Cyanobacteria and cyanolichens: can they enhance ...

Great Basin Naturalist Great Basin Naturalist

Volume 53 Number 1 Article 8

4-2-1993

Cyanobacteria and cyanolichens: can they enhance availability of Cyanobacteria and cyanolichens: can they enhance availability of

essential minerals for higher plants? essential minerals for higher plants?

Kimball T. Harper Brigham Young University

Rosemary L. Pendleton Shrub Sciences Laboratory, Intermountain Research Station, USDA Forest Service, Provo, Utah

Follow this and additional works at: https://scholarsarchive.byu.edu/gbn

Recommended Citation Recommended Citation Harper, Kimball T. and Pendleton, Rosemary L. (1993) "Cyanobacteria and cyanolichens: can they enhance availability of essential minerals for higher plants?," Great Basin Naturalist: Vol. 53 : No. 1 , Article 8. Available at: https://scholarsarchive.byu.edu/gbn/vol53/iss1/8

This Article is brought to you for free and open access by the Western North American Naturalist Publications at BYU ScholarsArchive. It has been accepted for inclusion in Great Basin Naturalist by an authorized editor of BYU ScholarsArchive. For more information, please contact [email protected], [email protected].

Grcat Basin Naturalist·,)·'3tl), pp. ,')9-72

CYANOBACTERIA AND CYANOLICHENS, CAN THEY ENHANCEAVAILABILITY OF ESSENTIAL MINERALS FOR IIIGHER PLANTS?

I ,Kimball T. Ha'lJer and Hoselllary J,. Pcndlchlll~

ABSTH:HT-In lJOtb field and ?;reenhouse sLudies, (':'alH)hact(~ria aJld cyaJlo]ichpns of cold-telllpf'rate rlf'SPltS oftP]}f'nhance growth and essential elClIl('llt uptake In- associated herbs. ThaL er'f('ct is associated \vith hetter seedling estah­lishment and larger s('cdlillgS, The following are IXlssible mechanisms for these cO'eds: (1) tIlt-' micwhiota COJlCf'ntratf'essential f'lel1lents in available forms ill soil surface layers, (2) tht' lIJicrobial surface covers are lIsually clarb:'r colored thanthf' soil itself and produc(, wanll{'r soils (luring coo] s("aSOlls when soil waleI' is most availahle, (:1) the gelatinous sheaths ofseveral e:~l1lOhac:terial )?/'lwra common 011 alkaline deserts cUlltaill chelatiug compounds, <l1ll1 (4) conditions that favorpersistent microbial gnmths OJ] soil snrbcf's ,11 so bvor maintenance of larger rx)pulatiollS of ll!icroorgallisllls that fon]]mycorrhizal and/or rhizosllPath ;Issociations with seed plants. There is c"idellce tllat associat('d animals llIay 1)(' llutritionall)bClwfited by the enhanced mineral content of forage plants gnlwing in wf'll-dev(·loped (':'anohadcrial C!"llstS.

Key Icords; aypto{!.rllnic crusts, lIIineral uptakc. plant !l1ItritiO/l, /lImit grOlrtfl, iIlycorrfli::lle, dcserts, ('y(//w!Hlcferil(Collema, Microco1cllS.

Dense growths of cyanobacteJia, lichens,xerophytic algae, mosses, and lIlicrofungi are acommon feature of soil surfaces in semimicl andarid temperate regions workhvide (FJiedmann'md Galun 1874, Haq)erand Marble 1988, West1990). Such soil covers are cormnonlv nlUcll

~

darker than associated surfaces without suchgrowths (Fig. 1). Previous work on these Cl}T­

tohiotic surface growths (or "clJ1Jtogallliccrusts" as they arc often c<.Jled) has focused 011

their ahility to "flx" nitrogen in a hiologicallyavailahle f(mn (MacGregor and Johnson 1971,Mayland ct a1. 1966, Shields 'md Durrell 18(4)and to stahilize soil surfaces against water(Bmth 1941, Fletcher 'md Martill 1948, Fritsch1922) orwind erosion (MacKenzie ,mel Pearson19(9). Other studies have shovvn that nitrogenfixed hy cyanohacteJial components of the (1)1)­tobiotic crusts is available to higher plants(Fuller et a1. 1960, Mavlamj alld MacIntosh

~

1866, Stewart lfl(7).The intent of this paper is to assemble exist­

ing data aIllI report new data bearing on theinnuence of cyanobactelial-lich assemhlages ofcryptohiota on mineral nutrition of associatedvascular plants.

PEHTINENT LITETIATUnE

A large literature documents the role of cy­anobacteJia, cyanolichens, and free-liVing hac­teria (nonsymhiotic) in desert soils (Halper andMarble 1988, \Vest 1990). Since legulIles amIother Va.scllhrrphmts that fonn symbiotic, nib."ogen­fixing associations 'with baetelia are unCOllnnonin cold-temperate deserts, their importance assources ofbiologically available nitrogen is mini­mal in such environments (\Vest 1981). Rychertet ;,J. (1978) concluded that "blue-green algae"­lichen C'rusts fix significant amounts of atmos­phelic nitrogen in desert soils (they estimateflxation of 10-100 kg N'ha-1yr-1 in the CreatBasin of North Amelica). Available infonnationsuggests that nOllsymbiotiC', heterotrophiC nitro­gen fixers are responSible for fixation of only Q

kg N'ha \T I in North AmeJican deserts (1\y­chelt et a!. lm8, Stevn and Delwiehe lmO).The input of availahl(: nitrogen in annual pre­cipitation is apparently low, with estirnates run­ning from 4--{i kgha I yr I (West Hl78) to 1-2kg-lta I )'r- 1 (Sclllesinger 1991). SC'hlesinger(1 8\)1) notes that available N in dry-fall exceedsthat in rainfall in some areas, but \Vest andSkujins (1978) argue that since deserts produce

_~Dt'partLl1t'n(dBolany "",I Hani<" ~(';('ll"" B,igh"n, YOlll'g Ullin·"ily, Pn"'(J, L'lalJ ').f(~)2.- Shn,h S<:i'" K<" 1,a],oralo,-,,- lll!<'nnountaill H"'''''''ch Stat~",. LJ SDA 1"<", "~I S.. 1"'k.·, 7:).) '\ o,ih .')0(1 F:a,t. 1'1''''-'', Utah 1)-1(;1.--;'

.58

(ill GREAT BASI;\) :'\,\TUH.-\LlST 1Volume ,S:)

'.•, .,' ..• • •.... " ..• ' ... ,>... ~,, .

•

,. • • • .,. ••• • • • '" ". • ••

• •

-

Fig, J. Contrast lX'twecn natural soil color and the dark-colored cryptobiotic gro\\th on the soil surface. Such surfacegroWl hs of cy,lllOhactl'ria, the- hhwk lichen Colll'l!ltl. amI :wrophytic IllOS,9'S are COlllllJOn 011 dt'SPli soils ill the iJltt'fHlOll11tain

\\"(",t of \"orth .\merica. Soil shU\nl is rleycloped frolll am,jellt lacustrine deposits. w('stern lHal!.

rnore dust thall thev receive, N in dust (theprilll<U) SOUfCt-' ()f 1\ in (lry-hll1) prohahl:' rqxe­

sents a net CXIX)rt fnnn deserts. Df'nitJificatiollprocesses are active in deselts and 1ll<l:' c(!ual orexceed rates or fixation (Skujins amI KluhckHJ7B).

It is estimated that \'ascular plallts tah- up10-1 ,) kg N ha lyr- I in the cold deserts of t1leGreat Basin (\Vest and Skujins 19(7), \Ve caninfer from \\'ork h\ Homm,:\' et al. (197K) that. .al1lllLal uptake of N by vascular plants in thewartn deserts of the northcm :Nloha\-e is slightlygreater than that reported by \Vest and Sklljins(1977) for cold deserts, III either case, annualuptake is greater tllan the comhi ned total ofnewnitrogen added hy precipitation and heterotro­phic bactetia. Since tbe ,lTllOll1lt of N fhecl hysyrnhiotic organisms is apparently tIivi~J ill de­serts of both the Greal Basin and the nO/them:\Iohave, the inp"t of fixed I\ hy cyanobaetctia­lichen crusts is of significance in those desPlts(Rychert et al. 107B).

•The influt'nce of CI:IJtohiotic crusts on avail-

able forms ofcsscnti,ll clclll('nls other than N in

su rbce soils of deserts lias been documented in,1 \'<uiety of rcpOlts. Such studies often show thatavailable V exchangeable K, and Sl1rf~tcC soilorganic matter increase apace \\lth crypto1Jioticcover (Anderson et aL 198:2, KleineralHI Harper1079 • 1977a, 1077h. McKl1i~ht10BO). Thetrcndfor clements otlwr than N is apparently rebted,at h"'ast in pmt, to the tendency of the cryvtobi­otic CJ11sts to trap soil fines (silt and day) and tosequester essential elements in li\-ing cells(Fletcher and \lartin ]Y4S, Kleiner and IIarper1979).

