Persistence and antibiotic immunity of bacteria from a wetland used as a medical waste landfill.

Disclaimer

The information in this document has been funded wholly (or in part) by the United States Environmental Protection Agency (U.S. EPA). It has been subjected to the U.S. EPA's peer and administrative review, and it has been approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.

Introduction

Currently, three of 50 EPA Superfund sites in Washington State are former landfills. The Tulalip landfill in Marysville, Washington was the only site contracted to accept unsterilized pathological wastes from Seattle area hospitals from 1964 to 1978. This site is of particular interest to both EPA and the general public since it is also a marine wetland.

Municipal waste entering landfills may contain an undetermined number of bacteria, viruses, and parasites capable of causing disease in humans and animals. Previous landfill studies have focused on numbers of indicator bacteria (total coliforms, fecal coliforms, and fecal streptococci) or enteroviruses in feces and leachates, but have not estimated the total pathogen load or identified the major sources of these pathogens. This may be due to lack of detection methods for many of the pathogens in solid waste or leachate as well as potentially low numbers of viable but non-culturable organisms (1). Moreover, the isolation of viable organisms from a landfill site only demonstrates that they survive under landfill conditions and does not imply that the organisms are active in situ (2).

When solid waste is placed in a landfill, it begins to adsorb moisture depending on climatic conditions and the construction of the landfill. Leachate movement does not start until the waste has become saturated. Significant leachate movement does not commence for an estimated two to five years after landfill closure (3). In cases where water moves within a poorly constructed or unlined landfill, the leachate can transport pathogens outside the landfill (4). These low levels of pathogens can become a potential health risk by contaminating groundwater sources or bioaccumulating in sediment and biota, such as shellfish (5).

The Tulalip landfill [ILLUSTRATION FOR FIGURE 1 OMITTED] has been identified in previous investigations as a potential human health hazard and threat to fish and bivalves in adjoining wetlands (6). This conclusion was partially based on the fact that during its operation, non-sterilized medical and laboratory wastes were disposed at this site without proper containment. As a consequence, leachate flowed unrestricted into surrounding sloughs. During its operation, the [TABULAR DATA FOR TABLE 1 OMITTED] site lacked any semblance of a "sanitary" landfill and could be more accurately termed an open wetlands dump. Municipal and hospital waste was transported to the landfill by barge. Refuse was unloaded from the barges during high tide within barge canals [ILLUSTRATION FOR FIGURE 2 OMITTED], and crude nets were used to keep debris from being transported into Puget Sound by ebbing tides.

Similar studies at other landfill sites have found high and sustained levels of both bacterial indicators and opportunistic pathogens in leachate more than nine years following landfill closure (7). Elevated levels of some bacteria at the Tulalip site have been sustained far beyond this period of time, probably due to the rich nutrients and neutral pH of the wetland milieu as well as a constant oxygen recharge via daily tidal fluctuations. Bacterial resistance to antibiotics has been demonstrated in terrestrial and aquatic environments at other sites, particularly those contaminated with wastes from hospitals (8). Based on findings from the initial site investigations, site analyses included the identification of four common opportunistic pathogens and antibiotic susceptibility testing with common clinical antibiotic and chemotherapeutic agents used at Seattle area hospitals in the 1970's.

Pseudomonas aeruginosa is the bacterium responsible for ear and urinary tract infections, blue/green pus infections, dermatitis, folliculitis, "swimmers ear" and "pink eye." In addition to accounting for 60 percent of water contact sport infections, this organism produces enteritis in children (9). The wide use of antibiotics has thrust this organism into the limelight because it is susceptible to only a few of them.

Staphylococcus aureus is carried by one-third to one-half of the healthy population, [TABULAR DATA FOR TABLE 2 OMITTED] especially in the nasal cavity (10). A few individuals constantly carry the organism, but most are transitory carriers. Endemic hospital strains differ from the usual Staphylococcus population because they are generally more resistant to antibiotics, especially penicillin. As a result, synthetic analogs of penicillin have been developed to combat penicillinase-producing S. aureus (10).

