PROTOCOL FOR ALTERNATE TEST PROCEDURES FOR COLIFORM BACTERIA
        IN COMPLIANCE WITH  DRINKING WATER  REGULATIONS

PRESENCE/ABSENCE  LIQUID CULTURE METHODS FOR FINISHED WATERS
                        Version 1.2

                       December,  1995
           U.  S.  ENVIRONMENTAL PROTECTION AGENCY
            OFFICE OF RESEARCH AND DEVELOPMENT
          NATIONAL EXPOSURE RESEARCH LABORATORY
                  CINCINNATI, OHIO 45268

-------
         PROTOCOL FOR ALTERNATE TEST PROCEDURES FOR COLIFORM BACTERIA
                 IN COMPLIANCE WITH DRINKING WATER REGULATIONS
          PRESENCE/ABSENCE LIQUID CULTURE METHODS  FOR FINISHED WATERS

I.   INTRODUCTION

1.1  Regulatory Background

1.1.1  The Administrator, U.S. Environmental Protection Agency  (EPA), approves
analytical methods for all contaminants regulated  under the  Safe Drinking Water
Act  (SDWA).   When EPA  publishes  a regulation under  the  SDWA,  the  regulation
generally provides for  at least one method for detection and/or quantification
of that contaminant.  After any regulation 1s published, the Administrator may
approve additional methods or modifications to approved methods,  based on a staff
recommendation, after satisfactory comparability tests by the proposer, following
the Alternate Test Procedure  (ATP) protocol.

1.1.2  The June 29, 1989 and January 8,  1991  regulations on  the  microbiological
characterization of finished drinking water samples require the determination of
the  presence or  absence of  total  conforms  rather than  their quantitative
enumeration.1''-   These regulations also require that  when-total conforms are
present, the presence or absence of either fecal coliforms or £. coli must also
be determined rather than enumerated.  Consequently, the protocol for ATPs was
modified to evaluate this different type of method.  The methods  selected by EPA
for evaluation of proposed methods,  are designated as  Reference  Methods  in this
protocol.

1.1.3  If the data evaluation  demonstrates that the applicant's  method performs
at least  as well  as the currently  approved method,  the  Ecological Exposure
Research  Division of  the  National  Exposure  Research  Laboratory  (NERL)  in
Cincinnati   (formerly   the  Environmental  Monitoring  Systems  Laboratory  at
Cincinnati) will recommend approval to the Office  of Ground Water and Drinking
Water, which begins the regulation development process.  Regulation development
includes a Federal Register notice proposing to approve an ATP, public  comment
on the proposed method,  and  (depending on public comment)  a final rule published
in the Federal  Register that  approves  the method.  The regulation development
process may take about  one year or more.

1.2  Comparability Determination
  \            ^^—»«a^.

1.2.1  This  protocol  describes the method description and Comparability Study
data which EPA requires to evaluate an ATP for presence/absence liquid  culture
(LC) methods in microbiology.  The ATP program is intended to be flexible  in that
EPA's NERL-Cincinnati may tailor the criteria to a particular method. For this
reason, before beginning the comparability testing, the applicant is required to
contact  the ATP  Coordinator,  Ecological  Exposure  Research  Division, NERL-
Cincinnati,  U.S.  Environmental  Protection Agency, 26  West  Martin Luther King
Drive, Cincinnati, Ohio 45268, to assure agreement on the test  design.

1.2.2  Generally the  reference method  selected  by EPA for use  in the Compara-
bility Study will  be  the same type of test as the proposed  method. Consequently,
if an  LC  method were proposed  for the target organism,  the  corresponding LC
method for the target organism would be used in the Comparability Study.

-------
1.2.3  If a proposed method consists  of more than one configuration of tubes o"
bottles, the ATP Program of NERL-Cincinnati will require the testing of only one
configuration, providing that the proposer submits sufficient data showing tne
equivalence of other configurations.

2.   APPLICATION FOR ATP

2.1  General Requirements

2.1.1  The general  requirements for an  application for nationwide approval of a
new  or  revised method  for total coliforms,  fecal  coliforms, and/or  E.  coli
currently  include  the  name  and  address  of  the  applicant  and/or  authorized
representative;  the  microbiological  analyte  for which  the new  procedure  is
proposed;  justification  for  the proposed  new  method;   the  title,  company
identification number, the date of submission, and a complete write-up that is
a stand-alone  description  of the proposed method in the  required format (see
Section 2.3 below).

2.1.2   If the  applicant believes the proposed method  is very  similar  to a
promulgated method,  and/or represents  a minor optional change  of a reference
method, the applicant should also prepare  a two-column side-by-side description
of the sections of the refsrence and  proposed  methods and highlight differences
between  the  methods.   If the  method  is  a proposed  commercial   version  of a
previously  approved  method,  differences  in reagents,   interferences,  test
conditions, etc. should  be presented with available  performance data from the
proposed and reference methods.

2.1.3  NERL-Cincinnati will judge the  proposed method to be:  1) an acceptable
version of or an  optional minor modification of a previously promulgated method,
which does not require approval  as an ATP or 2) a significantly different method
which requires an application for an ATP approval.

2.2  'Every 'application  for  approval of  a  method  must"b'e'made  in  triplicate
(original + 2 copies) and forwarded to the Director,  Ecological  Exposure Research
Division, NERL-Cincinnati, USEPA, Cincinnati,  OH  45268.   Upon  receipt of the
Application, the ATP Coordinator will assign it an identification number, which
should be used  in all future communications.  NERL-Cincinnati staff will initiate
its technical reviews.  The  initial   review will concentrate on the clarity and
completeness of the description of the proposed method,  the  applicability of the
proposed   microbiological   principles   and   reactions   and  the  performance
characteristics  described  for  the method.    The  ATP  staff will  evaluate the
submitted  information  and  advise the applicant whether or not a comparability
study is required.

2.3  Method Description

2.3.1  Each method description must  include the following topics,  listed  in the
EMMC method format , in the order given.   The  purpose of the description  is to:
1) permit a fair comparison of  the proposed and reference methods  and  2) provide
a clean, clear description  of the method that can be easily used by laboratories.
The method should read  like  a  scientific  paper.

-------
 2.3.1.1   Scope and Application

      Include target organisms, type of test, e.g., membrane filter,  chromogemc
 test,  flubrogenic test,  etc.  and  the  sample types to which it  is  applicable.

 2.3.1.2   Summary of Method

      Include a brief outline of the method that describes its essential  features
 without  extraneous details.

 2.3.1.3   Definitions

      Include special  terms or  unique usage of terms.   Do not include  common
 microbiological  terms.

 2.3.1.4   Interferences

      Include  Information  and  data  generated   by  applicant  during   method
 development   using typical  samples  containing  a specific  quantity  of  an
 interference such toxic  materials, particulates,  non-target  organisms, etc.

 2.3.1.5   Safety

     Refer  to good laboratory  practices and use of  a  hood,  goggles,  and/or
 protective  clothing,   if appropriate.    Emphasize any  special  procedure  or
 precaution.

 2.3.1.6   Instrumentation, Equipment and Supplies

     Describe the necessary instrumentation, equipment and supplies and reference
 applicable manuals.

 2.3.1.7   Reagents,  Standards and Media

  v   Describe reagent, standard and media formulations and preparation.   Indicate
 shelf life of packaged materials and  special storage requirements.

 2.3.1.8  Sample Collection, Dechlorination,  Preservation, Shipment,  and Storage

     Detail  sample collection  and  handling  requirements.   Consider the  sample
 collector,   sample containers,   dechlorination,   sample  holding  times  and
 temperature conditions as specified In Standard Methods4.

 2.3.1.9 .Quality Control (QC)

     Indicate  the  specific  QC procedures  and  the  frequency  of performance
 required for the  proposed method. They should include sterility checks, positive
and negative controls, verification/confirmation of the target organism, media
performance  checks, duplicate analyses, etc.   Document that  a general  QA/QC
program is in operation  and that routine QC checks  are recorded  and action taken
 if a problem is  indicated.5'6

-------
2.3.1.10  Calibration and Standardization

     If applicable,  include the calibration steps that are performed on pH meter,
analytical balance, thermometer, autoclave, etc.

