rxEPA
United States
Environmental Protection
Agency
National Training
and Operational
Technology Center
Cincinnati OH 45268
EPA-430/1-78-014
August 1978
Water
Bacteriological
Methods in Water
Quality Control
Programs
Training Manual
-------
EPA-430/1-78-014
August 1978
Bacteriological Methods in Water
Quality Control Programs
This course is for laboratory personnel who can
perform basic bacteriological laboratory procedures
such as sample inoculations, transfers, weighings,
and related skills.
After successfully completing the course, the
student will have increased knowledge of all
aspects of sampling, analysis and data handling
for bacteriological samples as required by Federal
Register Guidelines for effluent monitoring and
other water quality programs.
The training incorporates classroom instruction
and activity sessions, student performance of
laboratory assignments and follow-up discussions.
U. S. ENVIRONMENTAL PROTECTION AGENCY
Office of Water Program Operations
National Training and Operational Technology Center
-------
FOREWORD
These manuals are prepared for reference use of students enrolled in scheduled
training courses of the Office of Water Program Operations, U. S. Environmental
Protection Agency.
Due to the limited availability of the manuals it is not appropriate
to cite them as technical references in bibliographies or other forms
of publication.
References to products and manufacturers are for illustration only;
such references do not imply product endorsement by the Office of
Water Program Operations, U. S. Environmental Protection Agency.
The reference outlines in this manual have been selected and developed with a
goal of providing the student with a fund of the best available current information
pertinent to the subject matter of the course. Individual instructors may provide
additional material to cover special aspects of their own presentations.
This manual will be useful to anyone who has need for information on the subjects
covered. However, it should be understood that the manual will have its greatest
value as an adjunct to classroom presentations. The inherent advantages of
classroom presentation is in the give-and-take discussions and exchange of infor-
mation between and among students and the instructional staff.
Constructive suggestions for improvement in the coverage, content, and format
of the manual are solicited and will be given full consideration.
-------
CONTENTS
Title or Description Outline Number
Federal Register Guidelines Establishing Text Procedures for the
Analysis of Pollutants 1
Bacteriological Indicators of Water Pollution 2
Examination of Water for Coliform and Fecal Streptococcus Groups 3
Media and Solutions for Multiple Dilution Tube Methods 4
Use of Tables of Most Probable Numbers 5
The Membrane Filter in Water Bacteriology 6
Membrane Filter Equipment and its preparation for Laboratory Use 7
Membrane Filter Equipment for Field Use 8
Principles of Culture Media for Use with Membrane Filters 9
Selection of Sample Filtration Volumes for Membrane Filter Methods 10
Detailed Membrane Filter Methods 11
Colony Counting on Membrane Filters 12
Verified Membrane Filter Tests 13
Collection and Handling of Samples for Bacteriological Examination 14
Testing the suitability of Distilled Water for the Bacteriology Laboratory 15
Residual Chlorine and Turbidity 16
-------
-------
FEDERAL REGISTER GUIDELINES ESTABLISHING TEST PROCEDURES FOR THE
ANALYSIS OF POLLUTANTS
I FEDERAL REGISTER GUIDELINES
A Authority
1 In 1972, section 304(g) of Public Law
92-500, required the EPA Administra-
tor to promulgate guidelines establishing
test procedures for the analysis of pol-
lutants that would include the factors
that must be provided in any state certi-
fication (section 401), or National
Pollutant Discharge Elimination System
(NPDES) permit application (section
402).
2 These test procedures are to be used
by applicants to demonstrate that efflu-
ent discharges meet applicable pollutant
discharge limitations, and by the states
and other enforcers in routine or random
monitoring of effluents to verify effective-
ness of pollution control measures.
B Establishment
Following a proposed listing there was a
period for reply by interested parties.
The first rulemaking was published in
the Federal Register on October 16, 1973.1
C Current Guidelines
Proposed amendments and update were
published in 1975. The current guide-
lines were issued in the December 1, 1976.
Federal Register.
D Format
The "Approved Test Procedures" are
given in a table which lists 115 parameters,
the methodology to be used to determine
them and either the page number in standard
references or else a source where the ana-
lytical procedure can be found.
1 Divisions
The parameters are listed alphabetically
including four subcategories of related
tests:
a bacteria
b metals
c radiological
d residue
2 Standard References
Those cited most often as sources of
analytical procedures for the listings
are the EPA Chemical Methods Manual,3
Standard Methods4 ASTM5 and U. S.
C
Geological Survey? Other sources of
procedures are given in footnotes to
the Table.
II EPA METHODS MANUALS
A Analytical Procedures
The EPA Bacteriological Methods Manual
(not FR referenced-manual in preparation)
was developed for their water quality
laboratories, using Standard Methods as
the basic reference.