Enterococcus faecalis will survive longer in brackish waters than coliforms but die out quickly in soil (5). E. faecalis has been proposed as an index of fecal contamination in bathing waters (11).

Clostridium perfringens is commonly used as an indicator of long-term fecal contamination. C. perfringens forms highly resistant endospores that persist under the extreme conditions of the marine benthic environment (12). This anaerobe is the most frequent cause of gas gangrene in humans. Once exposed to deep wounds, some strains multiply while secreting a potent exotoxin that produces extensive necrosis and considerable gas in tissues (13). Additionally, ingestion of enterotoxin-producing strains of C. perfringens in contaminated foods has been linked to gastro-enteritis.

The objectives of these investigations were to 1) monitor population trends of these bacteria in various matrices over time and 2) delineate changes in the antibiotic resistance profile of environmental isolates compared with a clinical strain of the same bacterium.

Materials and Methods

Sampling

Samples were collected in the field employing aseptic techniques. All sample containers were sterilized prior to field activities. A 15 percent ethylenediamine-tetraacetic acid dihydrate (EDTA) solution (J.T. Baker, Phillipsburg, NJ) was dispensed into all 250-mL polyethylene Nalgene sample bottles (Sybron Corp., Rochester, NY) as a chelating agent for heavy metals in seep and surface water samples. Sediments were secured using a sterile stainless steel clamshell and placed in pre-sterilized plastic specimen cups. Unbroken shellfish were collected using hand shovels and placed in clean 4 liter wide-mouth glass jars with teflon-lined lids. All samples were stored on ice (3 [degrees] C) and delivered to the laboratory for analysis within eight hours of collection. Analysis was initiated immediately following sample receipt.

Identification and Enumeration Procedures

Total coliform, fecal coliform, and fecal streptococci were enumerated employing the five-tube multiple-tube fermentation (MTF) test as described in Standard Methods (14). Briefly, 1:10, 1:100, 1:1,000, 1:10,000, and 1:100,000 dilutions were prepared, inoculated into lauryl tryptose broth (LTB) (Difco Laboratories, Detroit, MI), and incubated at 35 [degrees] C for 24 to 48 hours. Sterile wood applicator sticks were used to transfer broth from positive LTB tubes to brilliant green bile broth (BGB) (BBL Microbiology Systems, Cockeysville, MD) and Escherichia coli with 4-methylumbelliferyl-[Beta]-D-glucuronide (EC-MUG) (Difco) culture tubes with inverted fermentation vials. BGB tubes were incubated at 35 [degrees] C for 24 to 48 hours. EC-MUG tubes were incubated in a 44.5 [degrees] C water bath for 24 hours. Gas formation within the inverted vials confirm the presence of fecal coliforms. Ultra-violet light was used to determine the fluorescence of the culture tube media. Fluorescence confirms the presence of E. coli.

From the serial dilutions, aliquots were dispensed to azide-dextrose broth tubes (BBL) and incubated at 35 [degrees] C for 24 to 48 hours. Using sterile applicator sticks, all positive (turbid) fecal streptococci tubes were streaked on enterococcosel agar (BBL) plates and incubated at 35 [degrees] C for 24 hours to determine the presence of presumptive Enterococcus faecalis colonies. Typical colonies were picked from each plate and transferred to culture tubes of brain heart infusion broth (BHI) (Difco) and BHI with 6.5 percent NaCl, then incubated in a 44.5 [degrees] C water bath. If an isolate produced turbid growth in BHI and BHI with NaCl, the isolate was confirmed as Enterococcus faecalis using a BIOLOG[R] Microstation (BIOLOG[R], Hayward, CA).

