| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Crop and Soil Environmental Sciences, Virginia Polytechnic Inst. and State Univ., Blacksburg, VA 24061-0404
b Clarke County Planning Office, Berryville, VA 22611
* Corresponding author (chagedor{at}vt.edu)
Received for publication August 22, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: ARA, antibiotic resistance analysis • TMDL, total maximum daily load • VDH, Virginia Department of Health • DEQ, Department of Environmental Quality • CFUs, colony forming units • TSA, trypticase soy agar • DA, discriminant analysis • ARCC, average rate of correct classification • BMPs, best management practices
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Source tracking methodologies will find wide use in the total maximum daily load (TMDL) program (section 303, the Clean Water Act, USEPA, 1999a). A TMDL is a calculation of the maximum amount of a pollutant that a waterbody can receive (the sum of the allowable loads of a single pollutant from all contributing point and nonpoint sources) and still meet water quality standards. The calculation must include a margin of safety to ensure that the waterbody can be used for the purposes the state has designated (e.g., recreational uses, shellfish harvest, drinking water). The calculation must also account for seasonal variation in water quality. Allocations and allowable loads for pollutant sources implies that those sources can be accurately identified, and the inclusion of seasonal variation in the TMDL program will require longer-term fecal source identification studies than most of those reported to date (McClellan et al., 2000; McKenzie, 1998).
Methods to identify sources of fecal pollution are important because fecal contamination of water is still a widespread problem in the USA (USEPA, 1986, 1997). In Virginia, only one-third of >78000 km of streams and rivers have been adequately monitored, and to date 3486 km of streams and 253 km2 of estuaries (of those monitored) are listed as impaired (Friends of the Rivers of Virginia, 2001). The leading cause of impairments (>60%) is violation of the fecal coliform standard, with nonpoint agriculture the most widely suspected (but not proven) source. This situation is typical of many states (USEPA, 1999b).
In a previous report, sources of fecal pollution were determined for Page Brook, a small watershed upstream from Spout Run (Hagedorn et al., 1999). Antibiotic resistance analysis (ARA) of >4000 enterococcal isolates indicated that the majority of the isolates were of livestock (cattle, Bos taurus) origin, and no human signature was found. State regulatory authorities applied the results to fecal coliforms, and fencing portions of the stream to reduce livestock access lowered fecal coliform populations by an average of 94% and demonstrated that ARA had correctly identified the major source of fecal pollution in the stream. Such an interchange of indicators was successful in the previous study; however, until additional research makes it clear that such interchanges are really appropriate, the case could be made that either source tracking should be performed with fecal coliforms or the enterococci should be used as the freshwater indicator of water quality. Interchangeability may not be an issue, because Harwood et al. (2000) reported that enterococcal and fecal coliform databases classified known source isolates with similar accuracies and both databases identified the same sources in water samples and produced the same ranking among the various sources.
Page Brook is one of three tributaries that combine to form Spout Run. The Spout Run watershed contains the community of Millwood, and Spout Run is also the name of the stream that flows through the community. Preliminary sampling of wells by the Virginia Department of Health (VDH) and stream samples by the Virginia Department of Environmental Quality (DEQ) identified Spout Run as contaminated by fecal pollution and resulted in the stream being added to DEQ's impaired stream segment list (Virginia Dep. of Environmental Quality, 2001). These agencies designated Millwood as an at risk community, and Millwood was chosen for this study due to its suitability as a probable source of human contamination in Spout Run.
The objectives of this project were: (i) to use ARA to develop a library of enterococcal profiles from known sources in the Spout Run watershed around Millwood, and classify the known sources as human vs. livestock vs. wildlife; (ii) compare profiles of enterococcal isolates from stream samples (unknown origin) against the library to determine if there was a human signature among the sources of fecal pollution in the stream; and (iii) determine what proportion of the stream isolates were from human, livestock, and wildlife sources.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Water Quality Monitoring
Stream samples were collected monthly (May 1999–May 2000) and analyzed for fecal coliforms, enterococci, and selected water chemistry parameters (Eaton et al., 1995). Samples were transported to the laboratory and filtered within 6 h of collection when possible, but no later than 12 h after collection.
