Plant, Cell and Environment {\QQS) 18, 85-90

^̂ N natural abundance of vascular rainforest epiphytes:implications for nitrogen source and acquisition

G, R, STEWART,' S, SCHMIDT,' L,L, HANDLEY,^ M, H, TURNBULL,' P, D, E R S K I N E ' & C, A, JOLY^

'Department of Botany, University of Queensland, Brisbane Qld 4072, Australia, 'Scottish Crop Research Institute, Itivergowrie,Dundee DD2 5DA, UK, and-^ Departtnent of Botany, University of Campinas, Catnpinas 13081, Brazil

ABSTRACT

The foliar natural abundance of *̂ N was analysed to com-pare the potential nitrogen sources of vascular rainforestepiphytes and associated soil-rooted trees. Leaves of epi-phytes collected from six rainforest communities in Brazil,Australia and the Solomon Islands were depleted in '*Nrelative to the trees at each site. Epiphyte 5'*N was as lowas -6'4%c, while trees were generally enriched in '̂ N (0-7to 3*5%c). These results indicate either that epiphytes usenitrogen sources depleted in '̂ N or that discriminationagainst '̂ N is an intrinsic function of epiphyte physiology.At three sites, epiphytes could be grouped into those hav-ing both Iow 5'^N and low leaf-nitrogen content and thosepossessing hoth high 5'^N and high leaf-nitrogen content.The second group had 6'^N values in the range sometimesattrihutahle to N2 fixation (-2 to Woe). There was no corre-lation hetween growth form and 6''''N. It is concluded thatepiphytes may utilize '^N-depleted nitrogen from atmo-spheric deposition and N2 fixation.

Key-words: attnosphetic deposition; N2 fixation; naturalabundance ''^N; nittogen soutees; rainforest; vascular epi-phytes,

INTRODUCTION

The natural abundance of '"''N has been used to determinepreferential sources of inorganic nitrogen for plants,including hoth soil and atmospheric sources (Virginia &Delwiche 1982; Pate, Stewart & Unkovich 1993). In themajority of teported instances, soil nittogen was enrichedrelative to N2 (Shearer & Kohl 1989) due to kinetic frac-tionation during soil nitrogen cycling (Handley & Raven1992). The 5'^N values of plants ate a function of bothpreferential use of isotopically distinct nitrogen sourcesand assimilatory ftactionation. S'̂ 'N values recorded forterrestrial plants from various ecosystems range from -6 to+ 15%c (Virginia et al. 1989; Popp 1993) with low valuesof -17oo reported for plants growing in very young vol-canic soils greatly depleted in ''̂ N (Vitousek, Shearer &Kohl 1989).

Correspondence: G. R. Stewart, Depattment of Botany. Universityof Queettslatid, Brisbatie Qld 4072, Australia.

There appear to be no '"̂ N data for epiphytes, althoughepiphytes are extremely diverse and can represent a largeproportion of the total biomass of some ecosystems. Intropical tainforests, epiphyte weight has been estimated tobe as tnuch as 35-50% of tree leaf biomass (Edwards &Grubb 1977; Tanner 1977). Plants frotn many plant fami-lies grow epiphytically and exhibit numerous adaptationsto their gtowth environtnent. Adaptations for nutrientacquisition include ttash-basket (Aspletuum) and tankgrowth forms (Bromeliaceae), and leaf features such as tri-ehomes (Tillandsia) and velamen (Orchidaceae) (Benzing& Renftow 1974; Benzing 1990).

Potential nitrogen sources for epiphytes must be ofeither atmospheric or canopy origin once seed reserveshave been depleted. The probable nitrogen sources arethree major nitrogen pools: (i) canopy-derived nitrogen,(ii) atmospheric deposition, and (iii) nittogen derived ftomatmospheric N2 fixation. Canopy-derived nitrogenincludes both solubles leached frotn canopy leaves, stetnsor epiphytes and litter, and gaseous NH3 released by leaves(Parton et al. 1988). Atmospherically derived nitrogenincludes exogenous dry and wet deposition and N2 fixa-tion; N2-fixing tnicroorganisms have been reported on theleaf surfaces of epiphytes (Sengupta et al. 1981; Brighignaet al. 1992), Another possible nitrogen source is nitrogenderived from animals, either directly as excreta and deadbiomass or thtough insect symbioses such as ant-plants,where host plants are thought to receive nutrients frotntheir ant partners (Huxley 1980),

Soil nitrogen is the tnajor source of nitrogen for non-N2-fixing tenestrial plants; here we leport investigations intothe nitrogen sources of epiphytes. Our 5'^N data for arange of vascular rainforest epiphytes show that they con-tain isotopically light nittogen. We discuss the implica-tions of this with respect to the nitrogen sources utilized byepiphytes.