2.3.1.11  Procedure
     Detail the sample preparation and analytical steps in the proposed method
write-up.   Exceptions are those, routine microbiological procedures,  such  as
membrane filtration of samples,  that are  known to professionals and that may be
incorporated by reference.

2.3.1.12  Data Analyses and Calculations

     Describe the procedures for data analyses, calculations, interpretation.and
reporting of results.

2.3.1.13 Method Performance Characteristics (sensitivity,  specificity, recovery,
and precision)

     Provide available  information  on the performance characteristics  of the
proposed method and the procedures by which they were determined.  Specificity
data  should  demonstrate  the  ability of the proposed method to  recover and
distinguish the target organism from other organisms in the sample. The proposer
should have data on method sensitivity, the variability of  replicate analyses and
data on recoveries of known numbers  of target and  non-target  organisms by the
proposed method.  Summaries of these evaluative data should be included.

 ,    (However, in addition to the data provided above  in the method description,
a separate formal  Specificity Study will  be required as part  of the ATP process.
The method  for determining the  specificity  of  a  method is  presented  in the
Appendix.)

2.3.1.14  Pollution Prevention

     Pollution  prevention is  any technique  that  reduces  or eliminates the
quantity  or  toxicity  of waste  at  the  point  of  generation.     It   is the
environmental management tool  preferred over  waste disposal  or recycling.  Uhen
feasible, laboratory  staff should use a  pollution prevention  technique such as
preparation of the smallest practical volumes of reagents, standards and media
or downsizing of the  test units in a method.

     For  information about  pollution prevention  that  may  be  applicable  to
laboratories and  research institutions,  consult "Less is Better:   Laboratory
Chemical Management for Waste Reduction," available from  the  American Chemical
Society's Department  of Government Regulations  and  Science Policy,  1155 16th
Street N.W., Washington,  DC  20036,  (202) 872-4477.

     If the proposed method prevents or reduces exposure  to  toxicity, pollution
of the  laboratory or the general environment including  reduced  generation of
wastes, cite here.  Also  indicate non-applicability.

-------
2.3.1.15  Waste Management

     The  Environmental   Protection  Agency  requires  that   laboratory   waste
management  practices be  conducted consistent  with all   applicable  rules  and
regulations.   Excess reagents and samples and method process wastes should  be
characterized  and  disposed  of  in an  acceptable  manner.    The  Agency  urges
laboratories to protect the  air,  water,  and land by minimizing  and controlling
all  releases  from  hoods  and bench operations, complying  with  the letter  and
spirit of any waste discharge permit and regulations,  and by  complying  with  all
solid  and  hazardous  waste  regulations,  particularly   the hazardous   waste
identification rules and land disposal  restrictions.  For further information on
waste management consult "The Waste Management Manual for Laboratory Personnel,"
available from the  American  Chemical Society.

     Describe the proper disposal methods for waste reagents, materials,  supplies
and  samples.

2.3.1.16  References

     .Cite those source documents and publications which-are.necessary  sources  of
information to properly perform the method.

2.3.1.17  Tables, Diagrams,  Flowcharts,  and Validation Data

     Provide as needed.

2.3.1.18  Proprietary Information

     Mark  proprietary  Information  in  the  proposed  method  description   as
"Confidential".  EPA staff will treat  proprietary information according to the
regulations outlined in Subparts A and B in Part 2 of Title  40  of the  CFR.

2.4  Study Approval

2.4.1  Method comparability  data  and quality control data will  be required for
each application of a  new or significantly modified method.   The applicant  is
urged  not  to  initiate the  Comparability  Study  until   EPA   has  completed   an
evaluation  of   the  method  description   and   the  preliminary  performance
characteristic information and has approved the Comparability Study design.

3.   COMPARABILITY  STUDY DESIGN

3.1  Summary of Design

3.1.1  Typically, the  comparability data are  generated  from  a  single  finished
water which has been  spiked  with different (e.g.,  10)  wastewaters or  polluted
surface  waters  containing  the  target organisms  (total  coliforms  and  fecal
coliforms or E. coli).  The  spiked samples  are  analyzed  then diluted to a 10s/
100 ml level of target  organisms.   In  a preliminary study,  spiked samples are
exposed  to  a  chlorinating  agent to stress and reduce the  numbers  of target
organisms.   After an appropriate exposure period, sodium thiosulfate is  added  to
deactivate the chlorine in the samples and the  target organisms are enumerated
using a membrane filter procedure to determine the dilution  that will  yield a
density of  1-10 organisms per 100 mL  of sample.   Based  on these data, serial

-------
 dilutions of each  of the  chlorinated  samples are prepared  using one of  two
 options.   In the Comparability Study,  presence/absence analyses are performed on
 each of the serial dilutions by both the reference  and proposed methods so as to
 bracket  the  1-10 target  organisms/100 ml  and  assure a  reasonable split  in
 positive  and negative  results.   The identities,  preparation, dilutions,  and
 analytical  observations on the samples  are recorded and submitted to the  NERL-
 Cincinnati  staff for evaluation.   See the schema on the following page.

 3.1.2 A  minimum of  twenty  (20) replicate analyses are performed by each method
 on the dilutions selected for each of the 10 spiked samples:  Replicate analyses
 on the  three or more dilutions of each  spiked sample must  be  performed  on  the
 same day  for both the reference and proposed methods.

 3.1.3  See the  following example of a Comparability Study Design.

                           Example of  ATP Study Design

                                Natural  Sources            No.  Samples/
 	of Organisms	Source	
 Reference Method
     and                                10                        1
 Proposed  Method


                            Replicate Tests        Minimum Comparability
                           of 100 ml Volumes         Results Required
 	For Each Sample	per Method	

,Reference Method                 20                       200

 Proposed  Method                  20                       200


 3.2  Laboratory Participation

 3.2.1  Since the purpose of the study is to compare the proposed method to the
 reference method with minimal variability due to individual analyst error,  the
 number  of laboratories participating in the study  should  be  minimized.   It is
 strongly  recommended that  only one lab  perform  the analyses.   This laboratory
 must be certified under the drinking water laboratory certification program.  An
 applicant having a  vested  Interest  in  the method,  instrumentation,  apparatus,
 reagents, media, or associated  kits, may not perform  the  comparability study
 analyses  in the applicant's laboratory.

 3.3  Quality Control Data

 3.3.1  QC checks must be performed on each day of analyses with  known positive
 and  negative  cultures,   sterility   controls,  and verification   of colonies.
 Maintain   records  of  QC   checks  performed.    See  the  EPA Manual  for  the
 Certification of Laboratories Analyzing Drinking Water5 and the Intralaboratory
 Quality Control Guidelines in Standard  Methodl^.

-------
            SCHFMA OF COMPARABILITY STUDY

                     Preparation
      Collect  Spike  (Effluent  or Polluted  Water)
        Determine Density of Target Organisms
          Spike Drinking Water to 10s Level
        Conduct  Preliminary  Chiorination  Study
         Determine Target Organism Densities
                 Comparability Study
 Chlorinate Spiked Sample to Stress Target Organisms
 Dechlorinate  Sample Over  Increasing Exposure Times
  Select Dilutions to Bracket 1-10 Organisms/100 ml
 Determine Target Organism Densities (Option A Only)
        Conduct Comparability Study Analyses
If No Acceptable Split in P/A Responses, Repeat Study

-------
 3.4  Sample Collection and Handling

 3.4.1  Each sample should be collected and  held  in a single sterile, wide-mouth
 bottle of heat-resistant plastic or borosilicate glass with a leak-proof screw
 cap or ground-glass  stopper.   The container must be  resistant  to the solvent
 action of water and survive sterilization without any deformity or  production of
 toxic materials.   Screw  caps  must not  produce any  bacteriostatic,  toxic or
 nutritive products during sterilization.  Each new lot of bottles and closures
 should be checked for these effects before use in the study
j..'
     2  Samples should be maintained at 1-4  C during transit and holding time.