Stream samples were collected from sites in upper Millwood, middle Millwood, and lower Millwood (Fig. 1). Sites located in upper Millwood include FC-61 and FC-62. Sites located in middle Millwood include FC-55 and FC-56. Site FC-58 (lower Millwood) is in the section of the stream that flows through a private preserve (no livestock) just below the community. Site FC-54 (below Millwood) is the section of the stream that flows through a horse farm that contains approximately 120 horses, located 0.45 km downstream from Millwood. Upstream locations outside of Millwood included FC-35, located on Page Brook above its confluence with Roseville Run to form Spout Run, and FC-43, located on Westbrook Run, and FC-44, located on Roseville Run, just prior to where these two streams converge.
Flow rates in Spout Run were determined with a Global Water flow meter model FP201 (Global Water, Gold River, CA), recorded as meters/second, then converted to liters/minute by measuring the stream cross-sectional area (Moberg and Rice, 1999). Turbidity measurements were made with the nephelometric method using a Hanna HI93703 Microprocessor Turbidity Meter (Hanna Instruments, Woonsocket, RI), with a range from 0.00 to 1000 formazin turbidity units, and results were reported as nephelometer turbidity units, or NTUs (Method 2130 B, Eaton et al., 1995). An Orion PerpHecT Meter Model 370 (ATI Orion, Boston, MA) was used with a PerpHecT Ag/AgCl Sure-flow Electrode, Model 9272, to determine pH (Method 4500-H+, Eaton et al., 1995).
Ammonium-N (Method 4500–NH3, Eaton et al., 1995) and nitrate-N (Method 4500–NO3, Eaton et al., 1995) were determined on a dual channel autoanalyzer. The autoanalyzer used was a continuous flow Quick Chem 8000 made by Lachat Instruments (Milwaukee, WI). Orthophosphate (Method 4500-P, Eaton et al., 1995) was determined with a Hitachi Model 100-20 spectrophotometer (Hitachi Instruments Ltd., Tokyo, Japan). Oxygen concentrations for the BOD5 test were measured with a YSI Model 57 Oxygen meter (YSI Inc., Yellow Springs, OH) with a YSI 5905 BOD probe (Method 4500-O, Eaton et al., 1995).
Membrane Filtration for Fecal Coliforms and Enterococci
Water samples (100 mL) were filtered through a 0.45-µm poresize membrane filter. The filters were then transferred to a 50-mm petri dish containing 5 mL of m-FC agar (Baltimore Biologics Laboratory, BBL) for fecal coliforms and 5 mL of m-Enterococcus agar (BBL) for enterococci. The filters were incubated for 24 h at 44.5°C (water bath) for fecal coliforms and for 24 to 48 h at 37°C (air incubator) for enterococci. Dark blue colonies were counted for fecal coliforms and red to burgundy colored colonies were counted as enterococci, and reported as colony forming units (CFUs) per 100 mL for both bacterial indicator groups. Details on isolation and identification of isolates has been previously reported (Wiggins, 1996; Hagedorn et al., 1999).
Antibiotic Resistance Analysis (ARA)
After the enterococci had been enumerated and colonies were available on membrane filter plates, sterile toothpicks were used to transfer individual colonies to 96-well microtiter plates. No set number of colonies was taken from any one plate, but rather colonies were selected randomly and equally (where possible) from the plates that constituted one sample. No more than 20 colonies were taken from any one composite known source fecal sample, whereas 48 were selected from stream samples. The 96-well plates (presterilized) were filled with Enterococcosel broth (Becton Dickinson, Cockeysville, MD) using an 8-channel multiwell pipettor, adding 0.2 mL of broth to each of the wells. Care was taken to ensure that each colony picked was separate and distinct. Each isolate was scraped from the membrane filter plate (or spread plates for known source isolates) with the toothpick and thoroughly inoculated into one well of the 96-well plate. The 96-well plate was placed in a plastic container to prevent the microwells from drying up, and then incubated at 37°C for 24 to 48 h.