Pseudomonas aeruginosa and Staphylococcus aureus were enumerated according to the membrane-filter (MF) method described in Standard Methods (14). Ten-fold serial dilutions of each sample were passed through a 0.45 [[micro]meter] porosity filter (Millipore, Bedford, MA) in duplicate to capture cells. Using sterile forceps, filters were placed in 50 x 9 mm plates (Becton Dickinson) with MPA agar and incubated 72 hours at 41.5 [degrees] C. Each target colony was examined for casein hydrolysis on skim milk agar (BBL). In like manner, a filter was placed in plates containing m-staph broth (Difco) and incubated at 35 [degrees] C for 48 hours. After typical yellow colonies were counted, a representative number of pigmented colonies were picked for further confirmation. Colonies that were catalase positive, coagulase positive and/or DNase positive on DNase agar (Difco) and Gram positive, demonstrating typical microscopic morphology, were scored as confirmed isolates.

Samples were analyzed for Clostridium perfringens employing the pour-plate overlay method as described in the FDA BAM (15). Briefly, 1 mL aliquots of each dilution were pipetted to tryptose-sulfite-cycloserine (TSC) agar plates, then overlayed with 15 mL of TSC at 50 [degrees] C, and incubated in an anaerobic chamber with [H.sub.2] and C[O.sub.2] generator envelopes (BBL) for 24 hours at 35 [degrees] C. Typical colonies were transferred to sporulation broth and confirmed by Gram stain (VWR Scientific, Media, PA) and in Fe-milk, motility-nitrate, and lactose-gelatin media. Confirmed C. perfringens isolates were tested for enterotoxin production using an OXOID Reverse Passive Latex Agglutination (RPLA) test kit (Unipath, Basingstoke, England). A 16-18 mL volume of modified Duncan and Strong Medium was inoculated with 1 mL of an 18-hour sporulation broth culture, then incubated at 37 [degrees] C for 24 hours to promote enterotoxin production. The medium was then filter-sterilized by passing through a 0.2 [[micro]meter] - 0.45 [[micro]meter] low protein-binding cellulose acetate syringe membrane filter (Sybron Corp., Rochester, NY). The resulting filtrate was used for the enterotoxin assay according to manufacturer specifications.

Shellfish were brushed with a sterile nylon brush immediately prior to shucking and analysis to avoid cross-contamination of the tissue and liquor contents. A 200 g composite sample mass (approximately 20 shellfish) was secured from a shellfish bed adjacent to the wetland and blended for 60-90 seconds. Indicator and C. perfringens analyses were performed using the resulting homogenate according to APHA protocols (16).

Antibiotic Sensitivity Testing

Antibiotic resistance was determined according to the Kirby-Bauer disk agar diffusion procedure (17). A total of 10 isolates per sample were evaluated. Briefly, an 18-hour pure culture grown in trypticase soy broth (TSB)(Difco) at 35 [degrees] C was swabbed onto Mueller-Hinton agar (BBL) to achieve a lawn of confluent growth. Antibiotic disks of standard concentrations were placed on each plate using a susceptibility disk dispenser (BBL). Plates were inverted and incubated at 35 [degrees] C for 24 hours, and the "zones of inhibition" were measured in mm. C. perfringens antibiotic resistance was assayed using a modified Kirby-Bauer method. C. perfringens isolates were spread on TSC plates and the disks dispensed just prior to pouring a 15 mL overlay of 50 [degrees] C tempered TSC. The plates were incubated upright in an anaerobic chamber at 35 [degrees] C for 24 hours.

Results

Overall, both total and fecal coliform counts decreased significantly in Ebey slough water samples from 1974 to 1988, while fecal streptococci showed a concomitant decrease between 1976 and 1988 (Table 1). Fecal coliform and fecal streptococci counts decreased dramatically from 1974 to 1994. E. coli demonstrated no biologically significant changes between the years 1988 and 1994 (Table 2). As indicated in Table 3, total coliform counts from four of the eight on-site leachate seeps increased over time. Because total coliforms are highly ubiquitous, these increases cannot be exclusively linked to landfill wastes. Fecal coliform and E. coli populations declined significantly during the period between 1988 and 1994. In contrast to the coliform data, fecal streptococci populations were heightened during this same six-year period in five of the eight seeps.