MATERIALS AND METHODS

Study sites

Leaf material of vascular rainforest epiphytes and associ-ated tree species was collected from four tropical and twosubtropical rainfotests. Sites in Brazil were situated in SaoPaulo State and comprised a tropical gallery forest (Brotas,

85

86 G. fl. Stewart etal.

22°15'S, 48°07'W, altitude 640 m) and a tropical semi-deciduous rainforest (St Genebra, 22°49'S, 47°06'W, alti-tude 669 rn). Sites in Australia were a tropical lowlandrainforest in north-east Australia (Cape Tribulation,16°25'S, 145 °E) and two subtropical rainforests in easternAustralia near Brisbane (Lamington National Park,28 °14'S, 152 °30'E, altitude 900 m; Mt Glorious/Mt Nebo,27°20'S, 153 °E, altitude 640 m). Samples from theSolomon Islands were collected from tropical rainforestson the islands of Kolombangara and New Georgia (8°S,157°OO'Eandl57°3O'E).

Plant material

Mature leaves were collected from epiphytes and fromsoil-rooted trees. The Brazilian epiphytes (rt=18; fordetails see Table 1) included members of the familiesBromeliaceae, Orchidaceae, Peperomiaceae, Polypodi-aceae and Caetaeeae. The Australian epiphytes (72=63)were members of the Aspleniaceae, Polypodiaceae, Orchi-daceae, Peperomiaceae, Pandaceae and Adiantaceae. Theepiphytes (n=10) collected in the Solomon Islands weremembers of the Orchidaceae and Rubiaceae. Leaves from

%N

Brazil (Brotas), tropical gallery forestEpiphytes Caetaeeae

PolypodiaceaePeperomiaceaeBromeliaceaeOrchidaceae

Total epiphytesSoil-grown trees

12(2)2(2)2(2)3(3)

10 (10)26 (26)

Brazil (St Genebra), tropical semideciduous forestEpiphytes Bromeliaceae

OrchidaceaePiperaceaeCaetaeeaePolypodiaceae

Total epiphytesSoil-grown trees

112(2)3(3)8(8)

18(18)

Australia (Cape Tribulation), tropical lowland rainforestEpiphytes Adiantaceae

OrchidaceaePandaceaeAspleniaceaePolypodiaceae

Total epiphytesSoil-grown trees

1113(1)6(2)

12(6)12(4)

Australia (Mt Glorious), subtropical rainforestEpiphytes Orchidaceae 6(1)

Polypodiaceae 8(1)Aspleniaceae 10(1)

Total epiphytes 24 (3)Soil-grown trees 8 (4)

Australia (Eamington NP), subtropical rainforestEpiphytes Orchidaceae 2 (2)

Peperomiaceae 2(1)Polypodiaceae 5 (3)Aspleniaceae 18 (2)

Total epiphytes 27 (8)Soil-grown trees 19 (12)

Solomon Islands, tropical rainforestEpiphytes Orchidaceae 7 (6)

Rubiaceae 3 (2)Total epiphytes 10 (8)Soil-grown trees 27 (25)

2-41-7(0-8)1-9(0-8)0-9 (0-0)0-8 (0-3)1-8 (1-3)2-6 (0-7)

1-10-82-31-6(0-4)2-2 (0-3)1-7 (0-6)3-1 (0-8)

3-60-41-11-3(0-7)1-6(0-7)1-2 (0-7)1-5 (0-4)

1 -0 (0-2)1-2(0-6)1-4(0-5)1-2 (0-5)1-6 (0-3)

1-0(0-2)1-4(0-0)1-3(0-5)1-8(0-6)1-6 (0 6)1-6 (M)

1-5(0-6)1-1 (0-4)1-4 (0-6)1-5 (0-6)