CH. 13b. 1.78
1-1
-------
Federal Register Guidelines Establishing Test Procedures for the Analysis of Pollutants
The bacteriological manual contains
information on sample techniques;
labels; chain of custody; log sheets;
selection of sampling sites and
frequency; and preservation and
transit of samples.
The EPA Chemical Methods Manual
(Referenced in FR) was developed
for their water quality laboratories,
using Standard Methods and ASTM
as basic references. In many cases,
EPA modified methods from these
sources or else developed methods
suitable for their own laboratories.
B Order of Processing
Before approving such applications,
the Regional Administrator sends a
copy to the Director of the EPA
Environmental Monitoring and Support
Laboratory (EMSL) for review and
recommendation. If the Regional
Administrator rejects any application,
a copy is also sent to EMSL. Within
90 days the applicant is to be notified
(along with the appropriate state agency)
of approval or rejection notifications
for purposes of national coordination.
The chemistry manual also contains
a section on sampling and preservation.
This is in tabular form and contains
information on volumes required for
analysis, the type of container that
can be used, preservation measures
and holding times. The current
Federal Register references this
Table for recommendations on these
aspects of sample handling for NPDES/
Certification purposes.
B Precision and Accuracy Data
Precision and accuracy data from
interlaboratory quality control studies
are given for most of the methods cited
in both of the manuals.
Ill METHODS NOT IN 1976 GUIDELINES
A Application to Use
A system has been established for
permit holders to apply for approval
to use methods not listed in the December
1, 1976 Federal Register. One supplies
reasons for using an alternative method
to the EPA Regional Administrator
through the state agency which issues
certifications and/or permits. If the
state does not have such an agency,
the application is submitted directly
to the EPA Regional Administrator.
IV REQUIRED ANALYSES
Which measurements are to be done
and reported depend on the specifications
of the individual certifications or permits.
A Mandatory for Secondary Plants
As of July 1, 1977 all municipal secondary
wastewater treatment plants are required
to measure and report pH, BOD (bio-
chemical oxygen demand), suspended
solids and flow.
B Additional for Secondary Plants
Measurements which also may be required
of secondary treatment plants are fecal
coliform bacteria, residual chlorine,
settleable solids, COD (chemical oxygen
demand), total phosphorus, and the nitrogen
series (total Kjeldahl N, NH3-N, NO3-N,
NO2-N).
C Municipalities and Industries
Other required analyses depend on local
factors for a municipality. Each industry
has requirements pertinent to the processes
involved.
1 Non-specific
Non-specific measurements to assess
overail water quality might be required
like acidity, alkalinity, color, turbidity,
specific conductance.
1-2
-------
Federal Register Guidelines Establishing Test Procedures for the Analysis of Pollutants
2 Organics
Various organic analyses might be
relevant such as total organic carbon,
organic nitrogen, phenols, oil and
grease, surfactants, pesticides.
3 Metals
Specified metals may be of interest.
Currently, the Federal Register lists
35 trace metals in the test procedure
guidelines.
4 Others
Cyanide, bromide, chloride, fluoride
and hardness are other measurements
that might be required.
V METHODOLOGY AND SKILLS
A Methodology
The analytical methods specified in the
Federal Register for these measurements
range from procedures using equipment
commonly found in most laboratories, to
procedures requiring sophisticated instru-
ments such as an organic carbon analyzer
or an atomic absorption unit.
B Skills
The degree of analytical skills required
to perform the analyses likewise varies,
as does the cost of having such analyses
performed by service laboratories.
VI OTHER ANALYTICAL CONSIDERATIONS
A Sample
The importance of securing a represent-
ative sample of the type (grab or compo-
site) specified by the permit cannot be
over-stressed.
B Record Keeping
Keeping complete and permanent records
about the sample is also essential. Such
records include conditions when the sample
was collected, chain of custody
signatures and details and results
of analyses.
C Quality Control
Whether the analyses are done in-house
or by a service laboratory, an Analytical
Quality Control Program should be estab-
lished. For chemistry fifteen to twenty
percent of analytical time (cost) should
be given to checking standard curves for
colorimetry, analyzing duplicate samples
to check precision and analyzing spiked
samples to check accuracy. Recording
precision and accuracy data on quality
control charts is an effective method of
using such data as a daily check on analyt-
ical performance. This can also be done
with numbers reported on "blind" samples
sent to service labs.