 4.   PREPARATION FOR COMPARABILITY STUDY
  y
 4.1  Preparation of Spiked Samples

 4.1.1  Since  it is unlikely that  finished drinking water samples can be collected
 which  contain  total  coliforms  and  fecal  coliforms  or  E.  coli  for  the
 comparability study,  a  single  drinking water sample can be  spiked  with these
 target organisms from different sources.  Recommended spiking sources are non-
 chlorinated secondary sewage effluents or polluted surface waters.  Both contain
 large numbers of total and fecal conforms  and non-coliforms; however, effluent
 has the advantage of providing a wide range of £.  coli strains, whereas polluted
 surface waters may be more variable in quality.  These sources  must come from ten
 or more geographically dispersed sites.

 To prepare spiked samples:

,4.1.1.1  Collect  at least  five  liters  of each  surface  water  or  effluent to be
 used to spike the drinking water.

 4.1.1.2   Obtain  a sufficient  volume  of  an  oxidant-free and  reductant-free
 (chlorine-free  and sodium  thiosulfate-free)  drinking  water sample to  make
 dilutions of the  spike water to obtain 10s target organisms/100 ml.   A single
 drinking water  must be used for  all dilutions  and  testing  of a spiked  test
 sample.  Refrigerate all waters at 1-4 C until  tests are' completed.

 4.1.1.3  Perform MF analyses of the spike source waters/effluents to determine
 the original  densities of the target organisms.

 4.1.1.4    Using   the  density  information,  add  sufficient  oxidant-free  and
 reductant-free  drinking water  to  provide 10s  each  total  coliforms,  fecal
 coliforms or E.  coli/lQQ ml.

 4.2  Chlorination of Spiked Samples

 4.2.1   The  microorganisms  in  the samples must be  stressed by exposure to
 chlorination  at  ambient  temperatures under  conditions similar  to  those in
 drinking water treatment facilities.  No two samples  are expected to produce the
 same levels of injured target organisms after chlorination.  After exposure, and
 prior to comparability testing,  analyses must be  conducted to confirm that each
 sample  contains  the  desired low  numbers  of  injured  total   coliforms,  fecal
 coliforms, or E. coli.

-------
 4.2.2   -The  disinfection process  Is  dependent  upon a  number- of physical  and
 biological  factors  including the type of sample to be disinfected (e.g.,  sewage
 effluent  or source  water),  the initial  titer of the target  organism(s)  present
 in  the  sample, the  chlorine demand of the sample, the type and concentration of
 chlorinating agent, the exposure  time,  s'ample  mixing,  pH, and  temperature8'9.
 These physical and biological factors should  be carefully monitored and recorded
 for each  sample  or  experiment.

 4.3 Preliminary Chlorination  Study

 4.3.1   Conduct a preliminary chlorination  study to establish  the exposure time
 needed  to reduce the  10  colony-forming units (CFUJ/100 ml level  in each spiked
 sample  to a  1-10 CPU/100 mi level,  using the  following procedure.  Determine the
 total residual and free residual chlorine concentrations  initially, at mid-point,
 and at  the  end  of the exposure time just prior  to  dechlorination.   Make these
 measurements using the approved N,N-diethyl-p-phenylened1am1ne  (DPD) colorimetric
 method10.   Record the measurements.

 4.3.2   Place 2 liters of the spiked  water  sample in a 3-liter glass container.
 If  a sample has  an  appreciable chlorine demand  (e.g.,  a  primary effluent or a
 sewage  sample),  add dilute  sodium  hypochlorite  solution until a  total residual
 chlorine  level  between 2.0 and 2.5  mg/L 1s  maintained  in the absence  of free
 chlorine.

 4.3.3   If a  sample has a low chlorine  demand,  avoid over-stressing or killing the
 organisms by prolonged exposure to free residual chlorine.   The free residual
 chlorine  concentration should  not  exceed 0.5-1.0 mg/L.

•4.3.4   The  exposure  time is directly  dependent upon the initial titer  of the
 target  organism  present, the sample  demand,  and the form of  chlorine present.
 For example, expose a sewage sample  to 2.0-2.5 mg  total  chlorine/L  for  over a
 range of  times such as 10, 20  and 30 minutes to reduce the E.  coli density from
 10   CPU to  a 1-10 CPU/100 ml sample.   However,  the required period  of exposure
 of  a sample with a  low chlorine demand of  0.5-1.0  mg free residual  chlorine/L,
 may be  significantly  shorter than  20-30 minutes.

 4.3.5   Add  an appropriate  volume of  a  diluted solution  of reagent grade sodium
 hypochlorite e.g.,  a 1:20 dilution of  5% (w/v)  stock solution,  to  achieve the
 desired level of chlorinating  agent.   Stir  the  sample  continuously during the
 Chlorination.

 4.3.6   Stop  the  action  of the chlorine at  the  end of the exposure  period by
 adding  0.8  mL of a  10% (w/v) sodium  thiosulfate solution/L  sample.

 4.4 Determination  of Bacterial  Density

 4.4.1   After deactivating  the chlorine  in  the  exposed samples,  analyze the
 samples to  determine the exposure times which achieve  residual  levels  of 1-10
 CPU/100 mL,  for  total  coliforms and  fecal coliforms/f.' coTi using the MP method
 and medium  designated for each target organism cited in Table 1.

-------
Target Organism
Total conforms
Fecal coliforms
E. coli
TABLE 1
Method
SM 9222B11
SM 9222D12
SM 9222B"
EPA 110513*
Medium
M-Endo LES Agar
M-FC
M-Endo LES Agar/
Nutrient Agar MUG
Count the colonies of target organisms after 24 h to determine the densities and
record, the exposure.time producing the desired counts	

4.5  Chiorination of Samples for Comparability Study

4.5.1   Proceed  with the chlorination of samples  for  the formal Comparability
Study using  the exposure time  and  necessary dilution to  produce  the desired
survival  count  of  1-10  CFU/100 mL for  total  coliforms and fecal  coliforms/
E. coli.

4.5.2  The proposer may  modify the  guidance  in this protocol for the spiking,
chlorination, dechlorination steps as seen fit with the provisos that:

4.5.2.1  The chlorination reduces the 10s CFU/100 ml in  the  spiked water samples
by <32£ZT6i)s to assure the presence of stressed organisms.  The final coli form,
fecal coli form  or f. coli  target  range for the Comparability Study  is 1-10
CFU/100smL and 100 mL sample volumes are tested.

4.5.3   Place  5  liters  of each  test  water in  a  glass  container.   Add  an
appropriate volume  of a  diluted solution of  reagent grade sodium hypochlorite
(e.g., a 1:20 dilution of 5% (w/v)  stock  solution)  to achieve the desired level
of chlorinating agent.

4.5.4  Determine  the total  residual  and  free residual  chlorine concentrations
initially, at midpoint and at end of the exposure time.  Make  these measurements
using the  approved N,N-diethyl-p-phenylenediamine  (DPD)- colorimetric method.
Record the measurements and submit with the study results.

4.5.5  Expose each sample to the disinfectant for  the selected time, based upon
preliminary data, to achieve an inactivation of 3-4 orders of magnitude  (99.9-
99.99%) of the initial 10s  population.  Stir  the sample continuously during the
chlorination.

4.5.6   Stop  the action of the  chlorine  at the end of  the exposure period by
adding 0.8 mL of  10% (w/v) sodium thiosulfate solution/L sample and mixing.
*Equivalent to Reference Method, NA + MUG  .
                                      10

-------
4.6  Dilution of Chlorinated Sample for Comparability Study

4.6..1.  Since .the  study is intended to compare the performance of the proposed
method  to the performance of the reference method,  the data used for comparison
should  be from the dilution(s)  which produces results closest  to an equal number
of positive and negative  MF results for the reference method.  The results for
the proposed method must  come  from that same dilution.

4.6.2   The sample dilutions are prepared according to one of two options:

     Option A utilizes.preliminary analyses to establish the density of target
organisms and hence requires a 24 h holding period for samples.

     Option B does not use preliminary analyses and, therefore,  does not require
a 24 h  holding  period.   However,  an  expanded  series of dilutions is needed to
bracket the density of the target organisms.