Cr~vtohiotic crusts are usually darker thanassociated soils in oeselts (Fig, 0. As a result,soil slllfal'l's cO\'c..Ted lw such Cl11stS ahsorh radi-

~

ant cneq..,l\- better than nearbv uncl11stecl soils, -. .,Hill afe wanner (Kleiner and Harper 1977b,Harper alllI \larble 1988). Temperature differ­ences may Iw at least SOC (llarper and f\ifarble1YKH) and are prohahly greatest in cooler sea­SOilS. In tilt: Great Basin, where the hulk ofbiol(ll!;icallv l1sable moisture aCCllll1ulat8s from, .winter storms. areas covered bv cvanobaetelial­Collcma lichen Cntsts can he e\1Jt'eted to be

1993J CY;\I\OBACTEHIA A;\"]) CYAI\OLICIIK'\S 61

signific<mtly wanner t1 tan i llterspersed 1I11­crusted soils. Both soil miLTobes and associatedvascular plants (especially shallowly rooted spe­cies amI seedlings) should experience acceler­ated spring growth on sites where soiltemperature is elevated hecause of well-devel­oped IT)vtoLiotic growth. Enhanced growth ofseed plants rooted in clyptobiotic CJ1.1StS mightbe expected because of more f~lVorahlc tcm­peratures and more fertile soils. Physiologistshave long recognized tll<1t lTlallY physiologicalprocesses have a Qw of 2-3; that is, as tempera­tures \vithin tllP rangp of easy tolerancp areincreased lOGe, metabolic rates for processessHch as enzymatic reactions, ion uptake, and iOll

transpOlt are donhlee! or trip lee! ( ;Iass 19H9).Algologists amI microl)iologists have docu­

mented the natnre of organic secretions frommicroorganisms and speculated on their cco~

logical roles in the organisUl-envirOillnentalcomplex. Algae are known to secrele p(1)'sac­chmides, amino acids, vitamins, growth factors,steroids, saturated and unsaturated fi.lts, andother beneficial or toxic compounds of unknownstructure (I ,efevre 19(4). The secretions oftenform a gelatinous film around the cells. Amongcy:mobacte,ia, such sheaths are COllltllOll amIinclude polys<lcch<uides, organic acids, anlinoaciels, and polypeptides (Llllge 1974). Inaquatic systems, cxtraccilltlar organiC' secretionsplaya variety of roles including food for Ilf·t­erotrophic organisms, che1ating agents that in­crease availability of essential elcmcnts(particularly iron and other trace elements),growth stimulators, toxic compounds that dis­courage herbivOlY (and are sometimes auto­toxic), and compounds that complex with andinactivate toxic agents (such as copper) in water(Lefevre 19fi4).

Lange (1974) demonstrated tlmt slteatltma­terials of cyanobacteria indude chelating agentsthat permit the organisms to grmv \igorously inwater having a high pH in which several essen­tial clements would othel\vise be available insuch 1m\' alllounts that active growtlt would beimpossible. lIe dem{)nstrated that somc sp(~cies

in each ofthe follOWing genera secreted enoughnatural chelators to produce growth equivalentto that of the same species in cultures that re­ceived artificiqI chclators: Anabaclla, Anacystis,Lynghya, Aficrocystis, and Nostoc. He alsodemonstrated that the natunJ chclators werewater soluble and that tlltrates of cultures inwhich chelating species had grown suppOlted

good growtl I of llolldlelatillg species. The latterspecies were unable to grow in the same waterif chclating species had not preViously grown init. Lange (1976) c()J]c1Ulled that the gelatinoussheaths of cyanohactf'ria prmide a microenvi­ronment around their cells, where cssentialnu­tlients can be concelltrated from anenvironnlent in which those clements exist atlevels too low to sustain growth. III alkalinedf'Sf'ltS thf' hygroscopic nature of the copiousgelatinous sheaths produced by eyanobaeteria(aJl(llllany associated algae, hade ria, and fungi)sllggf'st anothf'r, Jlf'rhaps f'ssf'ntial, n)le for suchextracellular secretions, that of retainingenough water around cells in dry l)f'liods toprevpnt If'thal dpsiccation.

A large literature documents the importantrole of mvcolThizal associations for mineral nu-

~

tritional relationships of vasculm' plants (Allen1991). Allen (1991) concl"ded that "major per­turbations almost always reduce the inoculumdenSity" of 1l1ycOlThizal fungi. Bethlenfalvay ctal. (198.5) determined that trampling of soils byhoofed grazers (cattle) reduced 1I1ycOlThizal in­oculum in crested wheatgrass pastures in Ne­vada. Koide and Moonev (1987) showed thatIXJcket gopher lmrrowing in othcJ"\visf' Ilndis­turlwd se1lJentine herblcmds in coastal Califor­nia also recItlced ll1\'corrhii',al i IIOCU lu 1Il. PelT\' et

• •al. (19HB) rf'poltf'cl that sprollting shmbs Inaymaintain a myc(llThizal infection through thestressful conditions induced bv \\~Idfirp. \Vllll­stein and Pratt (l9H}) discussed the develop­lllf'llt of rhizoshe'lths of Ory::.opsis hYlllc/wides(R. & S.) Riker (now classified as StipahYllu'/Ioides H. & S.), a deselt grass. Ory::.:opsisand several other grasses of arid, sandy soilsproduct' COl1SI)icuous rhiz,oslleatlls c(lnsisting ()fdpnsf' tangles of root hairs and intermixed het­erotrophiC bacteria capable of thing nitrogen(\Vllllstein f't al. 1979). Unpuhlished data in thefiles of the senior author suggest that rhi­zoshcaths apparentl.yeldHUlce mineralllptake ofStlj){l hYlllcnoides for several elements.

The foregoing literature survey snggf'sts thatCI)11tohiotic Crtlsts may Significantly alter theuptake of essential elements by associated de­sert seed plants. In this paperwp rpport prelimi­nml' results on the effects of cryptobiotic coversdominated bv cvanohacteria and Colle lila on, .tissue chemistry of associated seed pl<llltS.Collcmfl is a black-colored Hcllen in which thephotohiont is the cyanobacterium Nos{oc. Spe­cifically, we will consider the eHects of the

62 GREY!' BASI); l\XITHILIST [Volume .5:3

Characteristic BI()\\, sand Cvanobactcrial-CtJffmw salKI

METHODS

cryptobiotic cover on soil fprtility, soil tempera­ture in the cool se<L,>on, possible chelationeffects, and colonization of seed plant roots hy

- -mieroorgamsllls.

paste. Organic matter \vas estinlated by diges­tion with 1.0 N potassiulll dichromate. Total soilnitrogen W,-L') anal)-l'zed bi' the micro-Kjeldahlprocedure. P!JOSp!JOlUS \V<.l'> deterrnined withthe iron-TCA-lIlolvbdate method on a soil ex­tract taken \vith 0.2:'\ acetic acid. Exchangeablebases were freed [rom the soil with 1.0 N am­monium chloride. Ion concentrations in the ex­tract \vere estimated by atomic absorption. Allsoil analvses \vere made in the Soil and PlantAnalysis Laboratory, Department cif Agronomyand HOlticnltmc, Bligham Young University,and all anahtical methods were based on thosereeommcndcd by Black et al. 096:5).

Soils for pot trials in the glasshouse werebulk-collected from the Sand Flats site, GrandCOl111ty, Utah, in January 199] and imillediatelyspread in a thin layer on a laboratory floor toair-dry. Samples from areas with and \vithoutClyTJtohiotic cover consisted of the surface .5 cmonly. Once dried, soils from each surf~lce t)-Vewere thoroughly mixed Ondlllling the bioticcover for that sample set) to ensure a uniti:lnllpotting mixture. ~ 0 fertilizer ,-unenumentswereadded. Subsamples from each surface type weretaken for subsequent analysis of physical andchemical characteristics. One liter of soil [romthe sample taken from each surface type wasplaced in a drained phL,>tie pot having a topdiameter of 1.5 CI11. Before pots were filled,drainage IHJles were covered \lith a coarse fiI>er-

•glass lnesh to preclude loss of soil. Soils fromeach surface t}lw \Vcrc replicated 10 times inindhidual pots. Pots were immediately placedin a grid on a :vater-tight table in a glasshouseand watered from the hottom with a 2..5-Cl1llaver ofwater that was drained off as soon as soilat the pot s1l1facp was thoroughly wetted bycapillarity. Pots were placed ill a 4 X .5 grid withgJid intersections 30 em apart'. CI)"11tobiotic andblow sand soil surface types were alternated intIle grid. Six presoaked seeds of Sorghum!w[cpclIse (L.) Pel's. were plantpd in each pot on11 February 1991. Pots werp irrigated with t<.lPwater as nepded to maintain nonstressful grow­ing eonditions.

To evaluate thp effects of cvanobacterialgrmvth tllroUghout the rooting z;me, we initi­ated a second trial simultaneouslv with the fore­going pot trials. In that trial \V~ fmed narrow,glass-walled planters with 0.9 liter of the samecr)]Jtobiotic-covered soil. Each planter was 1..5cm wide, 40 em long, and 30..5CTrl deep and wasdivided into two compartments ofefIual size by

. -

79. ]

(J.g.3

0.021

P('rcpnt (oJ(,)

7,H

$-)2,7

~itro~f'n O.OOS

On.;unit: lIlatkr o..'')!