VII SUMMARY
The December 1, 1976 Federal Register
promulgates guidelines establishing test
procedures for the analysis of pollutants
which might be required for certification
(PL 92-500, section 401) or for NPDES
permits (PL 92-500, section 402). The
issue lists page numbers in standard
references where procedures can be
found to measure the 115 parameters
listed. It also updates the regulations
for application to use methods not cited
in the guidelines. The measurements
which must be made are specified by either
a state agency or by U. S. EPA. Apparatus
and professional skills to do the measure-
ments will vary. Representative samples,
complete records and analytical quality
control measures are all necessary elements
for producing reliable data.
REFERENCES
1 Federal Register, Vol 38, No 199, Tuesday,
October 16, 1973, Title 40, Chapter 1,
Subchapter D, Part 136, page 28758.
1-3
-------
Federal Register Guidelines Establishing Test Procedures for the Analysis of Pollutants
Federal Register, Vol 41, No. 232,
Wednesday, December 1, 1976, Title 40, This outline was prepared by A. D. Kroner,
Chapter I, Subchapter D, Part 136, page 52780. Chemist, National Training and Operational
Technology Center, MOTD OWPO, USEPA,
Methods for Chemical Analyses of Water Cincinnati, Ohio 45268
and Wastes, 1974, EPA, EMSL, Cincinnati,
Ohio.
Standard Methods for the Examination of Descriptors; Chemical analysis, chemical
Water and Wastewater, 14th ed., 1976, guidelines, self-monitoring requirements,
APHA, Washington, D. C. non-approved analytical methods, NPDES
Annual Book of Standards, Part 31, Water,
1975, ASTM, Philadelphia, Pennsylvania.
Methods.for Collection and Analysis of
Water Samples for Dissolved Minerals
and Gases, U. S. G. S. Survey Techniques
of Water - Resources Inventory, Book 5,
Chapter Al, 1970, U.S. GPO, Washington,
D.C.
Microbiological Methods for Monitoring
the Environment (In Preparation).
1-4
-------
WEDNESDAY, DECEMBER 1, 1976
PART II:
ENVIRONMENTAL
PROTECTION
AGENCY
WATER PROGRAMS
Guidelines Establishing Test Procedures
for the Analysis of Pollutants
Amendments
-------
52780
RULES AND REGULATIONS
Title 40—Protection of Environment
CHAPTER I—ENVIRONMENTAL
PROTECTION AGENCY
SUBCHAPTER D—WATER PROGRAMS
[FIL 630-4]
FART 136—GUIDELINES ESTABLISHING
TEST PROCEDURES FOR THE ANALYSIS
OF POLLUTANTS
Amendment of Regulations
On June 9, 1975. proposed amendments
to the Guidelines Establishing Test Pro-
cedures for the Analysis of Pollutants
(40 CPR 136 > were published in the FED-
Ls.ii. REGISTER '40 FR 245351 as required
by section 304 of the Federal Water
Pollution Control Act Amendments of
1972 (&6 Stat. 816, et seq.. Pub. L. 92-500.
1972> hereinafter referred to as the Act.
Section 304(g) of the Act requires that
the Administrator shall promulgate
guidelines establishing test procedures
for the analysis of pollutants that shall
include factors which must be provided
in: (1) any certification pursuant to sec-
tion 401 of the Act. or (2> any permit ap-
plication pursuant to section 402 of the
Act. Such test procedures are to be used
by permit applicants to demonstrate that
effluent discharges meet applicable pol-
lutant discharge limitations and by the
States and other enforcement activities
in routine or random monitoring of ef-
fluents to verify compliance with pollu-
tion control measures.
Interested persons were requested to
submit written comments, suggestions, or
objections to the proposed amendments
by September 7. 1975. One hundred and
thirty-five letters were received from
commenters. The following categories of
organizations were represented by the
commenters: Federal agencies accounted
for twenty-four responses: Suite agen-
cies accounted for twenty-six responses;
local agencies accounted for seventeen
responses: regulated major dischargers
accounted for forty-seven responses;
trade and professional organizations ac-
counted for eight responses; analytical
instrument manufacturers itnd vendors
accounted for seven responses; and an-
alyticu; service laboratories accounted
for six responses.
All comments were carefully evaluated
by a technical review committee. Based
upon the review of comments, the follow-
ing principal changes to the proposed
amendments wore made:
A i Dtftn'-tions. Section 130.2 has been
amended tj update references: Twenty
commenters, representing the entire
spectrum of responding groups pointed
out that ihe references cited in §§ 136.2
'f'. 136.'.!'«>. and 136.2'hi were out-of-
date: 5§ 136.a'f'. 136.2'g'. and 136.2'hi,
re'pecuvely. have been ani'nidr'.l to show
the following editions of the standard
references: "14th Edition of Standard
Methods for the Examination of Water
and Waste \Vatrr." "1974 EPA Manual
of Method* "for the Analyse of Water and
Waste;' and ' Part 31, l!)Vfi Annual Bonk
of ASTM Standards."