0-3-1-2(1-5)-2-3(1-6)-5-2(0-1)-2-1 (0-6)-2-3 (2-1)

2-6 (1-5)

-4-9-2-6-1-3-0-5 (4-8)-0-7(1-8)-1-5 (2-6)

3-1 (1-2)

0-1-4-0-6-4-2-2 (0-8)-2-5(1-7)-2-7 (1-8)

2-3 (2-8)

0-3(1-3)-0-1 (2-2)-0-3(1-9)-0-1 (1-8)

0-7 (1-9)

0-6 (0-3)-0-1 (0-6)-0-8(1-8)

0-2(1-4)0-0 (1-4)3-5 (1-2)

-2-2(1-3)-3-0 (0-5)-2-4 (1-2)- M (1-7)

Table 1. Foliar 5''̂ N and % nitrogen (±SD)of epiphytes (n = number of samples;number of species in parentheses) and trees(pooled) from six rainforest sites

^N natural abundance of rainforest epiphytes 87

6

'^ -3

-6

6

-s- 3

z 0

'^ -3

-6

P < 0 001

T

fEpiphytes Trees

Brotas

P < 0 99

I

Epiphytes Trees

Mt Glorious

P < 0 001

11

I1

Epiphytes Trees

St Genebra

P < 0 001

i

Epiphytes Trees

Lamington

P < 0 001

11

Epiphytes Trees

Cape Tribulation

P < 0 97

T

Epiphytes Trees

Solomon Islands

Figure 1. Foliar S'̂ N (±SD) values for epiphytes and ttees from six rainforest sites.

non-N2-ftxing trees were collected from trees in the vicin-ity of the epiphyte samples and comprised a laige range ofspecies and families representing the dominant membersof each forest cotnmunity.

Nitrogen isotope analysis

The leaf samples were oven-dried at 60 °C on the day ofcollection; samples frotn the Solomon Islands and CapeTribulation were air-dried in the field and oven-dried afterapproxitnately 1 week. The samples were ground to a finepowder in a vibratory ball mill (Retsch MM-2, Haan, Ger-many) and the nitrogen content of the leaf material wasdetertnined by automated cotnbustion. Duplicate satnplesof approximately 120 fxg nitrogen were analysed for '̂ Nusing a continuous-fiow isotope ratio mass spectrometer(CF-IRMS, Tracer Mass, Europa Scientific, Crewe, UK)set to the single nitrogen mode. The ptecision of the instru-ment, based on multiple analysis (/!=133) of a laboratorystandard (Eucalyptus crebra leaves), is 0-21%c SD.

Statistics

The data were analysed using STATISTICA (Statsoft, Tulsa,Oklahoma). Significant differences were detertnined byANOVA followed by Scheffe's post-hoc test. TheKruskal-Wallis ANOVA by ranks was used for compaii-son of the 5'^N values and nitrogen contents of epiphytesat each site.

RESULTS

Epiphytes representing a range of morphological struc-tures and nutrient acquisition strategies exhibited both low8'''N and Iow leaf-nitrogen contents relative to associatedtree species (Table 1). Mean 5 ' ' 'N values for epiphytesfrom the six sites (total /!=91) were depleted when cotn-pared with associated tree species (Fig. 1), Epiphytes at allsites, except Lamington National Park, exhibited lowerleaf-nitrogen content than trees, indicating possible nitro-gen limitation in the epiphyte growth environment. Theseresults are similar to those found previously for speciesgrowing epiphytically when compared with those growingtertestrially (Ball et al. 1991). At the Brazilian sites, epi-phyte S'-'̂ N was -2-3%o at Btotas and -1 •5%f. at St Genebra,ranging frotn -5-3 to 0-0%c and -A-9 to 2-9%c, respectively.Corresponding mean 5'"'̂ N values for tree species were2 6%(. (Brotas) and 3- l%(i (St Genebra). The tnean S'-'̂ N ofepiphytes frotn ttopical rainforest at Cape Tribulation was-2-7%c, ranging from -6-4%c to 0- l%c, and a mean value of2-3%c was found for associated trees. Epiphytes frotn thesubtropical rainfotests were not as gteatly depleted in ' ' 'Nas epiphytes from tropical forests, with tnean values of-0-1 %o at Mt Glorious and 0-0%c at Lamington NationalPark, ranging frotn ^ - 5 to 2-6%o and -2-3 to 24%c respec-tively. The conesponding tnean 5'''N values for trees were0-7%o (Mt Glorious) and 3-5%c (Latnington NP). Epiphytesfrom the Solomon Islands had a mean 5''''N value of-24%o, ranging frotn -3-5 to -0-1 %o, and trees had anaverage value of-M%c. At all sites the 8'''N of epiphyteswas more depleted than the 5'^N of leaves frotn associated