The antibiotic-containing plates were prepared by adding filter-sterilized stock solutions (10 mg/mL) in sterile water (cephalothin, neomycin, oxytetracycline, streptomycin), 1:1 water/ethanol (chlorotetracycline, erythromycin, tetracycline, vancomycin), and 1:1 water/methanol (amoxicillin,), to autoclaved and cooled Trypticase Soy Agar (TSA; Becton Dickinson, Cockeysville, MD). Initial concentrations of 2.5, 5, 10, 20, 40, 60, 80, and 100 (g/mL were used to test the sets of antibiotics and concentrations reported by Wiggins (1996) and Hagedorn et al. (1999). The isolates were transferred with a stainless steel 48-prong replica-plater (Sigma Chemical Co., St. Louis, MO) from the Enterococcosel-containing microwells to a set of TSA plates containing the various concentrations of each antibiotic to be tested, and to a control plate containing no antibiotic. The plates were incubated at 37°C for 48 h and growth of each isolate on each concentration of every antibiotic was determined. An isolate was considered to be resistant to a given concentration of antibiotic if growth comparable to the controls occurred on that plate. Any isolate that did not grow on the control plates (containing no antibiotic) was not used in the analysis. To determine sources of fecal pollution, ARA was performed on enterococcal stream isolates on a seasonal basis during the months of May, August, October, and December 1999, and March 2000 to cover seasonal flow levels and water conditions.
Statistical Methods
Data on the ability of each known source isolate to grow in the presence of the various concentrations of the antibiotics were analyzed by Discriminant Analysis (DA) with SAS-JMP statistical software (version 3.2.2, SAS Inst., Cary, NC). Discriminant Analysis has been used to classify enterococcal and E. coli isolates based on source by several investigators (Bower, 2001; Carson et al., 2001; Harwood et al., 2000; Wiggins et al., 1999). Analysis by DA produces a classification set for every known source isolate. The average rate of correct classification (ARCC) is determined by averaging the percentages of correctly classified isolates for each source. A database is built for each known source (e.g., human, cattle, etc.). The DA procedure compares each set of isolates from an unknown source (stream sample) against the database of known sources and then classifies each isolate into one of the possible sources.
Hundreds of discriminant analyses were performed by varying the combination of antibiotics to identify the best combinations to use and the most advantageous way to subdivide the library. The classification table produced by the DA procedure was used to calculate the percentages of misclassified isolates and determine the average rate of correct classification (ARCC). The table is a source-by-source matrix in which the numbers and percentages of correctly classified isolates are found on the diagonal. The ARCC for a given combination of antibiotics was computed by averaging the percentages along the diagonal. The percentage of misclassified isolates for a given source was determined by adding the percentages of misclassified isolates in the appropriate row of the table, excluding the value in the diagonal, as suggested by Harwood et al. (2000).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
The sampling sites with the highest average fecal coliform counts included FC-35, the lower end of Page Brook, FC-56 in middle Millwood, and FC-58 in lower Millwood (Table 3) . Fecal coliform populations were higher for all nine sampling sites during the warm season–low flow months. Counts for the cool season–high flow months averaged below the recreational standard for all sites, and averaged above the standard for all but two sites for the warm season–low flow months. There were no obvious site differences in fecal coliforms when comparing sites above, within, and below Millwood, and fecal coliform populations in the millrace samples (FC-55 and FC-62) exhibited the same trends as the stream samples. Enterococcal populations were higher for seven of nine sampling sites during the warm season–low flow months, but the order of magnitude between the seasonal averages was not as great as with the fecal coliforms. The enterococcal populations were lower than the fecal coliform populations at four of nine sites during the cool season–high flow months and lower at seven of nine sites during the warm season–low flow months. There were two sites in middle (FC-56) and lower (FC-58) Millwood that contained warm season–low flow averages that exceeded 2000 enterococcal CFU/100 mL (Table 3). Seasonal increases in fecal bacteria under low flow conditions are typically a function of less dilution and more direct deposition by wildlife and livestock during the warm weather months (Hagedorn et al., 1999), although regrowth of fecal bacteria in sediments cannot be ruled out (Harwood et al., 2000).