4.6.3   Option A

4.6.3.1   After chlorination,  and deactivation of  the  chlorine and  prior to
comparability testing, conduct MF analyses to  confirm that each sample contains
the desired low numbers of 1-10 total coliforms, fecal coliforms, or E. coli/\OQ
ml using the media designated  in Table 1.

4.6.3.2  After initiating  the assay of the  target organism density of each of the
dechlorinated samples in 4.6, Immediately refrigerate the samples at 4-10 C until
density.data are obtained (about 24 h).  This  holding period .also simulates the
period  of drinking  water transit in a distribution  system.   Unpublished data
indicate that this holding period does not significantly affect the recovery of
target organisms,  cause die-off, or allow repair from chlorine injury.  Read the
plates  after 24 h  to determine  the approximate density of the target organisms.

4.6.3.3  Use the density  assay results to estimate the dilution of each sample
with oxidant-free and reductant-free drinking water that is necessary to reach
the target  organism density of  1-10  cells per unit test  volume.   Make  that
dilution and at  least two others that bracket, for exarole, halve and double that
dilution.   One -of these dilutions   should  contain the  desired  1-10  target
organisms per test volume.  Immediately conduct the comparability analyses with
each sample using these three dilutions.  •

4.6.4  Option B

4.6.4.1  Use  oxidant-free and  reductant free  drinking water to  prepare equal
volumes of five serial  five-fold  (1:5)  dilutions  of each  of the dechlorinated
samples immediately after initiation  of the MF density  assays.  Comparability
analyses will  be performed on each of the dilutions in each  set without delay by
both the approved and proposed methods.

4.6.5  If one  of the dilutions  does not produce an acceptable split in positive
and negative results for  the reference method, the proposer must return to the
original spike sample and begin again to prepare samples1for the Comparability
Study.   See Section 4 and the Schema.  Although the  intention of the ATP design

                                      11

-------
is to achieve  a  50%/50% split in positive and negative  results,  the split in
responses may be  as much as 25%/75% in either direction and still be acceptable.

5.  Comparability Study

5.1  Test Methods

5.1.1  The description  of the reference  methods  and guidelines for use of the
proposed method follow.   For valid comparisons, the  samples, sample volumes and
dilutions must be the same for both methods.  Each test unit  must be 100 ml, but
may be configured as  a  ten x  10  ml,  five x 20  ml or one x 100 ml volume.  The
spiked samples must  be  stirred continuously while  portions  are withdrawn for
testing.  Include the  testing of pure cultures of organisms of known positive and
negative responses for  total  coliform, fecal coliform and £.  coli cultures to
insure a proper interpretation of the results.  It is  particularly important to
include a non-f.  coli thermotolerant fecal coliform culture such as Klebsiella
pneumonias (negative control)  along with a NUG-positive f. co 7/culture to insure
a proper interpretation of the fluorescent responses.

5.2  Reference Methods

5.2.1 The approved reference methods used in the Comparability Study are usually
the same type of test as the proposed method.  The reference methods  to be used
are given in Table 2.
                                    Table 2

     Target Organism          Reference Method              Med i urn

     Total coliforms  (TC)     SM 9221B15                LTB & BGLB

     TC Completed Test        SM 9221B15                M Endo LES or
                                                        MacConkey agar/LTB
Fecal coliforms
f. COli
SM 9221B15/9221E.l16
SM 9221B15/EPA 110413*
LTB/EC
LTB/EC MUG
5.2.2  Reference Method for Total Coliforms

5.2.2.1  Inoculate LTB with the volumes of sample or dilution which will bracket
a total  coliform  level of 1-10 organisms/100 ml.   Use  the LTB tube or bottle
configurations approved by NERL-Cincinnati.

5.2.2.2  Incubate the  LTB tubes for 22-24 h at 35 C and examine for a positive
test of growth, growth and gas or growth and acid.  Record  results and inoculate
positive tubes into BGLB  for confirmation.

5.2.2.3  Incubate negative tubes for another 24 h  and examine for growth  and gas.
Record results and inoculate positive tubes into BGLB for confirmation.
'Equivalent to Reference Method,  EC MUG  .

                                       12

-------
5.2.2.4  Incubate BGLB tubes for 24 h at 35 C.  Examine for a positive test of
growth and gas  at 24 h.   Incubate  negative tubes for an  additional  24  h and
examine for growth and gas.  Record results.

5.2.2.5  Perform the  completed test on one of the 20 replicate  test units of the
dilution selected  for  one of 10 samples  included  in  the'Comparability Study.
Streak each BGLB positive tube on M-Endo  agar for isolation and  incubate  for 24 h
at 35 C .  Transfer a portion of  a total coliform (sheen) colony from each M-Endo
plate into LTB.   Incubate the LTB tubes for 24  h at 35 C.  Examine  for growth and
gas or growth and acid.  Record  results.  If no growth and  gas, incubate for an
additional 24 h  and examine as  above.  Record results.

5.2.3  Reference Methods for Fecal Coliforms  and E. coli

5.2.3.1  Inoculate tubes of EC and/or EC  MUG medium with growth  from positive LTB
tubes from 5.2.2.2-5.2.2.3 using the dilutions that bracket 1-10 fecal coliforms
or £.  C07//100  mL.   These dilutions  may differ from those used to obtain 1-10
total coliforms/100 mL  because fecal coliforms and E. coTi are only a  part of the
total  coliform  group.

5.2.3.2  Incubate  the  EC  medium at  44.5 C for 24 h and examine for growth and
gas, which constitutes a positive P/A test  for fecal coliform bacteria.

5.2.3.3  Incubate the EC MUG medium at 44.5 C for 24 h and examine  for blue-white
fluorescence under long wave light,  which constitutes a  positive P/A test for
E. co7f.

5.3  Proposed Presence/Absence  Method

5.3.1  Analyze  the  same  volumes of spiked finished waters' 'and/or dilutions as
tested for the  reference method, following  the instructions for the proposer's
method.

5.3.2  Incubate  the cultures as  directed in the method and  examine for  positive
and negative results.  Record all results.

5.3.3  Complete verification steps according  to method instructions.

5.4  Recording  Results

5.4.1  Record the individual  positive and negative results (raw data) for the
reference and proposed methods on a bench  form similar to Attachment  1.  Include
information on  the analyst, laboratory,  date of analyses,  supervisory approval,
etc.

6.   DATA REPORTING

6.1  All  the  data from  the Comparability Study, i.e., the  replicate observations
of  the samples  by both the  EPA-approved reference method and the  applicant's
proposed method and the quality control observations should be forwarded  to the
Director, Ecological Exposure Research Division, NERL-Cincinnati (address on page
1).  The final  test unit results should be recorded in the formats suggested  in

            •  '    '                    13

-------
the Attachments 2, 3, and 4 and forwarded to NERL-Cincinnati, see 5.4.1.  Include
information on the analyst,  laboratory, date of analyses, supervisory approval.
etc.  The individual test results (raw data) should be available for review by
EPA, if requested.  The evaluation of the application can be accomplished more
quickly by the NERL-Cincinnati program and  statistical staff if the information
is also forwarded on disks compatible with  an  IBM-PC computer.  The text on the
disc should be presented in the latest version of WordPerfect (currently  5.1) and
the data presented in WordPerfect or  in ASCII.

7.   DATA REVIEW

7.1   Upon  receipt  of an  applicant's data  sets,  NERL-Cincinnati  staff will
initiate Us  technical and  statistical reviews.   Appropriate criteria will be
used to determine the acceptability of the reference method data as a  basis in
the evaluation of the analyses by the proposed method.

7.2  For presence/absence methods, the Chi-square test will be used to compare
the number of  positive test units in the proposed method analyses with the number
of positive test units in  the reference method analyses.  This statistical test
will also  afford a comparison  of the results  of the reference  and  proposed
methods  for  each  of  the  samples.    The  Breslow-Day test  will  be   used  to
investigate the interaction between method and sample.