88 G. R. Stewart eta\.

1 -1

-1 -

-3 -

-5 -

IO

(a)

-7

•

1 -

-1 -

-3 -

-5 -

(c)

(

V tn

•

1

=1J

1 11 2 3

% N (dry weight)

1 - ]

-1 -

-3 -

-5 -

-7

(b)

1 -

-1 -

-3 -

-5 -

-7

(d)

I I \ I1 2 3 4

% N (dry weight)

Figure 2. Foliar 6'''N values and leaf nitrogen contents (% dry weight) for epiphytes from three rainforests: Brotas (a). Cape Tribulation (b)and the Solomon Islands (c) indicating groupings of epiphytes with both low 5'-'̂ N and %N or both high 5''̂ N and high %N, Groupings weredetermined according to Kruskal-Wallis ANOVA by ranks. The data are pooled in (d) and the regression of 5'̂ N on %N is shown.

trees. At the Mt Glorious and Solomon Islands sites thesedifferenees were not statistieally significant, although thesame trend existed.

A strongly positive relationship (r = 0-64; 5'''N=l-51x-4-62 at P<0-01) between 5''*N value and nitrogen contentwas found for epiphytes from Brotas, Cape Tribulationand the Solomon Islands (Fig, 2d). At these sites, epiphytescould be divided into two significantly different groups: arelatively '''N-enriched group and a relatively depleted one(Figs2a-c).

There was no eorrelation between growth form (litter-eolleeting speeies versus non-eollecting species) and 5'^N.Litter-colleeting epiphytes from the Australian sites{Asplenium and Flatycerium; 5'"̂ N = -2-4 to 0- I%o) did notdiffer in their 5'^N values from non-litter-eollecting epi-phytes {Dendrobium, Adiantum. Peperomia, Freycinetiaand Dictymia; 5'-''N = -3-5 to 0- l%c).

DISCUSSION

At five sites epiphytes were greatly depleted in '^N. Thereason for this must be either that '^N depletion of epi-phytes is an intrinsic funetion of physiology, or that itarises from the nitrogen source(s) that epiphytes use. Thelow 5'^N signatures of trees from the Solomon Islands arein accordance with '""N depletions recorded for young

pleistocene soils of volcanic origin (Vitousek et al. 1989).Here, most epiphytes wete not significantly triore depleted in•̂ N than tree species; only the epiphytic ant plants {Hydno-

phytwn and Mrytttecodia) were more depleted (>l%c)than the tree species, which might reflect the '"̂ N depletion ofanimal excreta (Wada, Mizutani & Minagawa 1991),

Discrimination against '"''N during nitrogen uptake andnitrogen assimilation would result in '^N depletion in tbeplant. This eould be related to nitrogen form, the fractiona-tion of leaf-absorbed versus root-absorbed nitrogen, differ-ent assimilatory pathways, or mycorrhizal status, NH4"̂ -grown plants are reported to be depleted in '""N relative tosource nitrogen (Yoneyama et cd. 1991). There is evideneethat non-mycorrhizal plants discriminate against '"""N lessthan do those with VA- or ecto-myeorrhizal associations(Handley el al. 1993; Pate et al. 1993). However, there isno reason to suppose a priori that epiphytes and trees inthe same ecosystem exhibit differential discriminationagainst '̂ N related to either nitrogen souree or myeor-rhizal status.