|
The library was developed with 1174 known source isolates, 203 from human and 971 from nonhuman (animal) sources (Table 4) . The ARCC was 95.57% for the human isolates (194 of 203 correctly classified) and 97.01% for the animal isolates (942 of 971 correctly classified). The ARCC for a human vs. animal split of the library was 96.29%. When the library was split three ways (human vs. livestock vs. wildlife), the correct classifications were 94.58% for human (192 of 203 correct), 93.73% for livestock (688 of 734 correct), and 87.76% for wildlife isolates (Table 4). The ARCC for human vs. livestock vs. wildlife was 92.02% (208 of 237 correct).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Wildlife was also problematic in Spout Run since wildlife-origin isolates predominated in 14 of 43 samples (Table 5). If best management practices (BMPs) are implemented to reduce the signatures of livestock and humans, the proportion of fecal loading due to wildlife will probably increase, and may be substantial enough to continue the stream's impaired status. Riparian zones reduce stream access for Canada geese, as they prefer stream access in open pastures. Established riparian zones can provide cover and food for a variety of wildlife [deer, muskrat (Ondatra zibethicus), raccoon (Procyon lotor), opossum (Didelphis marsupialis virginiana), rabbit (Lepus californicus), birds, etc.], can serve as an attractant to such animals, provide sufficient habitat to support large populations in a narrow border along the stream, and could lead to higher wildlife loading rates. Knowledge of the sources of pollution is important in determining the degree of risk for humans exposed to contaminated water and to assist in the development of BMPs to reduce fecal loading. Best management practices that may be appropriate for Spout Run could include stream fencing, establishing riparian-zone buffers, installing in-pasture watering stations, and improving waste treatment options for Millwood.
Antibiotic resistance patterns of enterococcal isolates, analyzed with DA, was a suitable method to differentiate and identify sources of fecal pollution in Spout Run, and the results presented here affirm the use of ARA in previous reports by Bower (2001), Hagedorn et al. (1999), Harwood et al. (2000), and Wiggins et al. (1999). Discriminate analysis was important in determining sources of fecal contamination in stream water where the classified isolates came from multiple sources. The high average rate of correct classification for the known-source isolates from Spout Run was an important finding as it meant that the library could be used to identify unknown-source isolates from stream samples.
ARA Library
The Spout Run watershed was small in size and known sources were intensely sampled for enterococcal isolates from fecal material to obtain the highest possible ARCCs (Table 4). For a known-source library to be able to correctly classify bacteria in an impaired stream segment, accuracy of classification is important, but other issues must be considered as well. The library must contain enough isolates to be representative of the different sources being classified. This is not so much a question of the specific number of isolates needed to provide better source identification and higher average rates of correct classification, but rather a question of representativeness of the library. Perhaps the best way to determine if a library is representative (for any source tracking method) is to regularly add samples of known source isolates to an existing library. If the ARCC and/or the individual correct classifications do not change significantly as new samples are added, then the library should be representative (Hagedorn et al., 1999).
Other investigators have reported ARCC values in the range of 50 to 90+%, and the Spout Run library (three-way combined ARCC = 92.02%, Table 4) is in the upper end of this range. Dombek et al. (2000) reported correct classifications of 78 to 100% on E. coli using repetitive DNA sequences, but the number of isolates tested was small (154 total, avg. of 22 per known source). Harwood et al. (2000) reported 34 to 88% correct classifications with ARA on 4619 enterococcal isolates, and 50 to 95% with ARA on 6144 fecal coliform isolates. Their known-source isolate collection was from a large geographical area in Florida and the lower rates of correct classification for some of the sources reflected that geographic diversity. Wiggins et al. (1999) reported correct classifications from 54 to 91% with ARA on 3032 enterococcal isolates, and demonstrated that the ARCC could be increased substantially by using a larger number of antibiotics and concentrations. Bower (2001) used ARA on 830 enterococcal isolates from a coastal watershed in Oregon and obtained correct classifications of 73% for human isolates and 89% for dairy cattle isolates. Parveen et al. (1999) ribotyped 238 E. coli isolates and reported an 82% correct classification rate when the isolates were classified between human and nonhuman categories. Samadpour and Chechowitz (1995) ribotyped 589 E. coli stream isolates in a 29-mo watershed study and were able to match ribotype patterns (against those in their library) for 71% of the isolates. Source classification on small numbers of isolates is one of the shortcomings of molecular methods as currently used (Dombek et al., 2000; Parveen et al., 1999; Samadpour and Chechowitz, 1995). With ARA (Bower, 2001; Harwood et al., 2000; Wiggins et al., 1999), technicians and students can be quickly taught to perform the procedure on several hundred isolates per week. In polluted streams that yield thousands of fecal coliforms per sample per month, some method is needed that best allows source determinations on a representative subset of the fecal population, whatever that might be (Harwood et al., 2000). To date, ARA appears to be the best method available for source identification on large numbers of isolates.
