K. H. (1990). The effect of fire on the soil in a degraded semi-arid woodland. I. Physical and micromorphological properties. Australian Journal of Soil Research 28, 755-777.

Harden, G. J. (Ed.) (1990-1993) 'Flora of New South Wales'. Vols 1-4. (University of NSW Press: Sydney.)

Harper, K. T., and Marble, J. R. (1989). A role for non-vascular plants in management of arid and semi- arid rangelands. In 'Application of Plant Sciences to Rangeland Management and Inventory'. (Ed. P. T. Tueller.) pp. 135-169. (Martinus Nijhoff and W. Junk: Amsterdam.)

Harper, K. T., and Pendleton, R. L. (1993). Cyanobacteria and cyanolichens: can they enhance availability of essential minerals for higher plants? Great Basin Naturalist 53,59-72.

Harper, K. T., and St Clair, L. L. (1985) Cryptogamic soil crusts on arid and semi-arid rangelands in Utah. Effects on seedling establishment and soil stability. Final Report to the Bureau of Land Management, Utah State Office, Salt Lake City, Utah.

Lichens in Soil Crusts

Harrington, G. N., Wilson, A. D., and Young, M. D. (1984). 'Management of Australia's Rangelands.' (CSIRO: Melbourne.)

Hyde, M. K. (1994). 'A Vegetation Survey of Disused Railway Corridors in the Mid-north Region of South Australia. July-November 1992.' (Nature Conservation Council of South Australia: Adelaide.)

Johnson, E. R. L., and Baird, A. M. (1970). Notes on the flora and vegetation of the Nullarbor Plain at Forrest, Western Australia. Journal Royal Society Western Australia 53,5641.

McCarthy, P. M. (1991~) The lichen genus Endocarpon Hedwig in Australia. Lichenologist 23,27-52. McCarthy, P. M. (1991b) 'Checklist of Australian Lichens.' 4th Edn (National Herbarium of Victoria:

Melbourne.) Northcote, K. H. (1980). Soils in aeolian landscapes in part of the Murray Basin of south-eastern

Australia. In 'Aeolian Landscapes in the Semi-arid Zone of Eastern Australia: Proceedings of a Conference Held at Mildura, Victoria, Australia on October 17-18, 1979'. (Eds R. R. Storrier and M. E. Stannnard.) pp. 102-122. (Australian Society of Soil Science Incorportated, Riverina Branch: Wagga Wagga.)

Pickard, J., and Norris, E. H. (1994). The natural vegetation of north-western New South Wales: notes to accompany the 1:l 000 000 vegetation map sheet. Cunninghamia 3,423-464.

Rogers, R. W. (1970). Ecology of Soil-surface Lichens in Arid South-eastern Australia. PhD Thesis, University of Adelaide, South Australia.

Rogers, R. W. (1971). Distribution of the lichen Chondropsis semiviridis in relation to its heat and drought resistance. New Phytologist 70, 1069-1077.

Rogers, R. W. ( 1 9 7 2 ~ ) . Soil surface lichens in arid and sub-arid south-eastern Australia. 11. Phytosociology and geographic zonation. Australian Journal of Botany 20, 215-227.

Rogers, R. W. (1972b). Soil surface lichens in arid and sub-arid south-eastern Australia. 111. The relationship between distribution and environment. Australian Journal of Botany 20, 301-316.

Rogers, R. W. (1974). Lichens from the T.G.B. Osborn vegetation reserve at Koonamore in arid South Australia. Transactions of the Royal Society of South Australla 98, 113-124.

Rogers, R. W. (1977). Lichens in hot arid and semi-arid lands. In 'Lichen Ecology'. (Ed. M. R. D. Stewart.) pp. 21 1-252. (Academy Press: London.)

Rogers, R. W., and Lange, R. T. (1971). Lichen populations on arid soil crusts around sheep watering places in South Australia. Oikos 22,93-100.

Rogers, R. W., and Lange, R. T. (1972). Soil surface lichens in arid and sub-arid south-eastern Australia. I. Introduction and floristics Australian Journal ofBotany 20, 197-213.

Scott, G. A. M. (1982). Desert bryophytes. In 'Bryophyte Ecology'. (Ed. A. J. E. Smith.) pp. 105-122. (Chapman and Hall: London.)

St Clair, L. L., Johansen, J. R., and Rushforth, S. R. (1993). Lichens of soil crust communities in the intermountain area of the western United States. Great Basin Naturalist 53,5-12.

Tongway, D. J. (1994). 'Rangeland Soil Condition Assessment Manual.' (CSIRO Division Wildlife and Ecology: Canberra.)

Tongway, D. J., and Ludwig, J. A. (1990). Vegetation and soil patterning in semi-arid mulga lands of eastern Australia. Australian Journal ofEcology 15, 23-34.

Tongway, D. J., and Ludwig, J. A. (1994). Small-scale resource heterogeneity in semi-arid landscapes. Pacific Conservation Biology 1,201-208.

Tongway, D. J., and Smith, E. L. (1989). Soil surface features as indicators of rangeland site productivity. Australian Rangeland Journal 11, 15-20.

West, N. E. (1990) Structure and function of microphytic soil crusts in wildland ecosystems of arid and semi-arid regions. Advances in Ecological Research 20, 179-223.

Williams, J. D., Dobrowolski, J. P., West, N. E., and Gillette, D. A. (1995). Microphytic crust influence on wind erosion. Transactions of the American Society of Agricultural Engineers 38, 131-137

Willis, J. H. (1951). Botany of the Russel Grimwade Expedition. Memoirs of the National Museum of Victoria 17, 75-76.

Manuscript received 13 March 1996, accepted 14 June 1996