Identification of Test Procedures.
Both the content and format of § 136.3,
"Table I. List of Approved Test Proce-
dures" have been revised in response to
twenty-one comments received from
State and local governments, major regu-
lated dischargers, professional and trade
associations, and analytical laboratories.
Table I has been revised by:
(1) The addition of a fourth column
of references which includes procedures
of the United States Geological Survey
which are equivalent to previously ap-
proved methods.
(2) The addition of a fifth column of
miscellaneous references to procedures
which are equivalent to previously ap-
proved methods.
(3) Listing generically related param-
eters alphabetically within four subcate-
gories: bacteria, metals, radiological and
residue, and by listing these subcategory
headings in alphabetic sequence rel-
ative to the remaining parameters.
(4) Deleting the parameter "Algicides"
and by entering the single relevant algi-
cide. "Pentachlorophenol" by its chemi-
cal name.
The Environmental Protection
Agency concurred with the American Dye
Manufacturers' request to approve its
procedure for measurement of color, and
copies of the procedure are now available
at the Environmental Monitoring and
Support Laboratory. Cincinnati (EMSL-
CI)
(3) In response to three requests from
Federal, State governments, and dis-
chargers, "hardness," may be measured
as the sum of calcium and magnesium
analyzed by atomic absorption and ex-
pressed as their carbonates.
(4) The proposal to limit measure-
ment of fecal coliform bacteria in the
presence of chlorine to only the "Most
Probable Number" (MPN» procedure has
been withdrawn in response to requests
from forty-five commenters including
State pollution control agencies, permit
holders, analysts, treatment plant op-
erators, and a manufacturer of analyt-
ical supplies. The membrane filter (MF>
procedure will continue to be an ap-
proved technique for the routine meas-
urement of fecal colifiorm in the pre-
sence of chlorine. However, the MPN
procedure must be used to resolve con-
troversial situations. The technique
selected by the analyst must be reported
with the data.
(5) A total of fifteen objections, re-
presenting the entire spectrum of com-
menters, addressed the drying tempera-
tures used for measurement of residues.
The use of different temperatures in dry-
ing of total residue, dissolved residue and
suspended residue was cited as not allow-
ing direct intercomparability between
these measurements. Because the intent
of designating the three separate residue
parameters is to measure separate waste
characteristics (low drying temperatures
to measure volatile substances, high dry-
ing temperatures to measure anhydrous
inorganic substances >. the difference in
drying temperatures for these residue
parameters must be preserved.
(E) Deletion of Measurement Tech-
niques. Some measurement techniques
that had been proposed have been de-
leted in response to objections raised
during the public comment period.
(1) The proposed infrared spec-
trophotometric analysis for oil and
grease has been withdrawn. Eleven com-
menters representing Federal or State
agencies and major dischargers claimed
that this parameter is defined by the
measurement procedure. Any alteration
in the procedure would change the def-
inition of the parameter. The Environ-
mental Protection Agency agreed.
(21 The proposed separate parameter
for sulfide at concentrations below 1
mg/1, has been withdrawn. Methylene
blue spectrophotometiy is now included
in Table I as an approved procedure for
sulfide analysis. The titrimetric iodine
procedure for sulfide analysis may only
be used for analysis of sulfide at concen-
trations in excess of one milligram per
liter.
(F) Sample Preservation and Holding
Times. Criteria for sample preservation
and sample holding times w ere requested
by several commenters. The reference for
sample preservation and holding time
criteria applicable to the Table I param-
eters is given in footnote (1 ~> of Table I.
(G> Alternate Test Procedures. Com-
ments pertaining to § 136.4, Application
for Alternate Test Procedures, included
objections to various obstacles within
FEDERAL REGISTER, VOL. 41. NO 232—WEDNESDAY, DECEMBER 1. 1976
-------
RULES AND REGULATIONS
52781
these procedures for expeditious ap-
proval of alternate test procedures. Pour
analytical instrument manufacturers
commented that by limiting of applica-
tion for review and/or approval of alter-
nate test procedures to NPDES permit
holders, § 136.4 became an impediment to
the commercial development of new or
improved measurement devices based on
new measurement principles. Applica-
tions for such review and/or approval
will now be accepted from any person.
The intent of the alternate test pro-
cedure is to allow the use of measure-
ment systems which are known to be
equivalent to the approved test proce-
dures in waste water discharges.
Applications for approval of alternate
test procedures applicable to specific dis-
charges will continue to be made only by
NPDES permit holders, and approval ol
such applications will be made on a
case-by-case basis by the Regional Ad-
ministrator in whose Region the dis-
charge is made.