During the U.S. Public Health Service and U.S. EPA field investigations at the landfill in the 1970's, P. aeruginosa was elevated slightly above background in slough waters around the former barge canal entrance. The greatest number of isolates occurred in samples taken at the bottom of the water column (Table 1). In 1988 P. aeruginosa was present at low levels in slough water, sediment, and leachate from the southern berm (Tables 1-3). In August 1994, P. aeruginosa was not recovered from leachate samples. Leachate and slough sediments also were sampled in November of 1994. Again, P. aeruginosa was not isolated from leachate; however, slough sediment samples secured adjacent to the north side of the landfill yielded colonies of the bacterium that expressed the typical resistant phenotype.

During the 1976 investigation, S. aureus populations were elevated in barge canal waters (Table 1). In 1988, increasing numbers of S. aureus also were found in some leachate seep locales (Table 3). Water column samples near the mouth of the barge canal contained S. aureus isolates with a slightly heightened resistance compared with background site isolates (Table 4). S. aureus was isolated from leachate seeps surrounding the landfill perimeter in August 1994, but was detected only in one leachate seep during the November 1994 sampling event. These environmental isolates expressed a decreased resistance to antimicrobial agents compared to a S. aureus clinical strain. S. aureus was not recovered from slough sediments during 1994 sampling events, including those collected near the entrance to the old barge canal.

The presence of Enterococcus faecalis was not investigated at the landfill until August 1994 (data not shown). E. faecalis was isolated from leachate samples in August and November [TABULAR DATA FOR TABLE 3 OMITTED] 1994. These cultures (10 total) revealed a pattern of resistance almost identical to the clinical (+) control. Furthermore, two of four sediment samples contained high counts of E. faecalis whose resistance pattern also closely matched the (+) clinical control. Clostridium perfringens populations were more abundant from seeps along the south and east landfill berm (Table 3). Recently, two of three sediment samples yielded C. perfringens resistant to nine of 12 antimicrobial agents (data not shown). The pattern of resistance closely matched that of the (+) clinical control.

One composite sample of juvenile Pacific Gaper clams (Tresus nuttallii) was collected approximately 300 feet from the northwest corner of the adjacent wetland during low tide. This species is commonly harvested in Puget Sound. Although adult Pacific Gapers are not known to inhabit the area surrounding the former landfill, recreational fisherman may harvest juveniles in the area as they range in size from 4.5-9.0 cm. The State of Washington's maximum contamination level (MCL) standard for fecal coliforms in shellfish is 230 MPN/100g tissue in nondepurated shellfish. Shellfish collected adjacent to the landfill harbored fecal coliform levels at 790 MPN/100g, well in excess of the standard.

Discussion

Coliform and fecal streptococci bacteria are termed "indicators" since their detection in a particular matrix often indicates the presence of more harmful bacterial pathogens and/or viruses. The coliform family is highly ubiquitous since all warm-blooded animals (especially humans) shed them in feces. Although wildlife populations will contribute a small proportion of fecal coliforms to the environment, they shed much higher proportions of fecal streptococci than human hosts. Hence, if fecal coliform counts are elevated, the source is usually assumed to be human fecal waste, while heightened fecal streptococci counts are a likely function of the wildlife (non-human) populations that inhabit the site.

Historically, the greatest source of the bacterial contamination at the Tulalip landfill has been at the end of the barge canal where syringes, tongue depressors and gauze pads could be seen floating in the water [ILLUSTRATION FOR FIGURE 2 OMITTED]. The higher concentration of P. aeruginosa in bottom sediments is worth noting since most Pseudomonads are aerobic, however, P. aeruginosa can respire anaerobically by using nitrate as an electron acceptor. These data seem to suggest a release of P. aeruginosa contaminated leachate below the intertidal zone. The apparent abundance of S. aureus in leachate is likely attributed to the warmer seasonal temperatures that often accompany logarithmic increases in overall bacterial populations.