One goal of future work will be to combine ARA with molecular methods to cross-validate both approaches and to assess where one method might be more suitable than the other. To assess method variability, ARCCs need to be determined on isolates from the same region over some longer period. In a 2-yr study using ARA in the Page Brook, there were no substantial reductions in ARCC for any of the known sources that were included in the library developed for that watershed (Hagedorn et al., 1999). While molecular methods may be more accurate in correctly classifying the specific type of animal (e.g., cows, sheep, deer, waterfowl, etc.), our approach of classifying isolates based on human vs. wildlife vs. livestock has been very useful to regulatory officials in Virginia where ARA has been used in 13 TMDL watershed projects to date (McClellan et al., 2000).
Samadpour and Checkowitz (1995) reported that lack of landowner cooperation was a serious obstacle to obtaining access to property for known source sample collection. The three-way classification in this report has proven to be a nonconfrontational approach that has been readily accepted by the public in the participation component of the TMDL process (McClellan et al., 2000). Landowner cooperation was obtained for every farm and property in the Spout Run watershed where access was desired, and the same level of cooperation was achieved earlier for the Page Brook study (Hagedorn et al., 1999). Also, the three-way classification [dogs (Canis familiaris) or pets could be added as a fourth category for more urban watersheds, and wildlife removed if necessary] allows source classification to be used in the modeling component of the TMDL process (McClellan et al., 2000) where load reduction allocations could be assigned to sources in a watershed based on the proportionality of source classification results (e.g., 10% human, 40% wildlife, and 50% livestock).
Stream Samples
Water chemistry and turbidity values for Spout Run were within the range that would be expected for relatively good quality water and did not appear to be problematic (Table 2). Even during the summer and fall drought when stream flows were reduced (Table 1) and livestock was more active in the stream, water chemistry values were only slightly higher than those from winter. Most noticeable was an increase in BOD5 during the fall, but the highest end of the range (9.6 mg/L) is still far from excessive. The most serious pollution problem in Spout Run was fecal contamination, because all 117 stream samples were positive for fecal coliforms. This result supports the identification of fecal contamination as the most widespread problem in Virginia's rivers and streams (Friends of the Rivers of Virginia, 2001). The TMDL program only allows streams to exceed recreational standards by no more than 10% of monthly samples (5% if a safety factor is included; USEPA, 1999a). Spout Run exceeded Virginia's standard (1000 CFU/100 mL) 31.6% of the time and was appropriately included on the state's impaired stream list.
Due to high fecal coliform counts, Spout Run is not suitable for recreational uses even though it is a designated trout stream and includes a popular swimming area near its confluence with the Shenandoah River. The application of BMPs to improve water quality in Spout Run might include stream fencing and development of riparian buffers to reduce livestock (and some wildlife) access into the stream, and evaluation of options for improving waste disposal and treatment in Millwood.