Applications for approval of alternate
test procedures which are intended for
nationwide use can now be submitted by
any person directly to the Director of the
Environmental Monitoring and Support
Laboratory in Cincinnati. Such applica-
tions should include a complete methods
write-up, any literature references, com-
parab\lity data between the proposed al-
ternate test procedure and those already
approved by the Administrator. The ap-
plication should include precision and
accuracy data of the proposed alternate
test procedure and data confirming the
general applicability of the test proce-
dure to. the Industrial categories of waste
water for which it is intended. The Di-
rector of the Environmental Monitoring
and Support Laboratory, after review of
submitted Information, will recommend
approval or rejection of the application
to the Administrator, or he will return
the application to the applicant for more
Information. Approval or rejection of ap-
plications for test procedures intended
for nationwide use will be made by the
Administrator, after considering the rec-
ommendation made by the Director of
the Environmental Monitoring and Sup-
port Laboratory, Cincinnati. Since the
Agency considers these procedures for
approval of alternate test procedures for
nationwide use to be Interim procedures,
we will welcome suggestions for criteria
for approval of alternate test procedures
for nationwide use. Interested persons
should submit their written comments in
triplicate on or before June 1, 1977 to:
Dr. Robert B. Medz, Environmental Pro-
tection Technologist, Monitoring Quality
Assurance Standardization, Office of
Monitoring and Technical Support (RD-
680), Environmental Protection Agency,
Washington, D.C. 20460.
(H) Freedom of Information. A copy
of all public comments, an analysis by
parameter of those comments, and docu-
ments providing further information on
the rationale for the changes made In
the final regulation are available for
inspection and copying at the Environ-
mental Protection Agency Public Infor-
mation Reference Unit, Room 2922,
Waterside Mall, 401 M Street, SW.,
Washington, D.C. 20460, during normal
business hours. The EPA information
regulation 40 CFR 2 provides that a rea-
sonable fee may be charged for copying
such documents.
Effective date: These amendments be-
come effective on April 1, 1977.
Dated: November 19,1976.
JOHN QTJARLES,
Acting Administrator,
Environmental Protection Agency.
Chapter I, Subchapter D, of Title 40,
Code of Federal Regulations is amended
as follows:
1. In § 136.2, paragraphs (f), are amended to read as follows:
§ 136.2 Definition:,.
n
, milli-
grams per liter.
2. Alkalinity, kg CaCO', milli-
grams per liter.
3. Ammonia (as N), milligrams
PIT liter.
BACTERIA
. t oliform (fecal)5, number per
100ml,
. Conform (fecal) * in presence
of chlorine, number per 100
ml.
Coliform (total) * number per
100 ml.
Coliform (total) * in presence
of chlorine, number per 100
ml.
Fecal streptoeoet i,' number
per 100 mj.
Benzidine, milligrams per liter.
Biochemical oxygen demand,
5 d (J)OL>i), milligrams per
liter.
Bromide, milligrams per liter__
Chemical oxygen demand
(COD), milligrams per liter.
' hloride, milligrams per liter,.
Electrons trie end point
(pH of S.2) or phenol-
pbtbakin end point.
Etoetrametric titration
(only to pB 4.5) manual
or automated, or equiva-
lent automated methods.
Manual distillation < (at pH .
9.5) followed by neseleri-
i»Uon, titratlon, elec-
trode. Automated phe-
nolate-
116
40 »(60T)
41 «(607)
1M
IK .
168
410 .
412
..
616 ..
*JPN;«m«iil.riin.- filter -.
do ' *
... do.«
MPN;« nu-ml.ran* filu-r '.'.'.'
with enrichment. . .
Ml'N;« ninnl.tane filter; ...
plate count.
. Oxidation— colnniaetric '..
Winkler (Azide modifies- .
tior.) or electrode method. . ..
Titrimetric, iodine-iodate
I>ichroinat« reflux
Silver nitrate; mercnric ni- _
trate; or automated eolori-
metric-ferricyanide.
H37 '' (45)
. , - 922
- . 928, 937
92S '(36)..
S16
033
943
!H4 ' (SO)
1H7
M3 ' (SO)
14 323 .1R
20 650 472 124
303 'Ail
'fj 304 Z65 .
31 M3 » (46).
'• (17)
l (610)
' (6151
FEDERAL REGISTER, VOL. 41, NO. 232—WEDNESDAY, DECEMBER 1, 1976
-------
52782
AND REGULATIONS
Parameter arid units
14. Chlorinated organic com-
pounds (except pesticides),
milligrams per liter.
1.5. Chlorine—total residual, milli-
grams per liter.
1*1. Color, pHitmum cobalt Units
or dominant wave length,
hue, luminance, purity.
17. Cyanide, total,H milligram:-;
per liter.