The Tulalip landfill was prohibited from accepting medical wastes in 1978. The waters adjacent to the landfill are currently closed to shellfish harvesting. The landfill is now capped and the barge canals plugged and filled. These containment measures will likely eliminate any direct exposure to the areas at the end of the barge canal where pathogen counts were the highest. Indirect exposures may occur, however, via trespasser contact with sediments and harvesting or ingestion of impacted shellfish (18). The accumulation of some enteric pathogens in sediments is largely due to a high organic matter which furnishes a niche for long-term survival. With Escherichia coli and [TABULAR DATA FOR TABLE 4 OMITTED] Pseudomonas aeruginosa, the expression of genes responsible for osmoregulation has been found to be enhanced in marine sediments containing organic matter (19). This may explain, at least partially, why sediments act as reservoirs for pathogens (20, 21).

Extrachromosomal plasmids carry the genetic information responsible for the phenotypic expression of antibiotic resistance (22). Plasmids can be passed from cell to cell in dense populations as well as being replicated in the progeny; thus, an entire population can quickly acquire the resistance characteristic (23). Walia et al. conducted a comparative analysis of antibiotic resistance and plasmid incidence among Gram-negative bacteria isolated from a landfill and patients with urinary tract infections (24). Of the clinical strains, 34 percent contained plasmids, as compared to 50 percent of the environmental strains. Sixty-seven percent of the plasmid-bearing environmental strains were resistant to three or more antibiotics. The antibiotic resistance and presence of large plasmids among related clinical and environmental isolates show significant correlation and present a real threat of spreading antibiotic resistance and catabolic genes by intergenetic transfer. Researchers also have noted that successful Gram-negative recipients of these plasmids are more likely to transfer the resistance factor than were the original donors (25,26). Moreover, the nutrient availability of the environment in which conjugation events occurred is a more important factor in its success than the genus and species of the recipient organism, making the effects of resistance transfer more broad-ranging than is often realized (27).

In earlier investigations, the level of C. perfringens in sediments surrounding the landfill was significantly elevated above background for the Puget Sound area. Although C. perfringens is ubiquitous in marine sediments, its presence from landfill samples is indicative of continued anaerobic decomposition and long-term organic contamination. Antibiotic resistant C. perfringens have only been isolated from slough sediments.

Overall, berm seeps from the south and east sides of the landfill continue to produce higher bacterial indicator counts. This would appear to make sense, since these segments of the landfill contain the latest waste deposition, which is 10 years more recent than refuse on the north side of the landfill. In contrast, during the most recent investigation, sediments along the north side of the landfill harbored significantly higher pathogen populations. The sample point yielding the highest count was near the entrance to the old barge canal. These results suggest that the former canal(s) may be functioning as a "path of least resistance" for migration of contaminated leachate into slough sediments.

Leachate and sediments high in S. aureus, P. aeruginosa and C. perfringens can promote infection through direct contact with skin, eyes and external orifices. Any such exposure(s) is(are) difficult to quantify, as one is required to measure concomitant increases in morbidity rather than mortality. Opportunistic pathogens are not expected to impact edible fish tissues directly, but ingestion of whole fish and handling of infected fish and shellfish may increase the risk of gastrointestinal or dermal infection.

The source of the fecal contamination found in Pacific Gaper shellfish is unknown. Because the site is heavily influenced by tidal fluctuations, fecal contamination could easily be introduced to these impacted shellfish beds by off-site sources (i.e., wastewater treatment facilities, houseboat sewage disposal); therefore, this contamination cannot be conclusively attributed to the landfill. Because the site and surrounding adjacent property are part of the Tulalip tribal reservation, commercial fishing practices are prohibited in the area. Highly visible warning signs have been erected around the perimeter of the former landfill to discourage trespassing. Hence, although the fecal contamination in these shellfish far exceed the State of Washington Maximum Contaminant Level (MCL) for shellfish meats, they are not expected to pose a significant risk to humans based on the restricted public access to the site.