Fecal contamination of Spout Run was seasonal, with the highest count occurring in the summer and fall, and the lowest in the winter. In July 1999 stream flow levels were reduced by at least 55 to 66% at all locations (Table 1) and livestock were evident wherever there was unrestricted access to the stream. More than 700 cattle (as well as substantial horse and sheep populations) are located on farms in the Spout Run, Page Brook, and Roseville Run basins. These animals (especially cattle) stand in and around the stream for prolonged periods during the summer, and higher fecal bacterial counts in the stream with subsequent degradation of recreational water quality was the inevitable result.
Source tracking methodology has the potential to provide agencies responsible for water quality and public health with a resource to determine sources of fecal contamination. Until sources of pollution can be reliably identified, and the information used in the TMDL process, risk to communities cannot be accurately assessed, and water quality improvement will remain a hit-or-miss affair that is not effective or cost efficient.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. W. Dickerson Jr., J. B. Crozier, C. Hagedorn, and A. Hassall Assessment of the 16S-23S rDNA Intergenic Spacer Region in Enterococcus spp. for Microbial Source Tracking J. Environ. Qual., October 16, 2007; 36(6): 1661 - 1669. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. B. Kaneene, R. Miller, R. Sayah, Y. J. Johnson, D. Gilliland, and J. C. Gardiner Considerations When Using Discriminant Function Analysis of Antimicrobial Resistance Profiles To Identify Sources of Fecal Contamination of Surface Water in Michigan Appl. Envir. Microbiol., May 1, 2007; 73(9): 2878 - 2890. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Stoeckel and V. J. Harwood Performance, Design, and Analysis in Microbial Source Tracking Studies Appl. Envir. Microbiol., April 15, 2007; 73(8): 2405 - 2415. [Full Text] [PDF]
|
|
|
|
|
|
B. Price, E. A. Venso, M. F. Frana, J. Greenberg, A. Ware, and L. Currey Classification tree method for bacterial source tracking with antibiotic resistance analysis data. Appl. Envir. Microbiol., May 1, 2006; 72(5): 3468 - 3475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Soule, E. Kuhn, F. Loge, J. Gay, and D. R. Call Using DNA microarrays to identify library-independent markers for bacterial source tracking. Appl. Envir. Microbiol., March 1, 2006; 72(3): 1843 - 1851. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Anderson, J. E. Whitlock, and V. J. Harwood Persistence and Differential Survival of Fecal Indicator Bacteria in Subtropical Waters and Sediments Appl. Envir. Microbiol., June 1, 2005; 71(6): 3041 - 3048. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Sayah, J. B. Kaneene, Y. Johnson, and R. Miller Patterns of Antimicrobial Resistance Observed in Escherichia coli Isolates Obtained from Domestic- and Wild-Animal Fecal Samples, Human Septage, and Surface Water Appl. Envir. Microbiol., March 1, 2005; 71(3): 1394 - 1404. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. K. Johnson, M. B. Brown, E. A. Carruthers, J. A. Ferguson, P. E. Dombek, and M. J. Sadowsky Sample Size, Library Composition, and Genotypic Diversity among Natural Populations of Escherichia coli from Different Animals Influence Accuracy of Determining Sources of Fecal Pollution Appl. Envir. Microbiol., August 1, 2004; 70(8): 4478 - 4485. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-H. Yang, R. T. Vinopal, D. Grasso, and B. F. Smets High Diversity among Environmental Escherichia coli Isolates from a Bovine Feedlot Appl. Envir. Microbiol., March 1, 2004; 70(3): 1528 - 1536. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Seurinck, W. Verstraete, and S. D. Siciliano Use of 16S-23S rRNA Intergenic Spacer Region PCR and Repetitive Extragenic Palindromic PCR Analyses of Escherichia coli Isolates To Identify Nonpoint Fecal Sources Appl. Envir. Microbiol., August 1, 2003; 69(8): 4942 - 4950. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Wiggins, P. W. Cash, W. S. Creamer, S. E. Dart, P. P. Garcia, T. M. Gerecke, J. Han, B. L. Henry, K. B. Hoover, E. L. Johnson, et al. Use of Antibiotic Resistance Analysis for Representativeness Testing of Multiwatershed Libraries Appl. Envir. Microbiol., June 1, 2003; 69(6): 3399 - 3405. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