18. Cyanide amenable to chlorin-
ation, milligrams per liter.
19. Dissolved oxygen, milligrams
per liter.
20. Fluoride, milligrams per Utn'.
51. Hardness--Total, as CaCO3,
milligrams per liter.
22. Hydrogen ion (pill, pH units.
23. KjAldahl nitrogen (ta N>,
milligrams per liter.
J4. Aluminum—Total, milligrams
per liter.
J5. Aluminum— Dissolved, milli-
grams per liter,
36. Antimony—Total, milligrams
per liter.
37. Antimony—Dissolved, milli-
grams per liter,
*8. Arsenic—Tola], milligrams
per liter.
29. Arsenic—Dissolved, mini
grams per liter.
30. Barium—Total, milligrams
per liter.
31. Barium—-Dissolved, milli-
grams per liter.
32. Beryllium—Total, milligrams
per liter.
33. Beryllium—Dissolved, milli-
grams per liter.
34. Boron—Total, milligrams per
liter.
35. Boron—Dissolved, milligrams
per liter.
30. Cadmium—Total, milligrams
per liter.
37. Cadmium—Dissolved, milli-
grams per liter.
38. Calcium—Total, milligrams
per liter.
30. Calcium—Dissolved, milli-
grams per liter.
40. Chromium VI, milligrams per
liter.
41. Chromium VI—Dissolved,
milligrams per liter.
42. Chromium—Total, milligrams
per liter.
43. Chromium—Dissolved, milli-
grams per liter.
Method
Gas chromatography !
1974 Utt ed.
KPA rtttMUrd
methods methods
References
(page nos.)
Pt. SI USGS
197S methods'
ASTM
Other
approved
methods
lodometric titration, ainper-
ometrie or starch-iodine
end-point; DVD colori-
metric or Titrimetric .
methods (these last 2 are
interim methods pending
.aboratory testing).
Colorimetrie: speetrophoto-
metric; or A DM I pro-
cedure.^
Distillation followed Uy
silver nitrate titration or
pyridinc pvra/olone (or
barbituric acid i colori-
mt'trie.
do
W inkier uVnle modifica-
tion) or electrode method.
1 >istillation * followed by
ion electrode: riPADNS;
oruutoniatedcomple-xone.
EOT A tit rat ion: -auto-
mated color! metric; or
atomic absorption (sum
of Ca and Mg as their
respective e&rbonatesj.
Klectronietric measurement _
Digestion aad distillation
followed by ne.ssleriration,
titration, or electrode;
automated digestion auto-
mated phenolate.
Digestion lfi followed by
atomic absorption 1( or by
eoloriraetric (Eriochrome
Cyanine -R).
0.45 micron filtration " fol-
lowed by referenced meth-
ods for total aluminum.
Digestion » followed by
atomic absorption.1*
0,4f> micron filtration " fol-
lowed by referenced
method for total antimony.
Digestion followed by silver
diethyldithlocarbamate;
or atomic absorption.'• "
0.45 micron filtration " fol-
lowed by referenced
method for total arsenic.
Digestion v* followed by
atomic absorption.19
0.45 micron filtration " fol-
lowed by referenced
method for total barium.
Digestion " followed by
atomic absorption »• or by
colorimetric (Aluminon).
0.45 micron filtration " fol-
lowed by referenced
method ff>r total beryllium.
Colorimetric (Curcumin)
0.45 micron filtration >: fol-
lowed by referenced metli-
od for total boron.
Digestion 15 followed by
atomic absorption » or by
colorimetric (Ditbizone).
0.15 micron nitration" fol-
lowed by referenced meth-
od for total cadmium.
Digestion » followed by
atomic absorption; or
EDTA Litration.
0.45 micron filtration >" fol-
lowed by referenced meth-
od for total calcium.
K^traction and atomic ab-
sorption; colorimetric (Di-
phenylearbazide).
0.45 micron filtration 1T fol-
lowed by referenced meth-
od lor chromium VI.
I >igestioii J* followed by
atomic absorption "or by
colorimetric (Uiphenyl-
carbazide >.
0-45 micron filtration 1T fol-
lowed by referenced meth-
od for total chromium.
36
30
101
322 278
332
320
51
56
65
59
61
68
70 __
23fl
175
165 __
182 ___
92
M ...
t
OS
97
09
13
443 308
450
38(1
3«1 307
393 305 -
614
202 161
WO ITS
J37 . ____
152
171
J8ft
»3
m
152
152
177
287
IL'li !(609)
!I3
94 3(617)
12ft '(606!
122 »(612)
II (!!!' .. .
« (3D
»(W)
52
53
148 S45 62 »(610)»(»7)
182
189
S45 66
89,105
75 .