Possible nitrogen sourees for epiphytes are (i) canopy-derived nitrogen, (ii) nitrogen derived from Nj fixation,and (iii) nitrogen derived from atmospheric deposition. Intropical rainforests the leaching of mineral nutrients frornthe eanopy ean be substantial, with 4 to 80kg N ha~' a~'reported (Bernhard-Reversat 1975; Jordan et al. 1980;

'̂ A/ natural abundance of rainforest epiptiytes 89

Herrera & Jordan 1981). Litter fall in mature lowland trop-ical rainforests has been estimated as 28-140kg N ha"' a '(Herrera & Jordan 1981; Brasell & Sinclair 1983;Vitousek 1984). While both leaching and litter fall maycontribute to epiphyte nitrogen nutrition, it is not obviousthat this would lead to '"̂ N depletion. Leachates rnight be'"^N-depleted as a consequence of kinetic fractionation.

N depletion of litter-derived nitrogen could arise frotnN / ' ' ' N fractionation during mineralization processes;

however it would be necessar'y to invoke mineralizationprocesses intrinsically different from those demonstratedfor tree litter lying on soil (Natelhoffer & Fry 1988). Inaddition, a small data set (L. Handley & D. Bergstrom, Uni-versity of Queensland, unpublished results) suggests thatcanopy litter is not a depleted nitr̂ ogen source for epiphytes.Fresh litter, decayed litter and downstream non-vascularepiphytes of a Moteton Bay fig all had a 5'^N of about 3%c.There was no '"̂ N depletion from source to sink.

N2 fixation in the phyllosphere is considered a possiblenitrogen source for tropical rainforests (Sprent & Sprent1990). The amount of nittogen fixed by bark tnicro-organ-isms and lichens has been estimated at between 35 and 200kg nitrogen ha"' a ' (Herrera & Jordan 1981), while nitro-gen fixed by canopy lichens may account for an input of1-5 to 8 kg N ha"' a"' (Forman 1975). Several epiphytespecies have been found to have N2-fixing bacterialmicrofiora in their phyllospheres (Sengupta et al. 1981;Brighigna et al. 1992). There is also evidence that N2 fixa-tion occurs in the epiphyte root substrate with estimatedvalues of 40jUg N2 fixed g ' DW a ' (S. Schmidt, Univer-sity of Queensland, unpublished results). The 6'''N signa-ture of nitrogen fixed via nitr ogenase is close to that of NT,but 5'"̂ N values of whole plant shoots lie in the range of-2to 0%o when plants are exclusively dependent on N2 fixa-tion (Bergersen, Peoples & Turner 1988). It is possible thatthe epiphytes with higher nitrogen contents and 5'^N val-ues in the range -2 to 0%c directly acquired nitrogen fromN2 fixation. For those epiphytes with low S'̂ 'N signaturesand low nitrogen contents, nitrogen sources other than N2fixation seem likely because their 8'''N values are belowthose generally associated with N2 fixation (Shearer &Kohl 1989).

The third possible nitrogen source for epiphytes is atmo-spheric deposition. Dty deposition is reported to supply25% of the nitrogen required for forest growth, with a fur-ther 15% of the requiretnents being introduced as wetdeposition (Lindberg etal. 1986). Nitrogen input from pre-cipitation has been found to be in the range of 11-22 kg Nha~' a"' for various tropical forests (Herrera & Jordan1981; Jordan et al. 1982), with the majority of nitrogenbeing deposited as NH4"̂ rather than NO3 . Greater inputsare repotted in Europe, where atmospheric depositionaccounts for inputs of 20-lOOkg nitrogen ha"' a"', with50-80% of the nitrogen deposited as NH4^ (Pearson &Stewart 1993). 5'^N for NH3/NH4"' in rainfall can be aslow as -I2%o (Freyer 1978), and generally shows '""Ndepletion relative to N2 (Moor-e 1977; Heaton 1987;Garten 1992). Similarly, S'-'̂ N values for NO^ range from

-6 6 to -3-l%o (Moore 1977; Freyer 1978; Heaton 1987).It seems probable that atmospheric deposition of depletednitrogen accounts for the observed low 5' ' 'N signatures ofthe greatly '"'N-depIeted epiphytes.