8.  METHOD RECOMMENDATION AND APPROVAL

8.1  After completion  of the technical and statistical reviews, NERL-Cincinnati
will  prepare  its  recommendation for approval/disapproval   of  the  new  or
significantly  revised method,  notify  the applicant  of its recommendation, and
forward the recommendation  to the Office  of Ground Water and  Drinking Water
(OGWDW), which has the responsibility for  proposing  the  method in the Federal
Register.   Following  a three-month  public  comment period, OGWDW  will review
submitted comments and prepare the final nationwide approval/disapproval decision.
and promulgation notice in the Federal Register.

8.2  Upon  approval  of the  method, the applicant will be  responsible  for the
publication and distribution of the approved method to anyone  requesting a copy.

9.   REFERENCES

1.   U.S. Environmental Protection Agency.   Drinking Water;  National  Primary
     Drinking Water Regulations;  Total Coliforms  (Including Fecal Coliforms and
     f.  co/j); Final   Rule.    Federal  Register  54(124):27562-27568, June 29,
     1989.

2.   U.S.  Environmental  Protection  Agency.  National  Primary Drinking  Water
     Regulations; Analytical  Techniques;  Coliform Bacteria,  Federal   Register
     56(5):636-643, January 8, 1991.

3.   Villa, 0. and L.  Reed,  Co-Chairs, EMMC  Methods  Integration Panel.  Final
     Version  of  Approved  EMMC Format (Memorandum to  Members of EMMC  Steering
     Committee, Methods Integration  Panel,  and  Work  Group, Tri-Chairs).  U.S.
     Environmental Protection Agency, February 14, 1992.  pp.  1-2.

                                      14

-------
4.   Standard  Methods  for the Examination of  Water  and Wastewater. 18th  Ed..
     APHA, Washington, DC, 1992, Section 9060A and B, p. 9-18 thru  9-20.

5.   U.S.  Environmental  Protection Agency.   Manual  for the Certification of
     Laboratories Analyzing  Drinking  Water:   Criteria and Procedures, Quality
     Assurance, 3rd ed.  EPA/570-9-90-008A.   Chapter  5,  Microbiology (revised)
     Office of Drinking Water, Washington, D.C., October 1991, pp.  37-48.

6.   Standard Methods	Section 9020B.

7.   Standard Methods.... Section 9020B, pp. 9-5 and 9-6.

8.   White, G.  C.  Chemistry  of Chlorine.   In:  Handbook  of Chlorination for
     Potable  Water.  Wastewater.  Coolino Water.  Industrial  Processes,  and
     Swimming Pools.  Van Nostrand Reinhold Co., New York, 1972.  pp.  182-227.

9.   Jolley,  R.  L.  and  J.  H.  Carpenter.    A  Review of the  Chemistry and
     Environmental Fate of Reactive Oxidant Species of Chlorinated  Water.  In:
     Water Chlorination:  Environmental Impact and Health Effects.  Volume 4.  Book
     1, Chapter 1.  Ann Arbor Science, Ann Arbor, 1983.  pp. 3-47.

10.  Standard Methods.... Method 4500-C1 G.  pp.4-45 and 4-46.
     NOTE:  To obtain  total  chlorine  in  one reading,  add the  full amount of
     potassium iodide at the  start  with the specified amounts of buffer reagent
     and DPD indicator.

11.  Standard Methods.... Section 9222B, pp.  9-54 thru 9-58.

12.  Standard Methods.... Section 9222D, pp.  9-60 thru 9-61.

13.  U.S.. Environmental Protection Agency.  Tests Methods.,for.Escherichia  co1i
     in Drinking Water:   EC  Medium with MUG Tube  Procedure  and  Nutrient  Agar
     with  Mug  Membrane  Filter  Procedure.   EPA/600/4-91/016.   Environmental
     Monitoring Systems Laboratory, Cincinnati, Ohio.  July 1991.

14.  U.S.  Environmental  Protection Agency.   National   Primary  Drinking  Water
     Regulations;  Analytical  Techniques; Coliform Bacteria,  Federal Register. 40
     CFR Part 141, 56 (5) 636-643,  January 8, 1991.

15.  Standard Methods.... Section 9221B, pp.  9-46 thru 9-49.

16.  Standard Methods.... Section 9221E.1, pp.  9-52 and 9-53.

17.  Fleiss, J. F.,  Statistical Methods for Rates and  Proportions. 2nd ed.,  John
     Wiley & Sons, New York,  NY, 1973.2
                                      15

-------
ATTACHMENT I.  EXAMPLE BENCH FORM tor PRESENCE/ABSENCE COLIFORM MFT ANALYSIS
LOD IV
TestO
Anoly
amw •
rganlsn
r»-
n>








s

ample * and Descrlptio
Sunervliai'a 1



Indicate: + • presence
Dlluti
ens :
}









it
11
11
11
u
u
u
17
10
If
M
hnri

}









10
n
n-
n
u
is
u
17
11
It
M
!••*

8
5


5







,







.



•
X
















»






X
!


i

.












•

1
•


X















*







X
V


s



















1
X







:

:


1

'


•


i


x





























1









10
11
11
11
M
11
U
17
10
It
M
—

1
1
1
1
4
1
t
7
1
f
to
11
11
11
14
11
U
17
11
If
n
st-
1
'


.M.I.,




Dote of Analyst
,.


, ,

• B absence

|








•



1
!
I



i
'•
X






1



,


1



1





*• jr
j

\


1
•

i






•


•
•



X









•


.

1





.


X
*


it







•
•



1
1
!

i


:
¥ S






1
i

1
.

,

1
•
.
i
i


i

x












































I









10
11
11
11
u
11
u
17
11
If
10
turn*

\









11
11
11
11
U
11
11
17
11
If
M
Sl«
I


5














1

,



X















-







x
a
\













'
i


i


+ s























•*• ^
U w
«• 5




















•»• >•









i
:
i

.
.



.
i

f S
                                                         USCPA, NIRL-Onclnnotl, 6/95
                                          16

-------
                                  ATTACHMENT 2.   EXAMPLE FORMAT
             RESULTS OF ATP COMPARABILITY STUDY OF METHOD FOR TOTAL COUFORM BACTERIA
Title of Proposed Method: 	   Lab Name A Location: 	
Analyst:  	;	   Supervisor's Signature:   	
                             Indicate: + = presence
Sam
Refer*
i
i
i
4
S
1
I
I
f
11
11
11
11
14
IS
U
17
II
1*
M
££
pie * and Descri|
Date of Am
nice Method (B<

,l~l.. / /
t»LB)
•1 _

J




















^X

II




















* /




















x
IslBsllClBMHl









II
II
II
II
14
IS
11
17
11
1*
tt
iZ£

Proposed Method
1 II II




















4- S




















4- S




















f S~

Sample*
Reference MI
Seta
•qpfc^i









u
u
it
u
M
IS
11
17
11
11
M
Sum*

Date of i
Bthod(B
ifedDIJvf
J! _ 1




















*/_




















f ^*
Knahrsls? ' '
GLB)
lions





















t- /

Proposed Method










n
u
u
u
14
IS
u
• 17
II
It
n
Sumal


L 	 ..II.... II















.
"



X




















x




















x

                                                                                USEM. NERL-Clndnnall. 6/95

-------
                                     ATTACHMENT 3.   EXAMPLE FORMAT
                  RESULTS OF ATP COMPARABILITY STUDY OF METHOD FOR FECAL COUFORM BACTERIA
    TIHe off Proposed Method: 	   Lob Name A Location: 	
    Analyst:  	;	   Supervisor'* Signature:   	.	
                                Indicate:  + = presence   - = absence
Sample * an
Da
H*fMwn"» ••-••.