104
148 345
in 28*
78
77
See foolnotfw at end o£ table.
FEDERAL REGISTER, VOL 41, NO. 732—WEDNESDAY, DECEMBER 1, 1976
-------
RULES AND REGULATIONS
52783
Parameter and units
Method
HT4 14th ed. (IM«* not.)
SPA itandwd '
methods methods Pt. *1
U8Q8
197S methods'
A8TM
Othtr
approved
methods
44. Cobalt—Total, milligrams per Digestion" followed by 1OT 148 345 80 »(97)
liter. atomic absorption."
45. Cobalt—Dissolved, milli- 0.45 micron nitration" fol-
grams per liter. lowed by referenced meth-
od lor total cobalt.
46. Copper—Total, milligrams Digestion » followed by 108 148 345 83 • (619) '• (87)
per liter. atomic absorption " or by 196 293 t
colorimetric (Neocu-
proine).
47. Copper—Dissolved, milli- 0.45 micron filtration" fol- _
grams per liter. lowed by referenced meth-
od for total copper.
48. Gold—Total, milligrams per Digestion" followed by _ _
liter. atomic absorption."
49. Iridium—Total, milligrams Digestion" followed by
per liter. atomic absorption."
50. Iron—Tola], milligrams per Digestion» followed by 119 148 345 102 •<«!»)
liter. atomic absorption u or by
colorimetric (Phenauthro- 208 126 - -
line).
51. Iron—Dissolved, milligrams 0.45 micron filtration IT fol- __
per liter. lowed by referenced meth-
od for total iron.
52. Lead—Tola!, milligrams per Digestion" followed by 112 148 345 105 »(619)
liter. atomic absorption lfl or by
colorimetric (Dithizone). 215
53. Lead—Dissolved, milligrams 0.45 micron filtration IT fol- _ „.
per liter. lowed by referenced meth-
od for total lead.
54. Magnesium—Total, milli- Digestion» followed by 114 148 345 109 »(819)
grams per liter. atomic absorption; or 221
gravimetric.
55. Magnesium—Dissolved milli- 0.45 micron filtration " dol-
grams per liter. lowed by referenced
method for total magne-
sium.
56. Manganese -Total milligrams Digestion" followed by 11« 148 145 111 >{•!•)
per liter. atomic absorption la or by 325,227
colorimetric (Persulfate or
periodate).
57. Manganese-Dissolved milli- 0.45 micron filtration'7 fol-
grams per liter. lowed by referenced
method for total manga-
nese.
58. Mercury—Total, milligrams Flumeless atomic absorp- 118 15« 388 "(51). .
per liter. tion.
59. Mercury—Dissolved, milli- 0.45 micron filtration » fol- „ __ _
grams per liter. lowed by referenced
method for total mercury.
80. Molybdenum—Total, mllll- Digestion" followed by 139 150
grains per liter. atomic absorption."
61. Molybdenum—Dissolved, 0.45 micron nitration" fol- .
milligrams per liter. lowed by referenced
method for total molybde-
num.
62. Nickel—Total. milligrams Digestion" followed by 141 itt MS 115
per liter. atomic absorption " or by , , „"'"
colorimetric (Heptoiime). "
63. Nickel—Dissolved, milli- 0.45 micron nitration " fol-_
grams per liter. lowed by referenced
method for total nickel.
84. Osmium—Total, milligrams Digestion" followed by
per liter. atomic absorption."
65. Palladium—Total, milligrams Digestion" followed by _,
per liter. atomic absorption."
66. Platinum—Total, milligrams Digestion" followed by
per liter. atomic absorption." "
67. Potassium—Total, milligrams Digestion" followed by 14S __ . 134 '(Ml
per liter. atomic absorption, colon- 3K .
metric (Cobaltinitrite), or 214 401 __ ""
by flame photometric. "
68. Potassium—Dissolved, milli- 0.45 micron filtration IT fol-
grams per liter. lowed by referenced meth-
of for total potassium.
SSI. Rhodium—Total, milligrams Digestion" followed by
per liter. atomic absorption."
70. Ruthenium—Total, milli- Digestion" followed by . ___
grams per liter. atomic absorption."
71. Selenium—Total, milligrams Digestion" followed by 145 159
per liter. atomic absorption." >«
72. Selenium—Dissolved, milli- 0.45 micron filtration IT fol- 1
grams per liter. lowed by referenced meth- -----
od for total selenium.
73. Silica—Dissolved, milligrams 0.45 micron filtration " fol- 274 487 398 189 ...
per liter. lowed by colorimetric
(Molybdos'ilicate).
74. Silver—Total," milligrams Digestion" followed by US 148 .. 14J HUM «{W)
per liter. atomic absorption " or by 248 '_ '
colorimetric (Dithiwme).