The epiphytes can be grouped according to '̂ N deple-tion, one group exhibiting higher nitrogen contents andhigher 5 N values, and one group having lower nitrogencontents and lower 6 ' ' 'N values. It is possible that epi-phytes in the first group receive at least some nitrogenfrom N2 fixation, because their 5'''N values are indicativeof N2 fixation and because these epiphytes also had highnitrogen contents relative to other epiphytes at the samesites. The second group uses depleted nitrogen sources andis more nitrogen-limited, having low nitrogen contents. Itis ptobable that this latter group utilizes nitrogen derivedfrom '"''N-depleted deposition. No other nitrogen sourcesor fractionation processes can explain the observed deple-tion of these epiphytes.

ACKNOWLEDGMENTS

We thank Rosetnaty Goodall and Gordon Moss for theirexcellent technical assistance and Dr Michael Olsen forhelp with collection and identification of specitnens. Thisstudy was supported by a special projects grant frotn theUniversity of Queensland. S. S. is indebted to the GottliebDaimler-und Karl Benz-Foundation and to DEET. L. L. H.is grateful to receive a British Council Travel Grant,

REFERENCES

Ball E., Hann J., Kluge M., Lee H.S.J., Luttge U., Orthen B., PoppM., Schriiitt A. & Ting LP. (1991) Ecophysiological compon-ment of the tropical CAM-tree Clti.?ia in the field. New Phytolo-

Benzing D.H. (1990) Vascular Epiphytes. Cambridge UniversityPress, Cambridge.

Benzing D.H. & RenfVow A. (1974) The mineral nutrition of theBromeliaceae. Botanical Cazette 135, 281-288.

Bergersen F.J., Peoples M.B. & Turner G.L. (1988) Isotopic dis-crimination during the accumulation of nitrogen by soybeans.Austtalian Jountal ofPlattt Physiology 15, 407^20.

Bernhard-Reversat F. (1975) Nutrients in through fall and theirquanlitative importance in rain forest mirier-al cycle. EcologicalStudies n, 153-159.

Brasell H.M. & Sinclair D.F. (1983) Elements returned to forestfloor in two rainforest and three plantation plots in tropical Aus-tralia. J(mnml of Ecology 71, 367-378.

Brighigna L., Montini P., Favilli F. & Carabez Trejo A. (1992)Role of the nitrogen-fixing bacterial microfiora in the epi-phytisni of Tillandis (Bromeliaceae). American Journal ofBotany 79, 723-727.

Edwards P.J. & Grubb P.J. (1977) Studies of mineral cycling in amontane rainforest in New Guinea 1. The distribution of organicmatter in the vegetation and soil. Journal of Ecology 65, 943-69.

Forman R.T. (1975) Canopy lichens with blue-green algae: a nitro-gen source in a Colombian rainforest. Ecology 56, 1 176-1184.

Freyer H.D. (1978) Seasonal trends of N H / and NO," nitrogenisotope composition in rain collected at Julich, Germany. Tellus30,83-92.

90 G. R. Stewart e\a\.

Garten C.T, (1992) Nitrogen isotope composition of ammonium and* nitrate in bulk precipitation and forest throughfall. International

Journal of Environmental Atialytical Chemistry 47, 33-^5.Handley L,L, & Raven J,A, (1992), The use of natural abundance

of nitrogen isotopes in plant physiology and ecology. Plant. Cettand Envirotitnent 15, 965-985,

Handley L,L,. Daft M,I,, Wilson J,, Scrimgeour CM,, Ingleby K,& Sattar M,A, (1993) Effects of the ecto- and VA-mycorrhizalfungi Hydnagium carneum and Glomus ctarum on the 5' ' 'N and8'"'C values of Eucalyptus gtobutus and Ricinus communis.Plant, Cell and Environment 16, 375-382.

Heaton T,H,E, (1987) " N / " ' N ratios of nitrate and ammonium inrain at Pretoria, South Africa, Atmospheric Environment 21,843-852,

Herrera R, & Jordan CF, (1981) Nitrogen cycle in a tropical Ama-zonian rain forest: the Caatinga of low mineral nutrient status. InTerrestrial Nitrogen Cycles (eds, F, E, Clark & T, Rosswall),Ecological Buttetin (Stockholm) 33,493-505,

Huxley C (1980) Symbiosis between ants and epiphytes, Biotogi-calReview 55, 32\-340.

Jordan CF,, Golley F,. Hall J,D, & Hall J, (1980) Nutrient scaveng-ing of rainfall by the canopy of an Amazonian rain forest,Biotropica 12, 61-66.