II
11
11
11
14
11
11
17
11
19
N
Stimcf
Sefei
dDescri)
to of Anc
od(ECM
efedDllui
•lion: 	
.tpi., / /
EDIUM)
Ifons
1 II II




















+ ^x















1




*x




















v^
tepfeofi









11
11
11
11
14
IS
U
17
11
If
II
££..
rPOpOSQQ enQiVN
C^lek^dtetfef Alf
Ml
•f «

<




















*/
\ II 1




















*/






















Sample * and Description
Dale off Analysis
Reference Method (EC MEDIUM)
•«»•«*
1
1
I
4
1
•
7
I
9
11
11
II
II
14
II
U
17
II
19
M
» 	 •» —
••••JlffK
















•




•1- .X





















V^





















f >X
!
* / ^
•••••VaHHI
1








11
11
II
II
14
II
U
17
II
19
N
Svmof
Imriti:



Proposed Method
Sefocfed Dilutions
\. .. II. . i




















^X




















4- >^
1 —















-
- -•

t- ./

00
                                                                                  USEPA. NERL-anclnnaN. 6/95

-------
                                      ATTACHMENT 4.   EXAMPLE FORMAT
                   RESULTS OF ATP COMPARABILITY STUDY OF METHOD FOR ESCHERICHIA COU
    Title of Proposed Method: 	   Lob Nome 4 Location: 	
    Analyst:  	     .	   Supenrftort Signature:   	
                                 Indicate:  -I- = presence   - = absence
Sample * am
Dal
Reference Me
»_ i —
wWf^N
fti|ifff»n









It
11
11
II
M
1)
14
If
11
If
M
s^i

IDescrif
AA| Ana
iflflMI*
l_l.. / /
Ihed (EC MUG)
nKfDffoflfofu
J! I'




















+ s





















'A





















f S

AV«COM









11
11
11
11
14
II
U
If
II
11
N
£Z£

Proposed Method

1 I















:




*/_
	 1






















1






















Sample * and DM
Date of J










11
n
11
11
14
11
U
17
11
It
M
Svm«f
SefKMdDlhil
i




















+ S
•riptlon:
Inatysls!
CMUG)
Mom
.. JL J




















4- S




















f ./

I I
•VfcGM
1
1
1
4
1
t
7
1
1
11
11
11
II
14
11
.U
17
II
It
M
Ml HI fli
Propo
•>— t —
•WwW
	 1




















•f iS
sed Method
•too* Dilution*
._.. II






••













+ .S



















f- ^s

u>
                                                                                    USEPA. NIIL-Clndnnall. 6/95

-------
                          APPENDIX

DETERMINING THE SPECIFICITY OF A PROPOSED PRESENCE/ABSENCE
              LIQUID CULTURE METHOD FOR TOTAL
        COLIFORNS. FECAL COLIFORMS, AND/OR E. coli

-------
                                   APPENDIX

   DETERMINING THE SPECIFICITY OF A PROPOSED PRESENCE/ABSENCE LIQUID CULTURE
         METHOD FOR TOTAL COLIFORMS, FECAL COLIFORMS, AND/OR E. coli

1.   INTRODUCTION

1.1    This   appendix  provides  guidance  for  determining  the  specificity  of
presence-absence (P/A) methods for total coliforms, fecal coliforms, or f. coli.
Specificity is determined by evaluating isolates from aquatic environments, which
are representative of target and nontarget organisms that may impact on  drinking
water, and using them in the determination of the false positive error and the
undetected target error of the proposed test.  A false positive error occurs when
a non-target organism produces  the reaction expected from the target organism.
A false -negative  error occurs  when  the target organism fails  to  produce the
expected positive reaction 1n the test procedure.  The false positive and false
negative errors may be  calculated according to the American Society of Testing
Materials (ASTM) Standard Practice 0 3870-79.'

1.2  The tests for specificity  may be conducted separately  from the formal ATP
Comparability Study or for economy may utilize positive  and  negative test units
from  the  Comparability  Study  Itself.   In  either  case,  the results  of the
specificity tests will  be evaluated independently of the  Comparability Study and
the specificity results will not be used to adjust or modify the Comparability
Study results.

2.   SPECIFICITY STUDY DESIGN

2.1  One water  sample  is  collected from each of ten different, geographically
dispersed wastewater  or ambient water  sources  that are known  to  contain the
target organisms (total  coliforms, fecal coliforms, or f. coli) for use as spikes
into a finished drinking water.  Preliminary quantitative analyses are conducted
to establish the density of the target organism in each source water.

2.2  Based on  the  results  of the density analyses,  the  source water volume or
dilution is determined  for spiking into the finished drinking water to yield 1-10
target organism CFU/100 ml.   This concentration of organisms should generate an
approximately  even split between  positive and  negative responses wherein a
positive response consists of one or more positive results-in  a  test unit and a
negative response consists of all negative results in a test unit.

2.3  The  proposer must perform  sufficient replicate  analyses  on the  selected
volume or dilution for each sample to generate  100  positive and 100  negative
responses.  Ideally, 10 positive and 10 negative responses would be obtained from
each of the 10 spiked drinking  water samples.

2.4   If the proposed  method can  analyze  for  more  than one target organism,
different sample or dilution volumes may be needed to achieve the desired even
split in positive and negative  responses for the different  organisms.

2.5   Each  specificity  analysis  must  be performed on a  drinking water sample
volume of  100 ml, consisting  of ten-10 mL,  five-20 mL, one-100  mL  or other
configuration equalling 100 mL  approved by the ATP Program  for the  study.

                                      21

-------
2.6  The testing  of  pure  cultures  of organisms  of known positive and negative
responses for total coliform, fecal coliform and E. coli Is included as quality
control  checks.   It  is  particularly  important  to  include  a non-f.  coli
thermotolerant  fecal coliform  culture  such  as Klebsiella -pneumoniae (negative
control)  along  with  a  MUG-positive  E.  coli  culture to   insure  a  proper
interpretation  of the fluorescent  response in the EC MUG test.

3.   DETERMINATION OF FALSE POSITIVE ERROR (See Figure 1)

3.1  Total  Conforms2

3.1.1  For each of the  ten  spiked  water  samples,  examine  the sample volume or
dilution selected in 2.2 and 2.3,  for units positive for total coliforms.  For
each sample,  transfer a loopful of  growth from the selected dilution of each of
the ten replicate positive 100 ml  test units  into tubes of LTB,  for a total of
100 transfers.  Incubate the  100 LTB tubes for 24 h at 35 C.  Examine for visible
growth with gas or acid production (a positive test).

3.1.2  If no visible growth after 24 h,  incubate  for an additional 24 h and again
examine for visible  growth.   If none, record as a  true negative response and
interpret as a  false positive test for total coliform bacteria by the proposed
method.

3.1.3  If growth is visible in  LTB  after  24-48 h, record results  and transfer a
loopful of culture into a tube of  Brilliant Green Lactose Bile  Broth (BGLB}2.
Incubate BGLB  tubes  at 35 C  for 24-48  h.   Examine  tubes for  growth  and gas
production.

3.1.4   If  no  visible growth and gas  in  BGLB after 24-48 h,  record  as  a true
negative response and interpret as  a  false positive response for  total coliform
bacteria by the proposed method.

3.1.5  Growth  and  gas in BGLB after  24-48  h is a true  positive response for total
coliform bacteria.

3.1.6  Record results.

3.2  Fecal  Coll forms3

3.2.1  For each of the  ten  spiked  water  samples,  examine  the sample volume or
dilution selected  in 2.2  and 2.3 for units  positive for fecal coliforms.  For
each sample,  transfer a loopful of  growth from the selected dilution of each of
10 replicate  positive test units into tubes of LTB, for a total  of 100 transfers.
Incubate tubes at 35  C for 24 h and  examine for growth and gas  or growth and acid
(a positive test).

3.2.2  If no visible growth after 24 h,  incubate  for an additional 24 h and again
examine for visible  growth.   If none, record as  a  true' negative response and
interpret  as  a  false  positive  response for fecal  coliform bacteria  by the
proposed method.

3.2.3  If growth is visible after 24-48 h, record results and transfer a loopful
of culture into EC. medium and incubate EC tubes  at 44.5  C for  24  h and examine.

                                      22

-------
3.2.4   Growth  and gas in  EC  medium after 24 h  is  a positive  test  for feca"
coliform bacteria.

3.2.5   No  growth and gas  in  EC medium after 24  h  is interpreted as  a false
positive response for fecal coliform bacteria by the proposed method.

3.2.6  Record results.

3.3  f. co7/*

3.3.1  For each  of the ten spiked  water samples,  examine the sample volume or
dilution selected  in 2.2  and  2.3  for units positive  for E. coli.   For each
sample, transfer a loopful  of  growth  from the  selected  dilution of each of 10
replicate positive test units into tubes of LIB, for a total of 100 transfers.
Incubate LIB tubes at 35  C  for 24 h and examine  for growth and gas or growth and
acid (a positive test).

