A C S S Y M P O S I U M S E R I E S 3 9 3

Biogenic Sulfur

in the Environment

Eric S. Saltzman, EDITOR

University of Miami

William J. Cooper, EDITOR

Florida International University

Developed from a symposium sponsoredby the Division of Environmental Chemistry

at the 194th Meetingof the American Chemical Society,

New Orleans, LouisianaAugust 30-September 4, 1987

American Chemical Society, Washington, DC 1989

Chapter 5

Sulfur Emissions from Roots

of the Rain Forest Tree

Stryphnodendron excelsum

Ecosystem, Community, and Physiological Implications

Bruce Haines1, Marilyn Black2, and Charlene Bayer2

'Botany Department, University of Georgia, Athens, GA 306022Analytical and Instrumentation Branch, Georgia Tech Research Institute,

Georgia Institute of Technology, Atlanta, GA 30332

Roots of Stryphnodendron excelsum trees in a lowland rain forestin eastern Costa Rica emit sulfur gases. Extrapolated annualestimates of emissions, based on S. excelsum tree density, are onthe order of 0.29 kg S.ha^.yr1. At the ecosystem level, this flux istoo small to account for the 11 kg S.ha^.yr1 SO4-S input-outputdiscrepancy and acid rain reported earlier. At the physiologicallevel, emission of €82 is stimulated by disturbance to the roots ofS. excelsum. Considering the known toxicity of C$2 to nematodes,root rot fungi, insects, and nitrifying bacteria we suggest that CS2emission may, at the community level, be a defensive mechanismagainst root predators, and pathogens and a nitrogen conservingmechanism.

Sulfur gas flux from living vascular plants to the atmosphere is a little studiedpart of the global sulfur cycle. Sulfur fluxes from other parts of the biosphere tothe atmosphere are more studied. For example, sulfur gas emissions are knownfrom marine systems (1-6). from coastal estuarine and marsh systems (7-16).from fresh water floodplain lakes (17), from temperate soils (18-20). and fromtropical forests (20-22). Laboratory studies have demonstrated sulfur gasemissions from soils (23-26). from intact higher plants (27-30). and plant parts(see review by Rennenberg 3_1). An inventory of reduced sulfur gas emissionsfrom soil, crops, and trees in the United States has been started (32.33.34).General reviews of the sulfur cycle are provided by Smil (3.5) and by Ivanov andFreney (36). From these studies, the geographic distribution of biogenic sulfurgas source strengths on a global basis and the phylogenetic distribution of sulfurgas emissions within the plant kingdom are still relatively unknown.

During a survey of sulfur gas emissions from a central American rainforest(37). Stryphnodendron excelsum Harms (Mimosaceae) was found to be asufficiently strong sulfur emitter that its location in the forest could be detectedby odor. The present study attempted to quantify sulfur emissions from 5.excelsum to the atmosphere.

0097-6156/89/0393-0058S06.00/0c 1989 American Chemical Society

5. HAINESETAL. Sulfur Emissions from Roots of a Rain Forest Tree 59

Study Site and Methods

Sampling was performed at the La Selva Biological Station of the Organizationfor Tropical Studies, Puerto Viejo de Sarapiqui, Provincia Heredia, Costa Rica,10°24-26 'N lat, 84° 00-02 'W long. Three S. excelsum trees ranging from 0.87 to0.96 m dia and from 22 to 32 m in height were selected within 100 m of thelaboratory to facilitate rapid processing of gas samples.