Sand (0/,)

lkaction (pIll

Calciulll 1.449 :3,71;..),

Copper 0.20 0.21

Iron S,I 7,S

Magnesiulll :39.0 .57,0

Mallgancsc 4.6 1.5A

PhospllOrus '):1 - .51.S:....- . ,PotassiUll1 fiq - i;7.0" . I

Sodillm h III

Zinc n.:) O,:}

---- ---

'1'.\ 1l1.F ]. Tlw i ntl1JeIlCl' of c\-anob"ctcria I-CIJ/!eIIUl covcron various ch,lracteristics of the O~3-('m sllrLltt, 1<1\I-'( ofsaud. For comparisoll, the same Ch,II',Il'tcristics for (,olllpa­rabl(' la~-('rs of nt'arb~- saml b'pt fret' of smfi.lCf' growl h hywind action afE' prl'st'llted. 'I'll(' \'ahws for each smfact'("ondition arc all an'rage of 1-/ S<llllplf'S (ullcdcd at threedif!i·rent sites ill soutlw<lstt'fll Utah. Two composite samplesWt'fl' tah'll Oil pach surface conditiun at \Viml \\'histlt'Campground, San Juan Co lint)'. ,111(112 composite sam\JlesWCf(' taken from hlo\\' santi aud IIp,,rh,, soiLs stahilizr·( ]A

cr:lltohiotie cover at two locations in Arches ~atiol1<l1 Park.Grand e(lll1)tv, Utah

Soil samples considered ill Tables 1 amI .'3eonsisted of composite samples of the surface 5em of the profllc taken at 10-12 randOlIl Iy cho­sen spots within each of the follmving surfacf,'conditions: an-'as that supported a well-devel­oped (i.e., >600/c) cover of c)-anohactelia andthe black lichen Col/emrl tenax, and areas onlva few meters away where prevailing winds o'rfoot traffic had prevented crust development.All soil samples considered were collected insouthc<'l,>tcl1l Utah at sites identified in the leg­ends of Tables 1 and 3_

Soil texture was determined using a hy'­dnnneter procedure. Soil reaction \vas takenwith a glass electrode on a saturated soil-water

CYANOHACTEHlA A:\ID CYA~OUCHEI\S 63

a redwood strip 1.5 X 0.5 X 30.5 em inserted atthe midpoint of the planter before adwng soil.The glass walls of one oompn.rtment of eachplanter were covered with aluminum foil toexdm.le ligll!. Planter,.; were drained and aeratec-lat their hottoms via a pettorated tygon tube (4mm in diameter) open to outside air at hothends. These plallters wert> irrigated and plantedwith presoaked SorghulII halepense sc-cd at amte of6 seeds percomp<utmcnt on 16 Fehnlary1991. This trial WiLli replicated 10 times. It wasnecessarv to water these phUlters on altenmtedays with 2.S0-300 ml of water. For conven­ience, tap water was dnmm and stored in aplastic bucket io the glasshouse until needed forirri~ation of planters. This water averaged 8­12<><: warmer than wilter taken directly from tiletap. The cumuillation ofwanner irrigation waterand less exposed soil surfal.'C from which water(.'Ould evaporate resulted ill rout temperaturesfor plants gmwn in narrow planters that aver­aged 3-8.5l>C wanner than those of Sorghumplants from the sal11~ se~d lot ~rown in the samesoil hut wntert=>d frorn the bottom. Temlx.~ratllrcdifferences were greatest immediately after ir­rigation of the pots with eold tap water.

After three weeks cyanobactcri::u growthcovered tlJe glass walls of planters not coveredby aluminum t()il. The cyanohacteria obviouslycompeted with Sorghum roots for essentiallllin­erals. Plants growl I ill planters that receivedlight throltghOllt tile rooting zone were smaller,and their leaves were discolored by reddish pig­ments in eontmst to adjacent plants grown un­der identical <.:onditiom except that light wasexcluded from the rooting zone. A<:, a conse­quence, glass walls ofall planter <_'ompartmentswere t:overed with aluminum foil 3 weeks afterplanting. The foil remained ill place until phmttop growth was harvested foranalysison 29 April1991. Plants gmwn in planter compartmentspreviously illuminated in the rooting zonequickly regained. nor1l1alleaf color and becameindistinguishable in siw from adjacent plantsbJTOwn in compartments with toil-covered root­ing zones. Chemical analyses of pr.mt tissuefrom these trials demonstrated that tissuechemistry from plants that had their roolingzone exposed to light for 3 weeks did not differsignificantly.lc)r any element considered fromthat of plants that had not received light at anytime in the rooting zone.

Plants in narrow planters and those in potsof bottoUl-inigated, Lyanohacterial-enriched

soil ""ere grown from the same sped lot in thesaint=> soil and were propagated in the sallieglasshuuse at the sam(~ time. These otherwiseidentical (..ouditiolls for growth were marked hya strong diflerenl.'e in root tempcmture. withputs averaging --16°C and narrow planters-2IOC. To determine the eflpe! of differeotrooting zone temperature-son mineral cOInposi­tion of Sorghum ahoveground growth, we ana­lyzed and compareu the chemical eompositionof top growth of pot-grown plants <U1U planter­grown plants (see Hesults).

Plant tissue was ovell-tlrieu at 60°C for 12 hI'and then ground in a steel rotary mill using a40-mesh sieve. Samples were stored until ana­lyzed in capped plastic vials. lisS1Ie nitrogen wasdetermined llsing micru-Kjeldahl procedures.Duplicate l.O-g tissue samples from each ex­periUlent<.J replication were digested in it 1:5solution ofconcentrated sulfuric and nitric acid.Content of bioessential clements in the dig:es­t;lte \V;L'i detennillt~(l llsin~ atomic ahsorptionprocedures (Page ct a!' 1982).

The degree of root infection hy vesiculararhuscular mycolThizae and other root sym­hionts such as nhi:ohium hacteria (associatedwith rOots of Lllpirllts) or Bacillus hacteria (as­sociated with rhizosheaths of Slipa hyrrLeTlOides)was detennilleu by microscopic examination ofroots of randOlnly selected plants gmwing inwell-developed cyanobacterial-Collenw crust oron nearby cumparable sites where wind actionor animal tramc (sometimes <u-eas trampled bypt:ople) had precluded growth o[ cryptobiota.\O\'ithin c<leh soil su rface type, plants were ran­domly selected using the <juarter method (Cot­tam 'lIld Curtis HJ50). Plaots 'vere mllectedduring early flowering (early May lfl92) in\ViL'\hington and Grand cuunties, Utah. Bromlls

teclOrtWl L., CnJptantha plerocanJa (Ton)Greene, Cryptnntlw cmssiw?"l'nln (T. & G.)Creene, Festllca octoflora \ValteT, LlIpi/HIS1-msiUfl.S Pursh, and Plrmtof!,o patagrmica Jacq.were <...'Ullected in \Va."hington County.Coleof!,yne ramosissima Torr., ...\.fentzelia afhi­awlis Doug. ex HfX)k., Stipa hymel1oides, amlStreptaJlt/lclla lot1gim!\triR (\Vats.) Ryclh. werecollected in Grand County

Using a shovel, we lifted the root ~"ystcms

from the sandy soih in a block 'md then f",~d

them froUl associated sand by hand. Plant topswere i1l11nediatelyexciserland roots were placedin 75% elhanol in labeled, screw-cap ~Iass

64 GREAT BASI'! NATURALIST [Volul11e m

• ••• 0

B NU0.

• N

.K

A NU0.

6'h6

6 6

o 0

.Fe

PC1

Plant nomenclature f()llmvs \\7eI5h et a1.(l9H7). Lichen nomenclature follows EGan(1987). Statistical significance of differences 1"W­hveell group means \Va.<; determined using anunpaired t-test model (Snedecor and Cochran19671. Significance of treatment effects in Tahle5 was dctennined using analysis of variance(ANOVA) with species treateel as blocks. Per­centage data were arcsine transf(ml1ed prior toanalysis by tlie GLM procedure of tlie SASstatistical package.

Centered, standardized principal compo­nents analysis (PCA) was used to analyze differ­ences among samples colleeted from clllstedsoils and unclllsted soils (Pielou 1984). Varioussoil chemical and physical parameters wereused in tllis analysis, which \vas comlucted usingthe Statgraphics package .

RESULTS

Fig:, 2. A: plot of the first two components of soil samplesfrom l'fyptnhiotically (Ttlstpd ,mel tlllcrusted surfac/:'s atthwl:' diffel"t-'JJt locations {st'(, Tahle J\, Sampl<:>.s from Arch~s

:'\iational Park are indicat<:'d lw triangles <lnd diarllnmk• ••

s<lmplt's from \\'ind \Vhistll:' Calnpgr011nd are represented0: drclPs. 13: plot of weighb for i ntlividllal parameters IIsedto obtain the first hvo ('ompOllents showll in Fi~lm:' 2A (.S('Cte'd for c\planation).

The effed of cl:)-ptobiotic grov.,th on surfacesoil chemistI)' and texture is particularly impres­sive for such variables as organic matter, soil N,exchangeable Mn, and "a",ulable" P (Table 1).Exchangeable Ca W,lS also much higher on av­erage in soils stabilized by cy<.Ulob,lderi<.u-lichsurface gro\\th. As in other studies (Fletcheranel Martin [948, Kleiner and Harper 1972,1977a, 1977h). our results show that soil silt andclay are much greater in soils stabilized by cryp­tohiotic grcJ\\th. The response ofCa, Mn, and Pmay be related to textural differences alone(BI~ck [968), but the increase in soil organicmatter and N is probably directly relateel to thepresence of cl}Ttobiota.

Principal components analysis of the basicdata on which Table 1 is based showed markeddifferences between cnlsted and uncnlstedsoils, even though three separate areas (hvo inArches .\\ltional Park and one at 'Vind 'VhistleCalnpground) were studied. The Hrst plincipalcomponent clearly separated the cnIsted soils ofthe most intensively sampled Arches site fromthe corresponding uncrllsted soils samples (Fig.2A). Clllsted and unclllsted soils of the otherhvo sites wefe also separated by the first com­ponent. Clear differences among sites were evi­dent in the separation that oc'Curred on thesecond principal cOlnponent (Fig. 2A). The per­centages of variance in the data accounted forby PCl anel PC2 were 48% and 22%, respec­tivelv.