3.3.2  If no visible growth in LIB after 24 h,  incubate for an additional 24 h
and examine for growth.   If none, record as a negative  response and  interpret as
a false positive E. coli by the proposed method.

3.3.3  If growth is visible after 24-48 h,  transfer one loopful of each culture
to EC MUG medium4 in a tube and incubate at 44.5 C for 24 h.

3.3.4   Examine  under long  wave (366nm) UV light.   If no fluorescence with or
without growth, response is interpreted as a false positive  test for E. co7i by
the proposed method.

3.3.5  If fluorescence occurs with growth, response is  a true positive response.
If  fluorescence occurs  without  growth,  it  is  judged autofluorescence  and a
negative response.

3.3.6  Record results.

3.4  Report all  results from the Specificity Study  of the 100 test units positive
for the target  organism,  in a  form similar to  the attached Example Forms 5-7.
Note that  the forms  have  only one entry space for  LIB  or BGLB  verification
results for  each transfer  for  the  proposed liquid culture method.   The  values
entered should be the sums  of the 24  and  48 h responses.

3.5  The. false positive  error will be  calculated  as follows:

     False Positive  Error  = A/B

     Where     A = False  positives which are  the number  of initial positive
                response  by the proposed method, which were  not  verified  by  the
                reference methods.

     and        B - Total number of initial positive  responses produced with  the
                proposed  method  (>  100).
                                       23

-------
4.   DETERMINATION OF  FALSE NEGATIVE ERROR  (See Figure 2)

4.1  Total Coliforms

4.1.1   For each  of  the ten spiked water samples,  examine the sample volume or
dilution selected in 2.2 or 2.3 for units negative for total col.iforms.  For each
sample, -transfer one ml of culture  from the selected dilution  of  each of 10
replicate negative test  units into tubes of LIB, for a total of  100  transfers.
Incubate at 35 C  for  24 h.  Examine  for visible growth with or without gas or
acid production  (a positive response).

4.1.2   If no  growth,  incubate for another 24 h at  35  C  and again examine for
growth.  If no growth,  record as  a true negative response.

4.1.3  If growth  is  visible in LTB after 24-48 h, transfer a  loopful  into a tube
of BGLB.  Incubate BGLB  tube at 35 C ± 0.5 C for 24-48 h.

4.1.4  Examine BGLB tubes.   Growth and  gas  after 24-48  h  is  a positive  response
and is  interpreted as a  false negative response for total coliform bacteria by
the proposed method.

4.1.5  No growth  and no gas in BGLB after 24-48  h for total  coliforms is a true
negative test.

4.1.6  Record results.

4.2  Fecal Coliforms

4.2.1   For  each  of  the ten  spiked  water  samples, if  the  test  is  for total
coliforms and  fecal  coliforms, examine the sample volume or dilution selected in
2.2 and 2.3  for  units  positive  for  total coliforms  and-negative  for fecal
coliforms.   For   each  sample,  transfer  one  mL of  culture  from  the  selected
dilution of each of 10 replicate  negative 100 mL test units into tubes of LTB,
for a total  of 100 transfers.   Incubate at  35 C  for 24 h  and examine for growth
and gas or growth with acid production (a positive response).

4.2.2   If the test  is  for  fecal  coliforms  only,  transfer a  loopful  of culture
into LTB  from  each  of  100 test  units  that are negative for fecal  coliforms.
Incubate at 35 C  for 24  h  and examine for growth.

4.2.3   If no growth,  incubate for another 24 h at  35  C  and again examine for
growth.  If no growth, record as  a true negative response.

4.2.4  If growth  is  visible in LTB after 24-48 h, record  results and transfer a
loopful into  a  tube  of EC medium.  This transfer may be  done  at the same time as
for the total  coliforms in  4.1.2.  Incubate at 44.5 C for 24  ± 2 h.  Examine for
growth and gas.

4.2.5  Growth and gas  in  EC after  24 H is a  positive  response and  is interpreted
as a false negative test for fecal coliform bacteria by  the proposed method.

4.2.6  No visible growth or gas in EC after 24 h for fecal coliforms is a true
negative response.

4.2.7  Record'results.                                    ""' '

                                      24

-------
4.3  f. co7?

4.3.1  For each of the ten spiked  water  samples,  if test is for total coliforms
and E. coli, examine the sample volume or dilution selected in 2.2 and 2.3 for
units positive for total coliforms and negative for E. coli.   For each sample.
transfer one ml of culture  from the  selected  dilution of each of 10 replicate
negative 100 ml  test units into  tubes  of  LTB,  for a  total  of  100 transfers.
Incubate at 35 C for 24 h and examine for growth and gas.

4.3.2  If the test is for E. co7;  only, transfer a loopful of culture into LTB
from each of 100  test units  that are negative for E.  coli.   Incubate  at 35 C for
24 h.

4.3.3   Examine  LTB  tubes  for visible  growth  with  or  without  gas  or acid
production.  If none,  reincubate  for  another  24  h and examine for growth,   if
none,  record  as  a true  negative  response  and  interpret as  a  false positive
response by the proposed method.

4.3.4  If growth  is visible after  24-48  h, transfer one loopful of each culture
to EC MUG medium in a tube and incubate at 44.5 C + 0.2 C for 24 h.

4.3.5  Examine EC MUG tubes for fluorescence under long wave (366nm) UV light.
Fluorescence with growth  is a  positive  response  and  is interpreted as a false
negative response for f. coli by the proposed method.

4.3.6   No  fluorescence  is a  true  negative result  response 'for E.  coli.
Fluorescence without  growth is  judged  autofluorescence and a negative response.

4.3.7  Record results.

4.4  Report all results from the Specificity  Study of the 100 test units negative
for the target organisms,  in a  form similar  to the attached Example Forms 8-10.
Note  that  the  forms  have  only one entry  space  for  LTB  or  BGLB verification
results for each transfer from the proposed liquid culture method.  The values
entered should be the sums of the 24 and 48 h responses.

4.5  The false positive error will be calculated as follows:

     False Negative Error = C/(B  - A + C)

     Where     A -  False  positives which  are the number  of  initial  positive
               responses by the proposed method which were not verified by the
               reference methods.

               B » Total number of initial positive responses  produced with the
               proposed method (> 100)

     and       C =  False  negatives which  are the number  of  initial negative
               responses  by the  proposed  method which  were positive  by  the
               reference methods.
                                      25

-------
REFERENCES

1.   American  Society  of  -Testing  Materials.    D 3870-79:    Practice   for
     Establishing  Performance  Characteristics for Colony  Counting Methods  In
     Bacteriology.  In: 1991 Annual Book of ASTM Standards. Vol. 11.02, Water
     Philadelphia, PA.

2.   Standard Methods  for the Examination of  Water  and  Wastewater. 18th ed
     Section 9221 B, APHA, Washington, DC  1992.  pp. 9-46-to 9-49.