Preliminary sampling suggested that roots of one of the trees extendedmore than 16 m from the trunk. Later sampling was performed at 4 m intervalson transects on each of the compass quadrants centered on two trees. For onetree near a forest edge, one transect line was run toward the forest interior for16 m. At each sampling point on each transect a 25 x 25 cm template was usedto cut leaf litter and roots. Litter was removed to a 250 ml Nalgenepolypropylene wide mouth centrifuge jar (Nalge Co. Rochester, NY). Soil andassociated roots were excavated to a depth of 10 cm and removed to the labwhere they were separated by 4 mm sieves without addition of water. Drysieving avoided the possibility of flooding and anaerobism changing the qualityand amount of reduced sulfur gas emission. The total root mass was sorted byodor into sulfur emitting roots and non-emitting roots. Sulfur emitting rootswere placed in 250 ml polypropylene centrifuge jars and incubated at ambientrainforest temperature. The time course of sulfur gas accumulation in thecentrifuge jars was quantified by use of a Perkin-Elmer Sigma 4B gaschromatograph (Perkin-Elmer, Norwalk, CT) fitted with a sulfur specific flamephotometric detector, a 2 mm internal dia. 9 m teflon column packed with 5%polyphenyl ether + 0.5% H3PO^ on 40/60 teflon, at 90C, with 99 mis N2 carriermin-i. Reference hydrogen sulfide, carbon disulfide, dimethyl sulfide, and ethylmercaptan were supplied from permeation tubes (V1CI Metronics, Santa Clara,CA) inserted into a Tracor Model 432 Tri-Perm Permeation Calibration system(Tracor, Inc., Austin, TX). Standards and unknown samples were pulled eitherfrom sample incubation bottles or from the permeation system to the gaschromatograph through multiposition zero dead volume sampling valves and a 2ml teflon sample loop using a 10 ml syringe. Sample loop, valves, andconnecting teflon lines were heated to 65°C to minimize surface adsorption ofsulfur gases.

Verification of CS2 as the principal sulfur gas from the roots of each of thethree trees sampled in this study was performed with a Finnigan OWA 3B gaschromatograph-mass spectrograph (Finnigan Corp., Sunnyvale, CA).

Three incubation regimes were used for sulfur emitting roots.1) Field moist roots. All root samples were physically pulled from field

collected soil and incubated at existing moisture conditions.2) Vigorously washed roots. Following incubations of field moist roots.they

were removed from the bottles and squeezed repeatedly under running water todislodge adhering soil particles in preparation for dry weight determinations.When the odor of increased sulfur emissions was noted during washing, onesubset of samples was re-incubated and reanalyzed by gas chromatography todetermine the magnitude of change in emission rate.

3) Gently rinsed roots. For another set of 12 root samples, followingincubation at field moist conditions, the roots were kept in the incubation jarsand the gas was displaced by 3 changes of water in immediate succession. Thiswas done to determine if wetting alone or both wetting and squeezing wererequired to stimulate sulfur gas emissions.

Following the various incubation regimes, all roots were washed free of soiland dried to constant weight at 65°C. Dry roots were sorted into diameter

60 BIOGENIC SULFUR IN THE ENVIRONMENT

classes of 0-1.9, 2.0-4.9, 5.0-9.9 and greater than 10 mm. Means, standarddeviations, and correlations were computed with the Statistical Analysis System

Field Moist Samples. Emissions of the sulfur gases hydrogen sulfide, carbondisulfide, dimethyl sulfide + ethyl mercaptan from forest floor litter, roots, andsoil calculated from the initial slopes of time course incubations are given inTable I. Rates are given for distances of 4, 8, 12, and 16 m from trunks of S.excelsum trees as g S. rn^.min'1. Summing outward from the tree trunk, theannual extrapolated emission rate multiplied by the areas of each successivecircle (Figure 1), estimated an overall emission of 100 g S. tree^.year1 from a200 mZ area. The fractional contribution of S. excelsum trees to the stem crosssectional area (basal area) of the forest is 0.06 (39)- From the extrapolatedannual emission rate and the proportional contribution of S. excelsum to thecross sectional area, an annual emission of 0.29 kg S. ha'1 .yr1 was calculated.This assumes constant emission rates instead of possible diurnal and seasonalcycling of rates.

Vigorously Washed Roots. The time courses of CS2 at field moist conditionsand after two vigorous washings are given in Figure 2a. Results from each ofthe two washings are replotted in Figures 2b and 2c synchronizing thebeginnings of the re-incubations in order to facilitate comparisons of initialslopes. For this particular set of roots, €87 was detectable in 5 incubations offield moist roots, but detectable in 7 incubations after the first wash and in 8after the second wash. Following rapid initial accumulation of €82,concentrations either increased more slowly, remained the same or decreased.