PC1Zn

Ca

sample jars. All root samples were taken fromthe surface 1.5 em of soiL

In the laboratory, samples were cleared andstained plior to microscopic evaluation of theamount of colonization on roots. Clearing andstaining were accomplished using techniquesdescribed by Koske and Gemma (1989). Usingtbe adapted grid line-intersect metbod of Am­bler and Young (1977) as discussed in Giovan­netti and Mosse (.1980), we reported degree ofmycorrhizal infection per specimen as the per­centage of root length having developed arhus­cules or vesicles. The percentage of root lengthhaving nodules or rhizosheaths was estimatedocularlv.,

1993J CVAM)BAerEHIA AND CVAN()UCIIENS 65

TAAU: 2. 'TiSSIIP f'lcmenla! wntcnl of plan Is of FrslIIC(I octoflonJ (il dilllillllli\'t> <lllllual) anel ,\kllt::.eUa lIlultiflora hIshort-li\'etll~n~lIniilllll-'rh) Kfl"""n On blo\\' sand ilnd lIearby sanu stabilizeu hy ('yanoh;ldcri;lI·C(Jllr~/l1ll crllsts. The F(~~1f1m

s!llnples Wcr(' taken at 'Vim) Whistlt> Campgrulllul in San Juall COllllt.\: Utah; tht.· MCII/:.difl sa.lllpl(~s \\'t're taken illC...ollrthous~ Wash DwX'S, Grand ClIllllty. Utah. All data ilrl" from Belnap and Harper (in Tl·\"iew). All tisSllf' t'oncentratinllSare exp(e.'t~(,.·das ammmt Ix'r unit dry weight of a}Xl\l'h'TOIll KI\-:wwth ami athK,lwd root tissut' bTl/sill-Ad fret-' ()f .~and. Smnplesize W'dS 5 (or each mean.

~-~-.

Festuca ocloflora Ment.ulia multiflora._~

Soil slIIfll(:e mnJiliull Soil surfac(' <:ouJifion

Cnl,lK II Jacterial Bluw Sigll. CVillloh,tderia!l U!ow Si~ll.

Elerntont lichcn co,\'cr ~al1d dili<.'l'. lichen lu\,cr sand difkr..._. ._.. --

N (%) 2.2.5 1.D.,) • 2.6] 2.18 ••r (%) 0.25 0.14 ... 0.20 0.24 ••K (%) l.H8 1.64 • 3.04 2.fi7 NS"C.(%I 0.65 0..'52 ... 2.50 15.17 NSMg(%) (US 0.13 •• 0.:,3 0.2.3 •ell (ppm) 11.0 lOA NS 8.6 9.0 NSFe (pplIl) 300.3 14\).4 • fi39.8 465.4 •Mil (ppm) fiO.3 74.0 NS 98.2 82.4 NSNa (ppm) 61.5 .SRS NS 77:2 63.4 •Zn(pptn) ·13.0 :J.1.0 NS 22.0 20.0 ~S

·M..~,., »J:Ilir.l"l,,~~·ollfK-"'Oltat p < .tL'>."Mu,,~ "'J:."m<':Int~ <liff"I\·"t It! I' < Ill.... """,,,~ ,;~"ifi(u"t~ cliffi,,,,,,t:tt,, < OOt~I""l"~ U(~ <i~"ili<"'"tt.- clilr""'nL

Plotting the component weights (Fig. 2B)yields an explanation for the separation ofCll1Sted and uncrusted soils. The first componentmost dearly represents the cliHeren(;cs betweenllncl1Isted and crusted soils. Uncrusted soilshave higher values on this axis, which meansthey have higher pH "nd S:lJId content (positivcweights) and lower Ca, Mn, P. Mg. N, and K(negativc weights). Iron and Zn tellded to hehigher in uncnlsted suils but were not a'i impor­tallt in sep<l.rating points on the Hrst axis. Theseeond component most clearly represents site­spedfte differences. One Arches site (triant;lesin Fig. 2A) had higher N, K. "nd Fe (positivelyweighted) and lower P, Mg. and sand (ne~atively

weighted) than the other two sites (Fig. 2B).The prinCipal ("'omponents analysis shows thatdespite some site-specific differences, crl1~ted

soils are generally higher in some essential min­erals (N, K. P, Ca, M~, Mn) than are ullcmstedsoils. Site diHerences were "uso dearly deline­ated in the analyses.

Table 2 clearly shows that c.yptohiotic enlStsdo have asignifiwnt influence on tissue contentofseveral bioessential elements in both Festucaoctoflora and Ment=elia 1I111ltiflvm (Nolt.) Gray.Since Belnap and Halper (in re"ew) show thatsoil textllral difTerenc'Cs are small «10% differ­ence in pen..'Cntage sand) between blow sand

and adjacent sands stabilized by surface growthof cryptobiota, the tissue content differencesseen in Table 2 would seem to be strongly influ­enced hy mi<.:roorganisms on the soil surface.Tissue content of both seed plants was signifi­c"ntly greater for N. Mg, and Fe when plantswere rooted in cyanobactelial-rich crusts. Al­though not "ul diflerences were statistically sig­nifjC<Ult. 9 of 1.0 elements were present in~rcater amounts in tissue of Festuca plantsgrown on eryptohiotic,~lystabilized surbc'Cs; 8of 10 elements were present in greater amountsin tissue of A.fcllt=.elia plants grov.rn on the cryp­tobiotic surfaces. Finally, thc data sll~est possi­ble competition between cryptobiota and seedplants lor P and Mn. We also note that theresponses of Festuca and Alcnt::.elia wereunalike in respect to J' and Mn upt"ke.

Glasshouse trials demonstrate the soil "fer­tilization" effect of the eyanob"cterial-Collel1lacover for growth of Sorghum (Jltbles .3. 4).Differences observed between chemistry ofcymlObacteria-free ,Old cyal\obacterial-Collemn­c'()Vered soih (Tahle 3) are smalL The cryptobi­otic-covered soil b,] slightly more N, P, K, "ndN". Thc soil Iree of cryptobiotic growth aver­"!led somewh"t higher in ea and Fe than thesamples supportin~cryptobiotic wowth.

66 GREAT BASIN NAI'UR\I,lST [VOhlTll(> .53

~--~ -

T\uL1", :3. Cllara('h'ristics of C\'<lllohaC'tcrial-C()//cllla­sl<lbiuzed sllrfact' (0--5 un depth! s'lnds a n<l sands (rmll sallIe

(lepth zow:' onl)' <I few dill aW<I;.' tll,lt \\'l'n" not culonizcd hyC'r)vtohiota. Thl-' lattf'f sauds had Ix'("n llllstahilized by hu­lllall tmmpling for at \('ast a (lecade, Samples ,Iff' [i'OIll SaudFbts, Cram! Count\', Utah. V"I11('s fur (',Itiom ,In' amountscxchange,lbk ill] ,0.\1 .:--JH~C1. These are the soils Oll \\'hicIJSorglwl/llw{(,pcllsl' tisslle ('onsidt'I"f'd ill T,lhlc 4 \\'as grown.

Surbce couditi()ll

Soil dWfilCt!'listic

Te\ tnrc l '7r s ,Illd1Hcaelion (pll)()rganic matter ('Ie.;'

'.!\(t;(,J(:a (Pillll)

ell (ppm)

F(~ ipplll'i\Ig (ppml.\'In (pplll]I' (ppm, av<tihble)K ,-ppm)"\<1 ipplll)Zn ippm)

Sample Sut'

Fr('(' ofcryptobiotic

growth

, ..J

1.000,0382J)70

(), -1-

57k:1.1:jl)

SJ

.S

IIC<l\\

('y.111 oJ Jacte'rial­CO!!ellI(f covcr

617.4

(UJSO.1H]] .S-l-,Sn.2,-Q, 'J

:J,-l­-1-7H']

111.9

crust-free soils on illfection of roots bv rhizo­symbionts slim\-' that 6 of 10 species e\idencedinfection lw root svmbionts (Tahle .5). Of thosespecies wbose roots showed some inf<x.:tioll, 4 of6 were annuals, and all infected alllll1<lls werecolonized bv vesicular arhuscular nlvcorrhizae, ,(VA\Ii. The dL'gree of infection was ahva:"sgreater for plants grown in cyaJlobacterial~

Col!enul enlsted soil. Seedlings of the shrnl>ColeoglJllc m IIwsissil1l{/ snppOited heavier VAl,,1infectioll than associated annuals, and the rpla­tive amount of root colonized bv the root sym­biont was over three tinles' grpatcr ,,:henseedlings (,lllerged [rom cll.1sted soils. Roots ofthe perennial grass Sti}Jfl hymelloides alwaysdeveloped rltizosheaths on hoth t)]1es of sur­fuce, bllt tIle degree ()f sheath developmcnt \vasgreatest (m cnisted snrfaees (Tclble 5). The over­all cfh,ct of cJ1Jsted soils on root illf(~ction bvsymbionts \vas positin: and statistically signifi­cant (hble ,5),

D1SCliSSIO'i

Civen the srnall physical and chemical differ­enc('s hetween soils (Table 3), the oftcn signifi­cant differences in Sorghum tissuc chemistrywhen grown on those soils (Table 4) were llIlex­pected. )Jitrogen and Zn were taken up insignificantly greater amounts on the eyano­hacteriaJ-Collelllfl-enriched soiL prodlll'ing(wer:3..5 tiIllCS more top grm\th than soils withno cr;ptobiotie growth (Fig. 3). The elementsCa, l~ Mn, and Na were present in signiHcantlygreater concentrations in tissue of plants grownin the cryptobiotic-free soil. These data suggestthat (1 ) I\ or Zn (probably the f()fJner) is limitingfor plant growth in the test soil and (2) there is\igorons competition between the cyanobacte­ria and Sorghum for Ca, E .MIl, and Na.