Jordan CF,, Caskey W., Escalante G,, Herrera R,, Montagnini F,,Todd R, & Uhl C (1982) The nitrogen cycle in a 'terra firme'rainforest on oxisol in the Amazon Territory of Venezuela, Plantand Soil 67. 325-332.

Lindberg S,E,, Lovett G,M,, Richter D,D, & Johnson D,W, (1986)Atmospheric deposition and canopy interactions of major ions ina forest, Sdence 231, 141-145.

Moore H, (1977) The isotopic composition of ammonia, nitrogendioxide and nitrate in the atmosphere. Atmospheric Etivironment11, 1239-1242,

Natelhoffer K,J, & Fry B, (1988) Controls on natural nitrogen-15and carbon-13 abundances in forest soil organic matter. Soil Sci-ence Society of American Journal 52, 1633-1640,

Parton W,J,, Morgan J,A,, Altenhofer J,M, & Harper L,A, (1988)Ammonia volatilization from spring wheat plants. AgronomyJournal 80, 4\9^25.

Pate J,S,, Stewart G,R, & Unkovich M, (1993) '^N natural abiin-dance of plant and soil components of a Banksia woodlandecosystem in relation to nitrate utilization, life form, mycorrhizalstatus and Ns-fixing abilities of component species. Plant, Celland Environment 16, 365-373.

Pearson J, & Stewart G,R, (1993) Tansley Review No, 56, Thedeposition of atmospheric ammonia and its effect on plants. NewPhytologist 125, 283-305,

Popp M, (1993) Ecological aspects of nitrogen nutrition. Progressof Botany 54, 448-^60,

Sengupta B,, Nandi A,S,, Samanta R,K,, Pal D,, Sengupta D,N, &Sen S,P, (1981) Nitrogen fixation in the phyllosphere of tropicalplants: occurence of phyllosphere nitrogen-fixing micro-organ-isms in Eastern India and their utility for the growth and nitrogennutrition of host plants. Annals of Botany 48, 705-716,

Shearer G, & Kohl D,H, (1989) Estimates of Nj fixation in ecosys-tems: the need for and basis of the ''̂ N natural abundancemethod. In Stable Isotopes in Ecological Re.search. EcologicalStudies 68 (eds, P, W, Rundel, J, R, Ehleringer & K, A, Nagy),pp, 342-347, Springer Verlag, Berlin,

Sprent J,I, & Sprent P, (1990) Nitrogen fixing organisms. Pure andapplied aspects. Chapman and Hall, London,

Tanner E,V,J, (1977) Four montane rain forests of Jamaica: a quan-titative characterization of the floristics, the soils and the foliarmineral levels, and a discussion of the interrelations. Journal ofEcohgy 65. &83-9\&.

Virginia R,A, & Delwiche CC, (1982) Natural ' ' N abundance ofpresumed N^-fixing and non-Nj-fixing plants from selectedecosystems, Oecologia 54. 317-325,

Virginia R,A,, Jarrell W,M,, Rundel P,W,, Shearer G, & Kohl D,H,(1989) The use of variation in the natural abundance of '''N toasses symbiotic nitrogen fixation by woody plants. In: StableIsotopes in Ecotogicat Research. Ecotogicat Studies 68. (eds, P.W, Rundel, J, R, Ehleringer & K, A, Nagy), pp,375-394.Springer Verlag, Berlin,

Vitousek P,M, (1984) Litterfall, nutrient cycling, and nutrient limi-tation in tropical forests, Ecotogy 65, 285-298,

Vitousek P,M,, Shearer G, & Kohl D.H, (1989) Foliar '"̂ N naturalabundance in Hawaiian rainforest: patterns and possible mecha-nisms, Oecologia 78, 383-388,

Wada E,, Mizutani H, & Minagawa M, (1991) The use of stableisotopes for food web analysis. Critical Review in Food Scienceand Nutrition 30, 361-371,

Yoneyama T,, Omata T,, Nakata S, & Yazaki J, (1991) Fractiona-tion of nitrogen isotopes during the uptake and assimilation ofammonia by plants. Plant Cell Physiology 32, 1211 -1217,

Received 29 March 1994: received hi revised form 28 June 1994:accepted for publication 20 July 1994