Gently Rinsed Roots. The field moist incubations showed rapid CS2accumulation followed by declining concentration. CS2 production was notcontinuous. The initial slopes for the €82 time courses were greater for thefield moist incubations (Figure 3a) than for post-rinse incubations (Figure 3b).€82 was detectable in 7 of the field moist incubations but continued in only 5 ofthem following rinsing.

Root Weight and S Emission per Unit Root Weight. Dry weights of non-sulfuremitting and of sulfur emitting roots (Table II) show about the same totalamount of roots but decreasing amounts of sulfur emitting roots with increasingdistance away from S. excelsum trunks. The dry weights of 0-1.9 mm dia. non-sulfur emitting roots were negatively correlated (r -0.3, P< 0.049, n=36) with thedry weights of 2-4.9 mm diameter sulfur emitting roots. We do not interpret thisas evidence for allelopathy. Emission rates first calculated as g.nr2 forest inTable I are recalculated using individual sample root weights to give sulfuremission rates per gram root dry weight in Table III to facilitate comparison ofdata with data from future studies.

Discussion

Results of this study have implications for ecosystem ecology, communityecology, and for physiological-evolutionary ecology.

Ecosystem Level. This study was designed to estimate the potentialproportional contribution of sulfur gas emissions from S. excelsum to the 11kg.ha'i.yr1 804-8 input-output discrepancy of the rainforest (40). A gas

5. HAINESETAL. Sulfur Emissions from Roots of a Rain Forest Tree 61

Table I. Potential Sulfur Emission, Mean (Standard Deviation)g S x 1O~9 nr-^.min"1 from Roots, Litter and Soil at 4 Distances from

5. excelsum Trees. Sample Size = 9 at Each Distance

Source Distance, m

Sulfur Compound 4

RootsCS2-S 2959(2010)H2S-S 41.6(83.2)

Litter(CH3)2S-S 0.006(0.22)

C2H5SH-SSoil 0(0)

8

1663(1417)3.5(10.7)

0.23(0.62)

0(0)

12

1049(1242)0(0)

0.18(0.49)

0(0)

16

281(416)0(0)

0.17(0.38)

0(0)

Total 3000(1981) 1666(1420) 1049(1242) 281(410)

62 BIOGENIC SULFUR IN THE ENVIRONMENT

Stryphnodtndron La Selva 1985-86

- l .57g-m- 2 x 13.5m2 » I9.7g• 0.87 g-m"2 x 37.7m2 - 33.0g• 0.55 g-m'2 x 62.8 mz -34.6g0.14 g-m'2x 87.9m2 - I2.9g

201 m2 I00.3g

0.29 kg I'll kg S04-S DiscrepancyFigure 1. Estimation scheme for sulfur emission from an averageStryphnodendron tree where emissions from roots, leaf litter, and soil weresampled at 4 m intervals on transects radiating out from trunk. Emissions ateach distance (Table I) were multiplied by the area of circular sampling bandbetween that distance and the previous distance, and the emissions summedthrough the circular bands for the whole tree.

5. HAINES ET AL. Sulfur Emissions from Roots of a Rain Forest Tree 63

, 5 0 0 ,v 'First wash" '

100

Time, min.Figure 2. Time course of CS2-S concentrations in incubations of S. excelsumroots. A) time course for individual incubations (identified by numbers)starting as field moist roots and later subjected to two episodes of vigorouswashing (W). B) time courses following the first washing replottedsynchronizing the starting times of re-incubation. C) replot for the second washas in B.

ICT9500 1000

Time, min.1500

Wa»h2000 9 300

Woih

Figure 3. Time course of €82-8 gas concentrations incubations of S. excelsumroots. A) time courses for individual incubations starting as field moist rootsthen gases displaced by 3 gentle rinsings with water (W). B) time courses afterwashing re-drawn with starting times synchronized.