It also seems clear that nutlient uptake bySorglwm was strongly aH(x~tcd by tenll)t'ratllrein the root zone (Table 4), Nitrogen, C,l, !\lg, Cu,and Na \\-'Cre taken up in significantly largeramounts from the evanobacterial-CollcHw soil,at the higher temperature. Increased telnpera­tllre seemed to intensi~v competition betweenthe cyanohaetelia and Sorghum for P, Fe, andMn,

Preliminary data coneeming the influence ofeV<lnobacteIial-C'ol!eIlUl CIllStS and associated

The results presented prO\,ide strong sup­port for the h!1)othesis that cr)ptobiotic soilsurface em'ers (at least those rich in cvanobac­telia) have Significant effects on uptakc' ofbioes­sential elements by associated seed plants. rnthis study we have considered only species (ordevclopmental stages) with a major pOItion o[their root svstem distributed in the snrbce 5 emof soil. The cr:rtohiotic snrf~lces appear to con­sistently enhance uptake ofsome elements (e .g.,N, K, Ca, :\-lg, anrl Zn), and to at least occasion­ally reduce uptake of other essential clerncntssuch as P, Fe, ~ln, and :\'a (Tahles 2, 41. Tllosceffects are apparclltly pmtially explained hy en­richlnellt of soils (increased "availabilitv" of es­sential dements) bv cvanobacterial-C:ollelllflcrnsts (Tahle 1,3), 11}: elc~'ati()n of soil teInpCfa­ture during cool seasons when moisture is rnostlikely to he readily available {(n- plant growth,and by greater likelihood of root colonization bymycorrhizal fungi and other root symbionts at. , . .sites that are stable enough to support well­developed cryptobiotic cnlsts, Onr preliminaryreslilts also suggest that cyanol)(lcterial-Collelllflcrusts Illav result in an enhanced availabilitv ofcertain elernents through accelerated df'cOl~po­sition or production of chclating compounds(i.e., ~-ln and P in Tahle I; eu, N, !\lg, and Zn inTable 4). Taken together, the data in Tables :3and 4 sllg,t;est that the enhanced uptake of~ hy

19931 (;"A;'V()RACTEHIA AND CY:\NOLlCHE:\IS 6i

T,\1I1 ~.; -i. Tissue chemistr: of ahovt'v;rulllu] g;mwtlt uf Sorghllllllwkpf'11Wl grown in c)'~lllohadf'ri:lI-C"lIel1la free S~llld Hl"

sand hc'ad'" con'red (0\'('1" ~(J':{;) b\" (Y<llluhaderia ami COI/CIlIiI ill pots v.,eHcu from tht> hottom. AI.~() Sh(JI.\'1l is tiS51lt'dlt'lllbiry ,;rSor;f.lwlIl ~nl\\-ll i •• tll(' ~:1I11~ c)mlObaderial-CoIlPlIUl-slahili'lf'd .<i;lIlIl ht'ld in decp, narrow rectaugular planters.A\'Crdgcs fuUO\\'('d by the same' !QI,\/{·n:a."f' letter do Hot differ signifkantl).' (p < .0.'5).

Soil and polliJ)~comlitiolls

4.N a 5.5 a

}ojfi.:l it 51.H a

64.-1 a 53.0 h

11·f.:1 a 32.6 b

15.7 a 22.6 b

1.3 a 4.8 b

Tissue (;ontent of

N

P

K

ea

ellFe

Mn

"Zn

Sorll;lllllll ),;t-I(Vpot(g-lop growth only)

No. of l"('plicr,tions

Av~. rootilll; Wiletemp (C)

evanoh:U:h-'riaJ- ffeesoi I in poLs

OJ;7 a

0.1·') a., ....., .,_,'::"CJ ,

0..'')7 a

0.J5 a

10

-IG

C\,lnoba(1erial soilill lXlls

COllt'cntration (% dry \\t.)

(l,7H b

0.12 h

2.4la

0.44 h

0.14 a

COIl(:cnlralion (ppm c1J}'\\1.) -

10

-lG

C\~lIloLJildcrialsoilin 1l;lrrOW plankr..

--

!.41c

0.09 l'

2.27 a

0.64 a

0.26 h

8.s h

2(.6 b

47.1 c

50.7 c

23.9 h

"'-21

plants grown on crusted soils must include aconsiderahle amount of N fixed simultaneouslyby ~yanobaeteria in the culture or N rele'lsed hymicroorganisms deL'Omposing the tissue of cy­anohacteria grown during the experiment.Orihrinal differences in soil were too small toaC(.'Ol1nt for tlw large rliflerenCts ObSCIVCd inplant uptake in the two cnltures.

Other studies have shown results similar tothose rep<llted here. Marble (1990) demon­strated that scalrin~the surface 1.0 em ofc-yano­bacteriaJ·GolLemlJ cmst around rosettes ofLepidiwrI numtanwn var. montallwn Nutl. twomonths prior to Howeling signifkantly reducedahovegr01md plant weight and tissue content ofK, Na, and ell at flowering time. Several otherelements (Ca, M~, Fe, Mn, and Zn) were alsolower ill tissue of plants around which the cryp­tobiotic CllJst had heen removed, hut those dif­ferenccs were not statistically signifieant (p >.(5). J. Belnap (unpublished) has ,malyzerl planttissue for the annual Streptantludln lotlf!,i-rostris(Bmssicaceae) and seedlin~s of CnIP.ngyne m-

mosi<;sima (Rosaceae) grown on weB-developedcyanol>acterial-CnllellU1. crusts and on emnpara­hIe soils without sud} enlsts. Her data sl10w thataverage plant sii'e or both species W<.l.'i signitl­calltl)' larger on enlsted soils, and tissue of bothspecies (xmtained Significantly more ,Ca, Mg,and Cu per unit dry weight when F;mwn Oll

cryptobiohc surfa<.:t::s. Belnap also has unpub­lished leaf ehemistlY data for adult shruhs ofColeugyne ral1wsi~sinUl and the long-livec..l,woouy-routed h(-:rh LepldiuUl I1wulanuUl var.jOlles;; (Rydb.) C. L. Hitchc. As one might havepredicted, adult plants of these species showeJno enhancement of essential mineral uptakewhile growing on cyanobacterial-rich cnlsts.The vast majority of their feeocr roots lie wellbelow thuse portions of the soil profile that areinflucnceu by cryptohiotie Cl1Ists.

As seen in Table 4, Sorghum growth wasseverely limited 011 soils free of ()'anobac(crial~Colle/)~I inoculum (Fig. 3). Nitm?;en seemedinadequate in sm:h soils to support healthygrowth ofSorghum. In faL't:. growth \-V'd$ so minimal

68 GREAT BASI" KATURALIST [Volume .5:3

r

•

• •.,

•

•

. 3\

•

.' ,, . , .• • • •

· I.

•

•

••

••

, . . ,,. . '

" ,, .- "

•

,

•

, ,

, : . ',-,.,:~~~"'~ '" .. -

"'j . ""'-.'

•

•

," .

•

•

,

•

•

•

•

•

•

Fig. :3. Sorghum Iw!ef!cllse plants after 60 day.~ of growth OIl a soil mixturf' cOllsisting of thE" surface .5 em of a sandcolonized by a IW<lYY gnJ\\ih of cyanobactelia and Collelilo tenax (RJ and tlw ,~aJne depth horizon from a llearh~' site keptfree of sllch growth hy f(Xlt traffic (Ll, See T,lble:3 for characteristics of tlw two sous.

e .

in those soils that: the species would probabl:'never have flowered and set seed on that: sub­strate. \Vhen Sorghum was gro\vl1 on a soilenriched hv inclusion of the c.vanobacterial-

• •Collema inoculum, growth was much greatereven though only ~ and Zn occurred in signifi­cantly greater concentrations in the tissue (i.e.,all other essential elements occurred in higherconcentration in phmts groyvn on clJvtobiota­free soil). These results suggest that eventhough ey,mobacterial-rieh crusts sometimesenhmlce uptake of several essential elements,plant growth (at least Sorghum growth) W"lS

limited by N (and perhaps Zn) only on the soilconsidered.

Evidence does exist that the cryptobiotic sur­face layers (especially those rich in cyanobacte­ria and/or cvanolichens) have a heneficial

"influence on seedling establishment under fieldconditions. Harper and Marble (1988) summa­rize the results of an experimental seechng offive species onto plots dominated by cyanobac­telia and the lichen Colle11lG. At time of seeding,ronghly half of 83 plots were randomly chosen

for a treatment in which the surface 1.0 em ofcl;vtobiotic Cll1St '\vas scalped away. Each spe­cies WtL'; seeded on 10-92 r<mdomlychosen plotsthrough a template at .32 locations per 1.0-m1

plot. After the first gro\'\ing season four of thefive species had more seedlings on plots wherethe cl)ptohiotic Cll1st was left intact; in totalthere were slightly over three times as manyseedlings on the average cmsted plot as on theaverage scalped plot. After three years a largerpercentage of the seedlings survived on crustedthan on scalped plots for a1l five species tested.