Conclusions

The results of this study showed that 1) although bacterial populations have decreased with time, the rate of this decrease was highly torpid in contrast to typical landfill environments where pathogens usually persist no longer than 10 years following landfill closure, and 2) studies on antibiotic resistance of pathogen isolates are highly variable over time depending on the pathogen being tested. Any changes in phenotypic expression of resistance factors is likely to be a result of natural environmental competition, rather than catabolic gene transfer from related clinical populations that somehow persisted for 20-30 years. Exposure to these impacted matrices may be significant as the treatment of environmentally acquired bacterial infections becomes increasingly more difficult.

The primary public health concern associated with the widespread establishment of multiple drug resistance among native bacterial populations is the threat this possesses to present and future antibiotic therapy. If native wild strains acquire the same degree of resistance to drugs presently employed for treatment of nosocomial infections, conventional therapy will become ineffectual for patients exposed to environmental strains of bacterial pathogens. South Africa and Central America have experienced serious pandemics associated with enteric pathogens possessing multiple resistance to such antibiotics as penicillin, tetracycline and chloramphenicol (28,29).

Based on the close proximity of slough sediments, and the predicted volume of leachate effluent released below the intertidal zone, antibiotic resistant strains may have found a niche for survival in the marine sediments surrounding the former landfill.

Acknowledgements

We would like to thank Linda Rosa, Lockheed-Martin Computer Support Technician, for her graphics assistance.

REFERENCES

1. Sobsey, M. D., J. L. Melnick and C. Wallis (1974), "Development of Methods for Detecting Viruses in Solid Waste Landfill Leachates," Applied Microbiol., 28(2):232-238.

2. Senior, E. (1995), Microbiology of Landfill Sites, 2nd ed., Boca Raton, Fl: Lewis Publishers.

3. Pahran, H. R. (1987), "Microorganisms in Municipal and Solid Waste and Public Health Implications," CRC Critical Review Envtl. Control, 17(3):187-228.

4. Huber, S. H., M. Abbaszadegan and C. P. Gerba (1994), "Study of Persistence of Enteric Viruses in Landfilled Disposable Diapers," Envtl. Sci. Technol., 28(9):1767-1772.

5. Keswick, B. H. and C. P. Gerba (1980), "Viruses in Groundwater," Environ. Sci. Technol., 14(11):1290-1297.

6. Vasconcelos, G. J. (1988), "The Tulalip Landfill Revisited," unpublished data, U.S. EPA, Seattle, Wa.

7. Brunner, D. (1983), Field Assessment of Site Closure, Boone County, Kentucky, EMCON Associates, EPA-600/2-83-058.

8. Bitton, G. J. (1994), Envtl. Microbiology, New York, NY: Wiley and Sons.

9. Vasconcelos, G. J. and N. C. Anthony (1985), "Microbiological Quality of Recreational Waters in the Pacific Northwest." J. of the Water Pollution Control Federation, 57(5):366-377.

10. Anderson, D. A. (1973), Introduction to Microbiology, Saint Louis, MO: C. V. Mosby.

11. Havalaar, A. H. (1993), A Bacteriophage Standard for Bathing Waters, Final Technical Report, Commission of the European Communities, Bilthoven, Netherlands.

12. Hill, R. T., M. S. Anikis, R. R. Colwell and I. T. Knight (1993), "Benthic Distribution of Sewage Sludge Indicated by Clostridium perfringens at a Deep-Ocean Dump Site," Appl. Envtl. Microbiol., 59(1):47-51.