64 BIOGENIC SULFUR IN THE ENVIRONMENT

c/r

J~

11! §QOS* 2o* c^o oe ^ c "̂i'l^i

^a §3

^Vs-a0"*O T3 dc D'3 a g.H.5 a! eSCO

Q ' — s <D &

fitsJH .O «^5* *""* ^^ *^j,/•;§ E ^£ Epo^

'^°X^!> Z gEpo^ SQ^X[ii•*-* _2 ^o t^to O ̂ Oc^f-1 E

H- i c^

3n)H

EiTo

E

C/3Q

VO

rH

OO

^

os

«

OOOS

OO

OS

00OS

OS

rj« a<D t/TS i)t/i-2J

o-oo^o"^5:5^o'°0

T^- T— < (^ O^^O Tj" rH T— <

^— sy— sS^s^o^o^^

CN CO CO O

C^- T-H C?\ T-H^o ^^ *™^ r^~

l*O *̂

CO 00 ^t O\

CN c<i co vq

X"Vy^-^X— s.rH CN* (NO"CO ^O i — ' OCN T-H ^H

^•^^^""sv^^v^^vO\ *4O rH Q\C-i (N CS ON(N o\or-j

o\ . . ̂^ ̂ ̂ rH«H°A

! 1<20

rq,-H

^^T-Hof

^<QrHSo'rH

rHrHoq'

0?

*̂1f^1—1

/-s£o"10

^^̂.•

rH^ — 'rorH

3

O.50)Q•o

§.'a 8oo B §

Em

issi

ons,

Mea

n i

Roo

t D

ry W

eigh

t'1

ie =

9 a

t E

ach

Dis

i

| <*&

.— i pCd >^ !3'•gooooU M

&H

H733L_t

E

Ss

nee

from

5. e

xcel

su

rtty2Q

rH

tN1-1

00

1̂"

•oG3OCX

1

0s

o

c?^o"•rTo'cs

It?gSo

. t~-CO v- (ic^^fO^

00 00tSOOOO N

UK

5. HAINESETAL. Sulfur Emissions from Roots of a Rain Forest Tree 65

emission of 11 kg S. ha^.yr1 to the atmosphere its subsequent oxidation to S04-2

and dilution in rain, might explain the acid rain (41) reported for this forest(40). A literal interpretation of emissions estimates calculated from rootsincubated at field moist conditions and extrapolated to annual estimates,thenadjusted for the proportion of S. excelsum trunk cross sectional area of theforest indicates annual emission rates of 0.29 kg S.ha^.yr1. Emissions from S.excelsum may be incorrectly estimated for three reasons.

First, maximum emission rates may have been missed. Emission rates wererapid at first, then declined. For individual samples, between 15 and 25 minwere needed to carry the soil from the forest to the laboratory, to extract sulfuremitting roots from the soil, and to place the roots in the incubation bottles.Incubations for determining rates were usually between 5 and 10 min duration.

Second, the decrease in concentration of CS2 in the headspace of theincubation jars, Figure 2, suggests that two reactions namely €82, productionand CS2 consumption were simultaneously in progress. As long as productionwas greater than consumption, the concentration of C$2, appeared to increase.When consumption exceeds production, the concentrations appears to decrease.With both reactions in progress, the gross gas production would be the sum ofthe two rates. Whether CS? was simply adsorbed to the surface of theincubation vessel or was oxidized to SO2 and SO^2 is unknown, thus themagnitude of the consumption rate is unknown. Potential CS2 emission ratesmay thus have been underestimated in these incubations.

Third, the stimulation of gas emission from roots by washing raises thequestion about the effect of the passage of wetting fronts through the rainforestsoil with each of the numerous rain showers.

Future quantification of sulfur gas emissions from S. excelsum roots shouldavoid possible mechanical damage to roots caused by their excavation from soilsas in the present study. Instead, plants need to be grown in soil in containers towhich simulated rainfall additions are made while sampling for sulfur gasesabove the soil surface. Once the relations of sulfur gas emission rates to thefrequency and amounts of soil wetting are determined, these data can becoupled with a rainfall frequency and quantity model to simulate potentialsulfur gas emission for S. excelsum on an annual basis.