In Kevada. Eckert et a1. (1986) planted seedsofsix species on three types ofsurfaces: (1) thosecovered by a sparse plant litter cover and he­neath a shrub canopy. (2) polygonal patternedSlllfaces covered by a \ig;orous grov.th of cr;p­togamic cover, and (3) polygonal pattenled sur­facE'S with little cryptogamic cover and ininterspaces between shlllbs as was surface type2. Five of the six species tested established 'LS

well on the type 2 surface as on other surfacetypes, or even better on t;pe 2 surfaces; theother species established best on type.3 surfaces

1993] CYANOBACTEHIA Ar\D C;YAN()!,IC:IIENS 69

T\BLE S. PCfCl'Iltap;e of" the root kn~tll COlolli/.e(l b.v rhizosyJllbiulltS Oil cyanobacterial crusted or comparahle soilswithout a biotic CfUSt. Hoots of fln' randomly chosen plants wefe l'x<lmilled per sj)('cies ami soil surfacf' l~lIlditi()ll. Theeffi.·ct or cyanobacterial crusts (treatment) on HlllOllllt of ('o]ollizati011 of routs !J\" rhizosyillhiollts is Siplific-dllt1y positiw'(F-\~d1\(' for Al'\O\'A ==" 7.D4.p < .OJ, dJ. = ] 1. F-vahH' for specit's cf[Pc!SW,LS 0'.:26 (p < ,01, d./". = 5).1'!rmfago patagoninlwas omitted fro!ll the lattt'f <lnalvsis.

~- ----

Associate 1Ilicrolw '\10 biotic nllsl CV<llloha{'tt'rial ('nlst----

th/!:o perC("llt of rnot length colonized

Hnmllls lectorlll11,

Colf'0I!.Yllc ralllosissillW-

Fest1lca ()cfll[lora

Colorado Plateau

\Vashingtotl COllllt~

Lllpinlls jJusilflls

Plantago }Ja(agonica

Stipa !lyme/wides

VA IlI\TOf.1

\ 'A Jll\cot".

VA lllH·or.

Rhi::,o!JiulIl

VA Ill\Tor.

Hhizusheaths3

,J

.5

T

"22

'\10 s<lJ11plf-'

4fl (all weak)

1

4

:34

1:3

100 (()(}[j, slrolla),"Hoob "r II", I(~]""i"g ']"'C'''' \I"'rt' 'l"di,'<1 Ionl nn llIH'orrloi"K' "n'" "h""T(',) C'''!/IJtrilltl", l'/cn~YII,,!,a. C'VI'/aIP/IUi cmwisqlfll". ,\l""f~dill "I!,;rwdi.<', andStl"I'f,/{wl/rril" IUIII!.U",\'1 ,'is, .-\11 of tlws.. "1"-"' 'i," "n' ")(1 III als.'\",,;,," lar "rl "",-"I" r' n1\1'In] ,i /,",, ",', .:: III h- ",(., III n~., "n.' "'~ 11'<'" I ,nl"lIt "'K']OC"' Wt'r...."", "i,,," 1 f,,,· (hi, 'pr'(·k-,.'Hh iz, ,,<I\('al),.' an' (~"N' la"g] ("' "fr<~,1 I,,,ir,. l'..II, "1',, 1>",-;1111.\' 1" ,",,,,,/xll-1 ikt· 1"oct, 'ril Ill' ""1'"1,],, "f" fh"li",,- :",d "dl""r.." f '" ,,,) !<,,,in, i\\ ',,11,1<" 11 a",] Pr"ll I~J" I '.

and did poorly on type 2 surf~{cps. Lesica andShelly (1 fJfJ2) reported that cryptogamic soil sur~face cover appeared to increase sunival of es­tablished plants of Arahi.') feclt/ufa Rollins(Brassicaceaf';) in Montana.

The forpgoing reports suggest that estab­lishment and survival of seed plants native toarid lands may often be enhanced by CI}lJtobi­otic cover on soil surfaces. As Harper amI Mar­ble (1988) show, several sciL'ntists havesuccessfully used inocula of cyanobacteJia toincrew·;e establishment ,1Od growth of agliclll­tural crops in various parts of the world. Accord­ingly, ohservations of positive interactionsbet\veellcyanobad{>rial-lich C111stS and seedlingestablishment and growth in natural arid landenvironments are not surprising.

Although the influencp of cyanobactcrial­rich soil Clllsts on essential llIinpral uptakp hyassociated sced plants appeared to be stronglybeneficial for N onh~ there is reason to helievethat enhanced tissue content of N, Ca, Mg, Na,and P may be beneflcial to 'L'isociateu herbivo­rous and granivorolls animals. Robbins (I9H3)notes that increased dietary protein consistentlyhastens growth and onset of reproductive ma­turity in herbivorous animals. Cyanobacterial. - -crusts consistently incrcaseu protein content ofassociateu shallow-moteu seed plants and seed­lings of deeper-rooted plants in this study (Ta­hIes 2, 4; Belnap personal communication).

Amnann (196.5), AUlllann and Enl1en (196.5),Belovsky (1981), and Robbins (19&,) suggestthat sodium in plant tissues is often inadequateto maintain healthy herhivores. Robbins's

•(19H:3) rt'\,iew of dietary sodium requirementsfor animals suggested that diets \\-ith less than0500 ppm sodiu m will eventually result in poorgrowth or death of anim..Js. \Ve note that cyano­bacterial crusts always enhanced phmt tissuecontent of Na in this study (Tables 2, 4). In omstudy, however, even plants grown on crusts didnot ('ontain the recormnended minimum con­tent or Na. Thus, animals must resort to local"mineral licks" to obtain aue(luate Na. Such lickslllay he widely spaccd on sandy uplands such asthosf-' sampled fiJI' this report. Small mammalssuch as the granivorolls lwteromyid rodentscommon Oll deserts considered here mav heespecially dependent on Na in plant tissue, ~incethey defend slllall territories that would rarely

• •include a lick where supplementar)' mineralscoulll he acquired. III sl1cll cases, increased tis­sue content of Na in plants grmving on cyano­bactelial-lich surface crusts lTlav be of critical

•importance to associated heteromyid rodents.

Hobbins (1 Y83) considered that "calcium de­flciencies arc prohahly the major mincn.J proh­lem encountered in captive wildlife." He notedthat Ca and P arc major eonstitnents of tIl('vertehrate skeletal svstem. In lllatllre anilnals,90% of ea and 809{.:of P occur in bone, which

70 GREAT BASI'" N.'TlJRALlST [Volume 53

has a Ca:P ratio of about 2; 1. Since these ele­ments are so intricately associated in bone, theyarc often discussed together. Birds use Ca notonlyin bone bnt alsoin eggshells, which are 98%CaCo.1 and less than 1% P. Osteoporosis is re­lated to deficiencies in ea and/or P in the dietor to major imbalances in their presence in thefood base. Osteoporosis has been reported forfree-ranging carnivores in Alaska, for reindeeron lichen-dominated ranges, and for the deserttortoise, a herbivore, from the wann deserts ofsouthwestern Utah (Jarchow 1987, Robbins1983). Carnivores may be especially prone toosteoporosis since flesh contains little CalcilllTI.

Osteoporosis in the desert tortoise is surprising;Jarchow (1987) considered the disease to he·aprincipal cause of death for the tortoise in Utah.He found the onset ofosteoporosis to be prema­ture and pathogenic in the animals examined.No disease could be shO\vn to be associated \\-1thosteoporosis; thus Jarchow (1987) concludedthat the condition was caused by dietarv defi-, ,ciencies. Since the principal food plants takenby the tortoise in southwestelll Utah (Hansen eta!. 1976) do not appear to hc deficient in Ca(Jarchow 1984), the limiting element is probablyP Jarchow's (1984) and our 0\<71 data (Tables 2,4) both show less P in plant tissue than is con­sidered necessary by Robbins (1983).

Since growth on cryptobiotic crusts has beenobserved to increa.<;;e plant tissue content ofP inFestuca octoflora (Table 2), Coleagyne ranwsi,­sima seedlings, Lepidium montanum var.jonesii(Belnap personal communication), and L. rrwn­tanum var. nwntantun (Marhle 1990), and tohave no effect on tissue P content in Streptan­thella longirostns (Belnap personal communi­cation), it seems possible that cryptobiotic crustscould affect dietaIV intake of P by the desert, ,

tortoise. Since our data show that P is also occa­sionally lower in plants growing on cyanobacte­rial crusts (e.g., Mentzelia in Table 2 andSorghum in Table 4), more research is neededto detemline whether widespread coverage ofcryptobiotic crusts would increase or decreaseavailable P in the diet of the tortoise. Inasmuchas this species is formally listed as eudangeredand does appear to be susceptible toosteoporosis in nature, it is important that theinfluence of cryptobiotic cnlsts on nutritivequality of tOltoise food plants be better under­stood. Cvanobactericu-rich crusts are known tooccur i~ southwestern Utall in habitats fre­quented by the tortoise (Brotherson and

Masslich 1985), <md their importance in thoseenvironments is also known to have been se­verely depleted by uses imposed by Europeanman (Anderson et al.1982, Callison et al.19&5).Restored vigor of those cnlsts may improve die­tary quality for desert tortoises.

Although Hobbins (1983) concluded that Mgrarely poses a dietary problem for herbivores,Gmnes et al. (1970) and Fairhoum andBatchelder (1980) suggest that less than 0.25%Mg in the forage base puts nnninant .animals atrisk for gnL<;s tetany, a nutritional disease result­ing in vasodilation, hyperinitability, and muscledamage, and possibly culminating in paralysisand death. Magnesium is an essential elementfor proper bone and tooth fonnation and is animport<mt enZ)11le activator for all animals. Ah­sorption of Mg by the digestive organs is appar­ently inhibited by high levels of Nand K in theforage; thus, grdSs tetany is most often observedwhen animals are feeding on lush spring grmvthof grasses. We note that Hansen ct al. (1976)found that annual grasses make up 68% of thedesert tOltoise's diet in southwestern Utah. Ourdata show that cryptohiotic crusts consistentlyenhanced Mg content of tissue of associatedplants (Tahles 2. 4). We snggest that the influ­ence ofcryptobioticcrusts on Ylgin plants eatenby the desert tortoise merits fUlther attention.,

ACKNOWLEDGMENTS

This work wa..<; partially supported bv a sub-• • •

contract from the Shrub Sciences Laboratory,Intennountain Research Station, U.S. ForestService at Provo, Utah. The contract was fundedby the u.s. ArmyCorps of Engineers , Construc­tion Engineering Laboratory, Champaign, Illi­nois. \Ve thank J. Belnap for permitting us toreview unpublished data from her files. ThePCA analyses presented in Figure 2 were per­fanned by Jeffrey Johansen, John Carroll Uni­versity, University Heights, Ohio.