3.   Standard Methods	 Section 9221 E, pp.  9-52 to 9-53.

4.   United States Environmental  Protection  Agency  (USEPA).   Test Methods  for
     f. coli in Drinking  Water.  EPA 600/4-91/016,  ORD,  NERL-Cincinnati, July.
     1991.
                                      26

-------
FIGURE 1.    SPECIFICITY TESTS FOB TOTAL Cm
           AND E.  COLI IJETIiODS FOR
                                                 FECAL CO
                                                 '
                       PROPOSED METHOD
                 FOR EACH POSITIVE RESPONSE:

           ONE LOOPFUL OF LIQUID CULTURE INTO LTB
                          INCUBATE
       LTB
       GROOTH-
       FALSE POSITIVE
                               LTB
                             GROWTH* w/wo GAS OR ACID
  GAS-
  FALSE
POSITIVE
  TOTAL
COLIFORM
                      GAS+
                 TRUE POSITIVE
                     TOTAL
                   COLIFORM
FLUORESCENCE
    FALSE
  POSITIVE
  £•
 FLUORESCENCE
TRUE POSITIVE
   £. £QLX
          GA
     FALSE POSITIVE
     FECAL COLIFORM
                                                AS-*-
                                          TRUE POSITIVE
                                          FECAL  COLIFORM

-------
FIGURE 2,
SPECIFICITY TESTS FOR TOTAL CoLIFORM, FECAL CoLIFORM
AND E. COLI METHODS FOR FALSE NEGATIVE ERROR
                         PROPOSED METHOD
                   FOR EACH NEGATIVE RESPONSE:

                ONE ML OF LIQUID CULTURE INTO LTB

                             INCUBATE
          LTB
          GROWTH-
          TRUE NEGATIVE
                  LTB
                GROWTH* w/wo GAS OR ACID
                                               EC HUG
     GAS-
     TRUE
   NEGATIVE
    TOTAL
   COLIFORM
         GAS*
        FALSE
       NEGATIVE
        TOTAL
       COLIFORM
      No
FLUORESCENCE
TRUE NEGATIVE
   £.
FLUORESCENCE
   FALSE
  NEGATIVE
  £•
         TRUE  NEGATIVE
        FECAL COLIFORM
                                  GAS+
                             FALSE NEGATIVE
                             FECAL COLIFORM

-------
                                ATTACHMENT 5.  EXAMPLE FORMAT
SPECIFICITY STUDY FOR FALSE POSITIVE RESPONSES BY THE PROPOSED METHOD FOR TOTAL COUFORM BACTERIA
Analyst:


tab Nam* « Location:
Supervisor's Signature:


Indicate: + « presence - = absence
Sam









II
tmm*
Sami









11
Z£
plo * and Description:
Data off Ana
vVs^BVIVflrCV aVoW
• 	 •-- / /
nOilal
II II










+ ^x**'










^x^


;







X^
•wlMM









11
.=£










^x<[

BGLB

1 II II










^X










^x










X!
Samplo * and Desertf
Data of Ana
LTB










• II
££
Samj
1
t
1
4
S
«
7
•
ff
11
££
iHon:
iydm: 1 1


II II










»• ^,x^
pto4>am
Dal











X










^^_
JDoscrif
to of Ana
LTB
EtadOffef










*^s^_










It
££,
BGLB


1 II II










^^_










*x<^
itiaiBMi
IIMMK
lysis: _ / /
VOfflal
II










tx










!X
••»•»•••









II
rx










^^

BGLB

< II II










*x










tX










*x^
                                                                         USIPA. NIRUOnalnnall. 6/»5

-------
                               ATTACHMENT 6.  EXAMPLE FORMAT
SPECIFICITY STUDY FOR FALSE POSITIVE RESPONSES BY THE PROPOSED METHOD FOR FECAL COUFORM BACTERIA
                                               Lob Nam* A Location:
A
tOljflYS

Swporvbor'* Signature


Indicate: 4- = presence - a absence
Sample * and DaMrtf










11
££
Sam,









if
££
^•laaej
^PWWJ
iflon*
i_i.. / /
LTB
*- -* fiii ti !••
ilffO ewleVlffOm
II II










+^^_
plefam
Dal
aTW^^R











+^_.










^x
IDoMrif










+ ^^
• !•»•»









11
££

ECMedlum
ap*XWWW aWiTNnWaaftp
1 II II










•f ^^X"










x^
iflon!
t~tm. 1 1
LTB
EtodDlnntoiH
II










x^










f ^S"
••>•»•*»









11
£1










f .X"

^ ECltodlum
1 II II










^X^


•







X!










X!
Sam)









if
££
pie 9 am
Dal
IDeMrii
iflon*

toef Amrfral*:^,/ '
LIB

II II










x.










tx<^
Sample * and Poaif
Date of Ana
LIB
•vfc*.









it
££










¥^_
•iM'«i









11
£=i
'§*,•• -a»_ ,J ^••--*«
eWtVCPerO t»f • VHOffM
1 II II










•*• ^x^










jxl
Wftt^kflA*
'inMU
VM« / '


II










;X!










tx^











x^











11
^










•f-.^^^X

EC Medium

II










^x!










•f ^^X











^<
                                                                                     M, «/9S

-------
                              ATTACHMENT 7.  EXAMPLE FORMAT

 SPECIFICITY STUDY FOR FALSE POSITIVE RESPONSES BY THE PROPOSED METHOD FOR ESCHERICHIA COU

                     	 Lab Mama A Locutions                        .
                          Indkate:  + = presence   - = abaence
       DaloalAiMlyBte.

           LTB
EC MUG
                                 II
                       and

                       Dateof Anafph
i
                                                                               fefa
                        EC MUG
                                                 II
M
                        M
      Soli
          LTB
    EC MUG
MbcNHflMfclfom
                           LTB
                        11
                                       II
                                       H
                         EC MUG
                                                II      II
                                                                         UMVW. NIIKlncbmatf. «/VS

-------
                               ATTACHMENT 8.  EXAMPLE FORMAT
SPECIFICITY STUDY FOR FALSE NEGATIVE RESPONSES BY THE PROPOSED METHOD FOR TOTAL COUFORM BACTERIA
Anolyvtf


un* mam* « bocanon:
SupandMrt Signature:


Indicate: + a presence - - absence
Sample * and Dosari f
Dale of Am
LTB
•*•«»
1
S
1
' 4
S
•
1
1
f
II
££
Sam,









n
••••f
nlon:
Jyflf. / /













^^_
II










XI










•f ^X^
•VSM*









II
£.1
BGLB
' SefocfMf DJlui
fin.,,
ion*
1 II II










x^










+ ^x^

Dal

LTB
•• il Hfli irfn •••
iw •viBvnfjfBB>
II II 1










[x1










jx1










^x
•>•!»









II
£^1










I- .X"

BGLB

1 II II


:
f






x^










+ ^s*










xi
Sam
•i»in«i
1
t
1
4
9
•
7
I
f
11
££
^PtWnj
pl« * and D«Mrif
Data of Ana
•powctocl Of lu
i










l^_











tx^l
IwlOllo
lyri»*_/ /
MOM











f ^X^
•**-









II
tn

BGLB

1










XI
II










tx
ple«>andDeKripflon:
B«fl»«J AM
limfet / /
LTB










it
££

II 11










^x










^X










x^










It
^L
BGLB

-------
                              ATTACHMENT?.  EXAMPLE FORMAT
SPECIFICITY STUDY FOR FALSE NEGATIVE RESPONSES BY THE PROPOSED METHOD FOR FECAL COUFORM BACTERIA
    THU of PrapOMd Method:
                          Indicate:  + = p*M«nce   • = ab*«nc«
                                                                    u«o*.

-------
                            ATTACHMENT 10.  EXAMPLE FORMAT
SPECIFICITY STUDY FOR FALSE NEGATIVE RESPONSES BY THE PROPOSED METHOD FOR ESCHERICHIA COU
            j U
n
A
IMMyMl

LabN
Supor




Indicate: + =» pretence - = absence
Sami









II
£z£
pl«* and Description:
. Data of Ana
LIB
lyflfft^ '













^x<;
II










^>^










^>^
*•*-









11
IZot

EC MUG

1 II II










+ ^^










*s^_










¥^^_
Samplo 9 ana DoMripnon*
Data of Ana
LIB
•*•_»









ll
££
lyj.. / /


u u i










|XI










XI


:







)X
•!• •»









n
tLl
EC MUG '

1 II II










+ ^^










^X










x^
Sam









ll
IZ^.
Sam
I
i
3
4
S
4
7
•
9
II
££
ptofandDoMrii
Data of Ana
LTB
iflOfK
-• — •-- / /


II II










+• .X'"










^>^1
pla«*andDoKri|
Data of Ana
•M^vWW^B aWMV










¥ ^^
•••••wapM









u
£lta
EC MUG


1 II II










x^^










^x^
,tloiMI
IHOfflz
Into: / '
MOM
II II










f ^sr


:
•
-





^x










x
B.|i»i.«.









11
£=£










^>^

EC MUG

< II II










x^










V^










TX
                                                                     USCM. NHL-dndnnatl, 4/VS

-------