Perhaps more important than the exact quantities of these emissions withthe three sources of uncertainty, are the qualities of the sulfur emissions.Qualitatively, C$2 emissions have community and physiological implications.

Community Level. Sulfur gas emissions from S. excelsum roots may influencethe species composition of biological communities through actions as anti-bacterial, anti-fungal, anti-nematode, anti-herbivore, anti-plant (allelopathy)agents. Anti-bacterial properties of CS2 can inhibit nitrification, the oxidation ofammonia to nitrate (42-44). The anti-fungal properties of €82 have been usedto control root rot fungi in agriculture and in forestry (45-47). It is a registeredfungicide and nematicide (48). Its anti-nematode properties have been used insoil fumigation in agriculture (49). Its anti-herbivore properties have been usedto kill insects in stored grains (50). The anti-plant or allelopathic effects havenot been investigated as far we know.

Physiological and Evolutionary Level. In field and pot studies Acacia pulchellaroots suppressed the soil fungus Phytophthora cinnamomi and promoted thesurvival of Eucalyptus marginata trees. This fungus is an important pathogen forEucalyptus. Investigating the Acacia-Phytophthora-Eucalyptus interaction,Whitfield et al. (51) found C$2 to be a major constituent of the Acacia rootvolatile compounds.

66 BIOGENIC SULFUR IN THE ENVIRONMENT

Carbon disulfide emissions are known from only five plant species: mincedleaves otBrassica oleraceae (52). intact leaves ofMedicago sativum L., Zea mays,L., Quercus lobata Nee (27) and steam distilled roots of Acacia pulchella (5.1).Continuation of surveys of plant species for sulfur gas emission will expand ourunderstanding of the phylogenetic and biogeographic distribution of the sulfuremission phenomenon.

The physiological controls of CS2 emission from plants apparently areunknown. We offer two interrelated suggestions about controls to CS2emission. Removal of CS2 from around the roots either by extracting rootsfrom soil (Figures 2 & 3) or by vigorous washing (Figure 2) or by gentle rinsing(Figure 3) leads to rapid CS2 emission followed by relatively steadyconcentration (Figures 2 & 3). This suggests a feedback mechanism in whichlowered CS2 concentration at the root surface leads to rapid CS2 release fromthe root.

A related control of CS2 release from 5. excelsum roots is disturbance. Inthe present study the gentle disturbance of rinsing with water stimulated CS2release (Figure 3). More vigorous washing stimulated higher rates of release(Figure 2). In the field, the odor of CS2 became stronger during the beginningof a rain storm. Given the toxicity of CS2, to nematodes and insects and thestimulation of CS2 release by disturbance, we suggest that disturbance to S.excelsum roots by root-eating nematodes and insects stimulates CS2 productionas a defense mechanism.

The emission of CS2 from these legumes may also be a mechanismremoving excess SC>4"2 from the rhizosphere where SO4-

2 might otherwiseinhibit the uptake of the Mo required for nitrogen fixation. Rennenberg (31)argues that H2S emission from plants is a mechanism against excess sulfuraccumulation in plants. While Rennenberg's study (31) dealt with H2S,emission of CS2 and other sulfur gases could have similar roles. Cole et al (53)showed that SCv2 can inhibit Mo uptake by algae and bacteria. Sulfuremissions from these trees may have multiple selective advantages.

Acknowledgements

Research supported by U.S. National Science Foundation grants BSR8104700,BSR8012093 and by U.S.D.A. Forest Service, Southeastern ForestExperimental Station, Coweeta Hydrologic Lab., Otto, North Carolina to theUniversity of Georgia. We thank the Organization for Tropical Studies forlogistical support. We thank L. Boring, D. Coleman, B. Michel, C. Monk, andM. Watwood for criticism of the manuscript. Literature on the pathologicaleffects of CS2 was provided by D. Freckman, S. Garabedian and R. Gross towhom we are grateful.