LITERATURE CITED

ALLF.:". :\1. F. 1991. The ecology ofmycorrhizae. CambridgeUniversity Press, Cambridge, United Kingdom. IH4

pr·AMBLEH, J. R, A1'\l) J. L. YOuNG, 1977. Techniques for deter­

mining root length infected by vcskular-arbuseularmycorrhizal'. Soil SC'ienec SocietvofAmerica Proc'et'd­ings 41: ,5.51 ~5S6.

A:\DEHSO!\. I). C., K. T. IL-\RPF:H. ,\:\,I) R. C. HOL\lGH~::\'

1982. Fact()rs intlilencing development of cryptogamic

19931 CY;\:\OBMTEBJA AND CY;\I\OLICIIE:\lS 71

crusts ill Utah deserts. Journal of Bange ~Lmag('lrlf:'llt

:1'5: J80--1&:5.Au "-1.\ J\" G. D, 196.5. :\Iicrotinc ahundall(,(, and soil s(xliulll

leV(·L~, JOllrnal of ~Ialllln;\l()i:-.ry·Hl: <')84-604.AUMAN:'>. G. ])"A!><]) J. T f<>,ILEN. 196.'). Hclatio)l OfpOpll­

latioll dCllsity to sodium availahilitv and sodium splec­tion by miCfC)tille rodents. :.iatnrt' '108: 191).-] sm.

HrCu)"sKY C. E. IDb'l. A possible poplllatioll fespolls(' orIlHXJSe to slX!illlll availabilih. Journal or ~\'Iamlllal(>g:'

62: G:]1-68.'3.BE L:\ A1'. j" \\1) K. T 11-\ HPI<: I{, l~J9.3. The inllul'lll'l' of

cryptohiolic soil crusts on ..If'lllt'ntal exmtell! of tissllt-'of two dpsprl seed plants. III JT\iew.

BETIIU;;-.;F,\!.\\Y G. I" H. A. E\"A:\S'\Nll A. r.. !.I-:SPFH­

\'\CE 1985. f\fvcorrhizal cololJizatio)l of creslhlwheatgrass as infl lWllccd hy grazing. AgroIlOlll~·.IOll rnal--. ")'~'l-').,,~I I. ~.>- _,Xl.

BLACK. C. A, W6H. SOil-plant rl-'latiollship. 2nd cd. Julm'Vill'\' and SOHS. 111(;.• :t\ew York. 7~)O pp.

BLACK C. A" D, D. E\'\'\s . .J. L. \\iIlITE L. E. ENS\II'i­(;~~H..-\'\D F E. C1.AHK. Ens ]965. Methods of soilanalysis. part J: Pllysil'al and mineralogic.J prolX'rtil:"s.inclll{ling statisti(;s of llleasn]"('llll'ut awl s<lllipling,American Society of AgronOlll\ Inc., 1-fadisOll. \\'is­cOllsin, Till pp.

BOOTII \V F. UJU. Algal:" as piou{'crs in plant SllcccssiOJland their importance in erosion coutrol. Ecology 21:3H----46.

BnoTll F I{SO N . .J. ])" .\\"D \V J. MASS LI ell 1~JH.'5. Vt'gdatiol1patterns in l"l'latiOll to slope position in the Castle Cliffarea of S(Hltl tern LJ tah. Grcat Basin l'\ atu raust 4S: ,')3.5­541.

CALI.ISO,\, J.,.I. D. BHOTlIFHSO,\ A'\D J. E, Bo\\ NS 198.'),The effects or fire on the blackbrnsh (Colcog!Jur' m­lIlosissinw) COIl1I tlullit~· of so\\thwf'stf'rn Utah. .I ournalof Hangl:" Matl.lg{'nwnt :31): !5.3,5---,53S.

COTTA.\L G., A"ll.J. T CL!HTIS W56. The nse of distancellwa.~llr(-'s ill phytoSOl'ju[ogical sailipling. E(dogv :~7:

4,'51-4bO.Fj;KEHT. H.. E., JIL F. F. PETEHSO.\'. 1-L S. ~h:UHJSSE. A'\D

J. L. S'n: I' II ).;:\ s 1HHr-i. Effccts of soil-surface n1orphol­oK'" on ellwrgence and sllrvinll of" seedlings in higsagt-'brush communities . ./()llmal ()f H<lllge Mallage­ment 38: 4],1----4:20.

Ec-\''\. R. S. I\)H7. :\ mih checklist of the lichcll-forHling.liehenicolollS and allipd fungi ofthe contillentallJ nitf,dStates and Canada. Brvo]ugist 90: 77 17,3.

FAIIUlOU{\', \f. L .. ,~,\I) A. H. RncIlFLDFIl. 191)0. Factorsinnuencillg maglwsilllll in high plains forage . .l0umal()f Range ManagelllCllt :3,1: ·J.3.5---43S.

FLETClII'; H. [. E., M\ D \V. I'. r>.l-\ HTI '\. 1945. Sortle pffects ofalgae mi(llllolds ill the rain c1"\\sl ofdesert soils. Ecology29: fj,')-100.

FH!EI)\IA\"'\. E. I.. AJ\D ~1. CALl:,\. ]874. Desert algae,!ichens. and fuugi, Pages 1(-i.,)-2 12 ill G. \V. BrowlL.IL.

ed.. Desert biology. Aca(lt>rnje I'n"ss, New York.FHJTSCIJ. F. E, H:J22. Thc tern'strial algal". Journal of I<:col­

0!0' 10: 220-2.3Ci.FL:LLEH, \V II.• R. K C,I!-.-lEHO'\ ..I'.;D N, RI1C:\ ]\:)60.

Fixation or nitrogen in <1('Sl'rt soils hy algae. \Vorkingpapers of .~('n'nth (ungress of Internation.J Society ofSoil Science for 19 August. Madison. \Visconsill.

GIOVAN'\V1TI \1., A\D 13..Moss!':. 1980. An cvaluation oftc:dmiqlles frJr measllring vc'siell]dr-arbllscular m~'cor­

rhizal illf(-.ctioll in roots, :t\ew Pl1\tologist S4'lS9~'500,

GLASS. A. D. M. 1989. Plant nutrition: an introduetiull toCllrwnt eOlll'{'pts, JOlles and Baltlett PlIhlislwrs, Bos­ton, ~IassachllsettS.13--l pp.

CJ\L'NLS J) 1..,1'. H. STOUT, A'\J) I. H. 13I\O\I''\EIL W70.Crass It'tan~' of lllHtinants. A(lv,llIcpS in A.grOlH))llY 21::3;~ 1-:174.

lL\"s~::\ H. M .. M. K. Jou:\SO\;, ;I'\l) T. H. VI\lDE\'E'\­

DEB W7n. Foods or the desert tOltoisl' ((;O/J!Wf'IIS

tI;-'/l'\'si;:i) in Arizona ,1Ild Utah. lleIJl,:tulo'f;ia ,12: 24"1­2:31.

I L\ \l PI'.Il. K. '1'., .~'\ l) J. B, \hIlI\LE. [UHf-\. A rulL- for nOJl\'as­cular plants in mar lug('lilenl of arill and sf'llliari([ range­lands. Pagps I:3S,](i\:l ill 1'. T. TllelleL cd .. VPgt~tati()1l

sdeneE' applieatlOlls for mngt-']aJld allalvsis anlllliall­a!4cllll'nl. KIllw(-'r Acadpmic l'uhlislwrs. J)onlrt'dlt.

'/-I11.Cl1o\\·. J. ]9H4. \'f'tt-'rinary managt'lllellt of the dt'sertturtuise, (;0}lhcrll.\' ai!.{/ssi::.i, at the Arii',ona-SonoranDest'rt Museum: a rational approach to dipt. l'agl'sWI-94 ill \1. W, Troller, e(1.- Proccetlillgs, 19M SY1IIPO­SiUlll of the Des('rt Thrtuise Conncil. LOll,>!; Beach,Call1()rnia,

____. 1!:lS7. Heport Ull im-('stigation of desert tortoiStlIIortality on the B('m'er Dam Slope, Arizona and Utah,Unpublished repmt to Arizona Strip ami Cedar Cit:·Districts. Blll"t-'.lU uf Land Malla~l:-'lTwnL St. George,Utah.

KLYI '\ 1':11. E. F,,\'\lJ K. T IL\I\1'FH 1\-l"l2. Em-ironllleut alldC0111111\mity orgallization in grasslands of (:<ln~"()JJlandsNational Park. E(X')lo~fY ,')3: 1!J-J-.309.

__ . 1877a. Soil prolx'r!ies in wlatiOJl to CT)ptog:mue l-,,-nJ1llldwv('r in c.tnyonLllld..... :t\ational Park. .lo\lrnal of Harlgl:"Managenll'nt ,10: 201-10.'5.

____. I~J"l7h. DeclIn('!ll'l' of fOllr major perennial grassesin relatioll to vdaphic LK'torS in a pristinc communit'v.Journal of Range Management ,30: 286---1.')8.