13. Salle, A. J. (1948), Fundamental Principles of Bacteriology, New York, NY: McGraw-Hill.

14. Am. Pub. Health Ass'n. (APHA) (1989), Standard Methods for the Examination of Water and Wastewater, 18th ed., APHA, Washington, D.C.

15. U.S. Food & Drug Admin. (1978), Food & Drug Administration Bacteriological Analytical Manual, 5th ed., Association of Official Analytical Chemists, Washington, D.C.

16. Am. Pub. Health Ass'n. (APHA) (1984), APHA Laboratory Procedures for Examination of Seawater and Shellfish, 5th ed., APHA, Washington, D.C.

17. Bauer, A. W., W. M. Kirby, J. C. Sherris and M. Turck (1966), "Antibiotic Susceptibility Testing By a Standardized Single Disk Method," Am. J. Clin. Pathol., 45(4):493-496.

18. Boening, D. W. and G. J. Vasconcelos (January 26-28, 1994), "Persistence of Antibiotic Resistive Bacteria in a Wetland Used as a Medical Waste Landfill," Proceedings of the 1st Annual ICF Kaiser Technical Conference, Denver, Co.

19. Grant, W. D. and P. E. Long (1981), Envtl. Microbiology, New York, NY: Wiley and Sons.

20. D'Souza-Ault, M. R., L. T. Smith and G. M. Smith (1993), "Roles of N-Acetylglutaminylglutamin Amide and Glycine Betaine in Adaptation of Pseudomonas aeruginosa to Osmotic Stress," Appl. Envtl. Microbiol., 59(2):473-478.

21. Gauthier, M. J. and V. A. Breittmayer (April 1-6, 1990), "Regulation of Gene Expression in E. coli Cells Starved in Seawater: Influence on their Survival in Marine Environments," International Symposium on Health-Related Water Microbiology, Tubingen, West Germany.

22. Vasconcelos, G. J. (1988), "The Tulalip Landfill - A Microbiological Perspective," unpublished report, U.S. Environmental Protection Agency, Seattle, Wa.

23. U.S. Environmental Protection Agency (1990), Microbial Decomposition of Chlorinated Aromatic Compounds, EPA/600/2-86/090, Washington, D.C.

24. Walia, S., A. Khan and N. Rosenthal (May 5-9, 1991), "Antibiotic Resistant Bacteria: A New Potential Health Hazard from Landfills," Proceedings of the 91st ASM General Meeting, Dallas, Tx.

25. Bell, J. B., G. E. Elliot and W. R. Macrae (1980), "Incidence of R factors in Coliform, Fecal Coliform, and Salmonella Populations of the Red River in Canada," Appl. Envtl. Microbiol., 40(3):486-491.

26. Shaw, D. R. And V. J. Cabelli (1980), "R-plasmid Transfer Frequency From Environmental Isolates of Escherichia coli to Laboratory Strains and Fecal Strains," Appl. Envtl. Microbiol., 40(4):756-767

27. McKeon, D. M., G. K. Bissonnette and J. P. Calabrese (1995), "Antibiotic Resistant Gram-Negative Bacteria in Rural Groundwater Supplies," Water Research, 29(8):1902-1908.

28. Grabow, W. O. K. (1976), "South African Experience on Indicator Bacteria, Pseudomonas aeruginosa and R+ Coliforms in Water Quality Control," ASTM International Symposium on Bacterial Indicators of Potential Health Hazards, Chicago, Il.

29. Gangarosa, E. J., J. V. Bennett, M. D. Bessudo, P. M. Hernandez, J. Olarte, P. E. Pierce, V. Vasquez, and C. Wyatt (1972), "An Epidemic Associated Episode?" J. Infect Dis., 126(2):215-218.

Corresponding Author: Dean W. Boening, M.S., Senior Toxicologist, Lockheed-Martin Services Group, Port Orchard, WA 98366 and Dept. of Science, Olympic College, Bremerton, WA 98337.