Literature Cited

1. Andreae, M. O.; and Raemdonck, H. Science 1983, 221. 744-7.

2. Kritz, M. A. Journal of Geophysical Research 1982, 87(C11), 8795-803.

3. Barnard, W. R.; Andreae, M. O.; Watkins, W. E.; Bingemer, H.; Georgii, H.

W. Journal of Geophysical Research 1982, 87(C11) 8787-93.

4. Andreae, M. O. In Biogeochemistry of Ancient and Modern Environments:

Trudinger, P. A.; Walter; M. R; Ralph, B. J., Eds; Australian Academy of

Science: Canberra, 1980; pp 253-9.5. Graedel, T. E. Geophysical Research Letters 1979, 6(4), 329-31.

5. HAINES ET AL. Sulfur Emissions from Roots of a Rain Forest Tree 67

6. Slatt, B. J.; Natusch, D. F. S.; Prospero, J. M.; Savoie, D. L. Atmospheric

Environment 1978,12, 981-91.

7. Steudler, P. A.; Peterson, B. J. Atmospheric Environment 1985, 19(9),

1411-6.

8. Steudler, P. A.; Peterson, B. J. Nature 1984, 311(5985), 455-7.

9. Jorgensen, B. B.; Okholm-Hanson, B. Atmospheric Environment 1985,19,

1737-49.

10. Hitchcock, D. R.; Black, M. S. Atmospheric Environment 1984,18,1-17.

11. Ingvorsen, K.; Jorgensen, B. B. Atmospheric Environment 1982,16(4),

855-65.

12. Bandy, A. R.; Maroulis, P. J.; Bonsang, B.; Brown, C. A. In Estuarine

Comparisons: Kennedy, V. S., Ed.; Academic Press: New York, 1982; pp

303-12.

13. Aneja, V. P.; Overton, J. H., Jr.; Cupitt, L. T.; Durham, J. L.; Wilson, W. E.

Tellus 1979,31,174-8.

14. Aneja, V. P.; Overton, J. H, Jr.; Cupitt, L. T.; Durham, J. L.; Wilson, W. E.

Nature 1979, 282(5738), 493-6.

15. Zinder, S. H.; Doemel, W. N.; Brock, T. D. Applied and Environmental

Microbiology 1977,34(6), 859-60.

16. Maroulis, P. J.; Bandy A. R. Science 1977,196. 647-8.

17. Brinkmann, W. L. R; Santos, U. de M. Tellus 1974, XXW1-2). 261-7.

18. Adams, D. F.; Farwell, S. O.; Pack, M. R.; Robinson, R. Journal of the Air

Pollution Control Association 1981,3.1(10), 1083-9.

19. Adams, D. F.; Farwell, S. O.; Pack, M. R.; Bamesberger, W. L. Journal of

the Air Pollution Control Association Journal 1979,29(4), 380-3.

20. Delmas, R.; Baudet, J.; Servant, J.; Baziard, Y. Journal of Geophysical

Research 1980, 85(C8) 4468-74.

21. Delmas, R.; Servant, J. Tellus 1983,35B. 110-20.

22. Delmas, R.; Baudet, J.; Servant, J. Tellus 1978,30,158-68.

23. Farwell, S. O.; Sherrard, A. E.; Pack, M. R.; Adams, D. F. Soil Biology &

Biochemistry 1979, H, 411-5.

24. Banwart, W. L; Bremner, J. M. Soil Biology & Biochemistry 1976, £, 19-22.

25. Banwart, W. L.; Bremner, J. M. Soil Biology & Biochemistry 1976, S,

439-43.

26. Banwart, W. L.; Bremner, J. M. Soil Biology & Biochemistry 1975, 7,

359-64.

68 BIOGENIC SULFUR IN THE ENVIRONMENT

27. Westberg, H; Lamb, B. In Environment Impact of Natural Emissions:

Aneja, V. P., Ed.; Air Pollution Control Association, Specialty Conference,

Transactions, TR-2: Pittsburgh, PA, 1984; pp 41-53.