KOIDI·: R. T,.\'\D H. A. \lo(),\·EY. 19f-J7. Spatial variation inin oClll (1111 potential of ves LClllar-arhllscl dar I nycorrlliz,1Ifungi callSi'd h~' formatioll of gopher Illounds, l'\('wPhytolop;ist107: 17:l-IH2.

KOSKI':, H. E, .\'\1) J. X (;1-:\1\1\. I~Jh'\-J. A lllodified pron'llllrf'fur staining roots to detect VA lll~·corr]lime. \-Iyco]ogi­callksearcll 92: "'H()--4KH.

L-\,!\'(;F, W. W7.... Chdating af';ents and hlue-green alg;'le.Caumlian .lOllrtlal of ~\licrohioIOhry20: ],31I-l:~2].

___. nJ"I(-). SpeCl1l:1tio11S on a possible ('ss('ntial fUllctionuf the ~{-'latillous sl1l'ath of hhK'-grepll algae. (:,lJladian.lonrnal of \Iicrohiology 22: 11f-J 1-1185,

Ll·;n:\'I-lJ-:. 1\-'1, W().!. Extracl'1lnlar products of alg:l('. Pages337-367 iii D. F. Jackson, cd., Algae awllllan. I'lenlll11Press, New York

LESIC:\. P.. .~!\Il.l. S. $lll'.l.LY 19~J2. Effects of (T)l)togalllicsoil crnst ou tl](' popn1ation dynamics oL1rahis!ecli mla(Hrassic'l('''':W), A11I('rican MilUalld Natllr'llist 11H: '=:;:j~

60.\1M:GHEC()H, A.1\ ...\\;1) D, K,/OII'\SO'\ I~J71. Capacity

of desert algal ernsts to Ax atmospheric' nitrogell. SoilScienc'(-' Societv of Anwric.l l'roe('('dill~s :lS: 1)4J----H44.

MACKE:\ZIF 11. j",\\"D]1. W. PI';,\l\so:'-! ]979, Prclilllinan. .

stndks Oil the potf'lltial usc of algae in tile stabilizationof sand \\'a~tr"s amI wind blow sitlutions. Hriti.~h Pllv­colo<t->'y Journal 14: 126.

MAlIBI,F .I. R. 1990. Haug('land microphyticcf11st Illanagp­llwnt: distrilllltiOll. grazing illlpads, an (I lllineral lll1tri­('nt relations. Unpuhlislwd dOdoral Jissertaliol1.Brigham Young University, Provo, Utah, 7h pp.

MAYL,I,\]). H. F,.nl) TIL \L..c1'\TOSIl. 1!:l(')(-i, A\,.lilahilit\of hiolo:..;i('allv fixed atlllosplwric nitro'-'('n-I.'5 to Iti<T!Jer'. . M ~

plants, J\atnre 100: 411--422.:\l-\YLIN!). H. F.. T I/. M'\Cl'\TOSII. M'm \V. II. FI'l.U:H.

19hfi. Fixation of isotopic llitrogell in a scmi-arid soil

72 CHEAT HASIJ\ NATURALJST [Volume ,'5,']

by algal crust organ is illS . Soi1Sciellce Socict}' of Amef­

ie-a Proci'f'dings 30: ,56-60.Me K ,,'I GilT. K _B, J~)SO. Factors illflm'>'llclllg size lind h)ph<1.1

pigmentation of soil microfungi populations: a stlHl)li'om !-,')'psiferous soils of a Utah desert. Unpuhlished1llaster's the.sis. HrighauJ Young University, Provo,ULl.h. 21 pp.

1'\(:1,:, A. L., H. II, \llr,LF:H..\'(1) D, H. Kn::\F:Y, EDS [08:2.

r....leth{xIs of soil anal)'sis. part 2: Chemical and micro­biological properties. 2nd cd, American Scx·jet) ofAgrOn()Ill:~ Inc., J\.bdisoll. \Viscullsin, IE,\) pp.

PEIU\\, D. A., \1. P. A\IAk\:\TJIL's, J. G. HOl{CHI<:l\s. S, L.HOI1C1H:HS, .-\'\0 H. E. BHAI:\El\l). 1989. Bootstrap­ping in L'COS)'stel11s, Bioscience .'39: :230-2:li,

PIEiOU, E. C. I L.\.S4, Tlw intelvrdatioll of ecological data.John \ViI",;: and Sons, 1\ewYork. 26.3 pp.

HOBH1'\S C. T. 1*>,3. Wildlife fecding ami nntritiOll. Aca­(lPlllic Prpss, Inc., :'\('\\' York. :34:3 pp.

ROr-.I\"F.Y, E. \'1., A. W,\I,l.,\CF..\'\D R. H. HU\"TFH 1975.Plant reslxlllsP to nitrogen fertilization in the northernMohave Desert and its relationship to water mallipnla­tions, Pages 232-243 in 'J. K \Vest ami J. Skujins, L'ds.,I\itrogf'n ill desert ecusystcms, Dowden, Hutchinsonand Ross, Inc., Str01J(isburg, PcnTlsyl\·aJlia.

Rye l! F HT. H., J. SK U jl \" S D. SoHE \" SE '\ __\:'\J l) D. POBC E 1.1.\,

1978, Nitrogen flxation b~' lichens and free-li\'ing llti­(Toorganislm in d(·serts. Pages 20---30 ill N. E, \Vest and1- Shijins, {'Lk, "Jitrogen in desert ecos.\stellls. Dow­den, Hutchinson and Hoss, fnc., Strolldshllfg. Pennsyl­vanHI,

Se II LF.S I' (; F. H. \ v. H . .1!*J 1. Biogcochemistr): an aualy;iis ofglobal change. Academic Prf'ss, ~ew York. 'H3 pr.

SllIFI.DS. L. rV1., AI\D L. \V [)L'Hl-lFl.L. ]964, Algae in rcht­tion to soil fertilitv. Botanic,ll Heview 30: 92-12R.

SKUjl'\S J .. "'\D B. KLLJ\E'" 1878. Nitrogen fix,ltioll anddenitrification in arirl soil Ctyptogamic erust mict"(wll­vironnwllts. P;lgc~ ,54,'3~5S2 hi \V, Knrmbein, cd.. En­\'ironll1ental bjL~l'(xhemistry and ge(nllicrobiolog~'.

Vol, 2, Ann Arbur Science. Ann Arbor, Michigan,

S'\FDLCOB. C, \V.. :\'\D \\-'. G. (.'OClIl{.\' 1967. Statisticallllethuds. 6th f'd. Iowa State Unin'rsity l'res.~. Ames.,')\).3 pp.

STF\\'AHT. W. J). P. J967. TmIlst"t'J" of biol(lf.';icall~·fixed nitro­t'Cl) in a sand dUll(" .slack reVioll. "J'atllW 214: 60:3---604." ~

STEY,\. P. L.,.-\~D C. C. J)EIWICUJ-" 1870.I'''-itrogCll fixatio)lby nons:mbiutic microorpmisms in ~OJllC Californiasoils. Em-ironntt'ntal SciencL' and Tcchnoloi0' 4: J122­Illf-;.

WE LS II. S. L .. :,\!, D. AT\\'()OD, L. C. III (;(; Il\'S A'\ J) S. COOD­

IUCli 1087. A Utah llora. Creat Basin Naturalist ~vletTl­

oir !\o. 9: 1-894,\Vl·sr :'-I. E. lY7S. Physical inputs of ltitrog;:'Jl to dcsert

eco.wstems. Pages 16,'5-170 ill 1\. E. \\'('st and J. Skn-. ,jins, cds .. Nitrogen in desert ceosyslcrllS. Dowd('lLlIntel tillSo)) ,1ml Ho.%, ]ne., Stroudshurg, Pennsvh-anhl,

__. 19S1. :\utrient L)'ding in desert eeosystems, Pages301--32·1 ill D. W, GorxlalL H. A. !'PITy, and K. ~L W,Howes, f'ds" Aridland ecosystetlls: structnre. fnnction­ing .mel managemcnt, Clmhridge Uni\'c)"sity Press.Ciulbridge, United Kingdoili.

.. . 1090. Structure and fllnction of lllicrophytil' soilcrusts in wildland ecosystellls or arid to scm i-arid re­gi()JIs. A(h-ances in Ecolo~ical Ikst'areh 20: ] 70-22:l

WEST, N, E., A'\n ]. SUljll\S 1977. TIlt-' nitrogen nd{' in. . .

1\Olth Alllerican cold-\\inter sentilleselt ecosvstclttS.Oecolo,:~ia Pbntarnm 12: 45_3,'3.

•____. 1975. Summary concilisions. and .\uggt'stiO\lS for

fllrtlH'r wsearch, 1',1I','s 244----25:3 in 1\. E, \Vest and J."Skujins, cds., :".Jitnlgen in desert ecosvste]l!S. O()\w!e)).. .

H lltchinson and Ross, Inc., Stromlsll1lrg. Pt'nnsyh-ania,\V\'I.LSTE1'\ L. H .. .\t. L. BHl'E.'Il\C A'\D \V B. BOLE\J,

197~J. NitrngL·n fixation :J_ssodatt'd with sand grainsheaths (rhizosheaths) of ('ert.lin xeric grasses, Physi­alogia Plantarum 4fi: 1-4.

\VUI,tSTE1'\, L. 11.,,\'\1) S. A. PKATT 19S1. Scanning plec­trolt 1l1l1'WSCOP;: of rhizusheaths of Ory;:;opsishyllll'lloit!cs. Anlerican l0tlrna] of BotaJ):' &'): 408-4 lU.