28. Hallgren, I; Fredriksson, S. Plant Physiology 1982, 70,456-9.

29. Winner, W. E.; Smith, C. L.; Koch, G. W.; Mooney, H. A.; Bewley, J. D.;

Krouse, H. R. Nature. 1981,289(5799), 672-3.

30. Spaleny, J. Plant and Soil 1977, 48, 557-63.

31. Rennenberg, H. Annual Review of Plant Physiology 1984, 35,121-53.

32. Mac Taggart, D.; Adams, D.; Farwell, S. Journal of Atmospheric Chemistry

1987, 5, 417-37.

33. Goldan, P.; Kuster, W.; Albritton, D.; Fehsenfeld, F. Journal of

Atmospheric Chemistry 1987, 5,439-67.

34. Lamb, B.; Westberg, H.; Allwine, G.; Bamesberger, L.; Guenther, A.

Journal of Atmospheric Chemistry 1987, 5, 469-91.

35. Smil, V. Carbon-Nitrogen-Sulfur, human interference in grand biospheric

cycles: Plenum Press: New York, 1985.

36. Ivanov, M. V.; Freney, J. R The Global Biogeochemical Sulfur Cycle: John

Wiley and Sons: New York, 1983.

37. Haines, B.; Black, M.; Fail, J., Jr.; McHargue, L.; Howell, G. In Effects of

Atmospheric Pollutants on Forests. Wetlands, and Agricultural Ecosystems:

Hutchinson, T. C.; Meema, K. M. Eds.; Springer-Verlag: New York, 1987;

pp 599-610.

38. SAS Institute, Inc., Gary, NC, 1985.

39. Hartshorn, G. S. In Costa Rican Natural History: Janzen, D. E., Ed.; The

University of Chicago Press: Chicago, IL 1983; pp 118-350.

40. Johnson, D. W.; Cole, D. W.; Gessel, S. P. Biotropica 1979,11, 38-42.

41. Haines, B. L. Oikos 1983,41, 139-43.

42. Rodgers, G. A.; Ashworth, J.; Walker, N. Zentralblatt fur Bakteriologie.

Parasitenkunde. Infektionskrankheiten und Hygiene 1980,135.477-83.

43. Malhi, S. S.; Nyborg, M. Plant and Soil 1982, 65(2), 203-18.

44. Bremner, J. M.; Bundy, L. G. Soil Biology & Biochemistry 1974, 6, 161-5.

45. Bliss, D. E. Phytopathology 1951. 41. 665-83.

46. Munnecke, D. E.; Kolbezen, M. J.; Wilburn, W. D. Phytopathology 1973,

63., 1352-7.

47. Filip, G. M.; Roth, L. F. Canadian Journal of Forest Research 1977, 7,

226-31.

5. IIAINESETAL. Sulfur Emissions from Roots of a Rain Forest Tree 69

48. Carter, E. P.; Betz, D. O. Jr.; Mitchell C. T. Annotated index of registered

fungicides and nematicides: their uses in the United States: Pesticides

Regulation Division, Agricultural Research Service; U. S. Department of

Agriculture, 1969.

49. Chandler, W. A. Plant Disease Reporter 1969, 53(1), 49-53.

50. Punj, G. K.; Girish, G. K. Journal of Stored Products Research 1969, 4,

339-42.

51. Whitfield, F. B.; Shea, S. R.; Gillen, K. J.; Shaw, K. J. Australian Journal of

Botany 1981, 29,195-208.

52. Bailey, S. D.; Bazinet, M. L.; Driscoll, J. L.; McCarthy, A. I. Journal of

Food Science 1961, 26., 163-70.

53. Cole, J. J.; Howarth, R. W.; Nolan, S. S.; Marino, R. Biogeochemistry 1986,

2, 179-96.RECEIVED December 27,1988

Reprinted from ACS Symposium Series No. 393Biogenic Sulfur in the EnvironmentEric S. Saltzman and William J. Cooper, EditorsCopyright « 1989 by the American Chemical SocietyReprinted by permission of the copyright owner