75. Silver—Dissolved," milli- 0.45 micron miration » fol-
grams per liter. lowed by referenced meth- " " "
od for total sliver.
7«. Sodium—Total, milligrams Digestion" followed by 147. 143 I (Ml)
per liter. atomic absorption or by 250 401
flame photometric.
77. Sodium—Dissolved, milli- 0.45 micron filtration " fol-
grams per liter. lowed by referenced meth-
od for total sodium.
See footnotes at end of lablf
FEDERAL REGISTER, VOL. 41, NO. 232—WEDNESDAY, DECEMIEt 1, 1*7*
-------
RULES AND REGULATIONS
1974
Parameter and unit* M^th^d EPA s
methods j
78. Thallium— Total, milligrams Digestion 15 followed by
per liter. atomic absorption.1*
70. Thallium— Dissolved, milli- 0.45 micron filtration17 fol- ..
grams per lit'-r. lowed by referenced meth-
od for tof.al thallium.
fcO. Tin— Total, uiill.'KMiiHS per Digestion I5 followed by
lile,r. atomic absorption.1*
81, Tin — Dissolved milligrams 0.45 micron nitration i: fol- ..
per liter. lowed by referenced meth-
od for total tin.
S2. Titanium— Toul, milligram?- Uigestiou 15 followed l>y
ptr lile.r. atomic absorption.16
83, Titanium- -OisM-Ued, milli- 0.4o micron filtration17 fol- .
gram* per lit-'r. lowed by referenced meth-
od for total lilaiiimn.
S4. V.madium- Totnl, miUigra:us 1 Motion. 10 followed by
PIT liier. atomic absorption )f' or by ,.
L-olnrinietrif (Gallic acid).
Ki Mms per Jii'T. ' lowed by ivfei.-ju'ed meth-
od for iiiiitl vanudinm,
W. Zinc Totnl, milligrams p- r Digestion -^ followed hy
liter. atomir absorption 18 or by .
folorinietric i Dithizonpi.
87, Zinc — Dissolve-! milligrams 0.45 mierun miration i: fol- .
per liter. lowed by referenced metli-
od for toial /i ne.
fcK. Nil rate r>- N . TM.'iliuram? p«-r '"admunn induction; bru-
llter. einP Sui!";ile: ;iutom;\tpd
eadmium i*r hyilni/iiii' re-
ft'.'. Nitrate ;,~ \ . ii.Jli^rums p--r Mumml or auimnat-'d eolori-
lit'T. metric i D:;i/.oti7;Uion).
*K). Oil and jrri':^.>. nrlligrum.- p»-r Liquiii-li'iui'l .'xtraetiun
lit'-r. with iridiloro-trifluoio-
**t. organ V curium; toiul 'To('.;. * ombus'ion- Infriircd
uiilligranis per liter. method.--
y-. Oi-pbniei'iiLroK'Mi nih N), milli- Kjeldahl niir«^"ii mn.ui
(ii'ams per liter. ammonia nitrogen.
•j3. ' nthophosphaic :,v- )'\ luilli- Manual or automated us» or-
grains per bier. hit: acid reduction.
grains per liter.
liter.
,f>. Phenols, milligrams per liter,. ('<..!. inni"lri<'. <4AAPi_. _ ..
grains per liter.
08. Phosphorus; tot;il :as P . Pcrsulfittf digesiion fol-
milliRrums per liin. lov\ed h> m:tnual_or auto-
CoiMiicr.
liter.
liter.
104. Total. milJmrams per liur ... ' -^\ imeinc. I031ol0r>( r...
105. Total dissolved itilterablci, ola,^ rib. r fiinatioii, 180° C-
milligrams pe.r liter.
106. Total .suspended (nonfilter- uiasf lil'. r [ii:i;iiioii, 103 to
able), milligrams per lit**1- 10,">L ( \
107. ^e).tleable, niillilii^rs per litej ^ olumeirk- «r gravinieirir _
or milJigrams per liter.
108, Total volatile, milligrams per *-ry\ in.eiri.-. .Wr '\ . _
liter.
101'. Sj^cilic coi»diK-(an.v, micro- W ii. ;if-(<.ne b>i'lij<- coudnc-
iiilios ptr et-nlimot'ir at 2o° tinii'irj.
per liier. "r auioman <1 roloiimetric
i barium chioranilate).
111. SiJiii'i' 'iis .w . T.^l,J^?nim^ per Tiirimetric— bidine for lev-
jji. - e!h pn-ater than 1 nig per
liter; Met)iyl>-he blue pliO-
t: ', iiiijligramt Tit rim*- trio, iodinr-iodate —
per liter,
113. Surfactant'-, Piiii/igt-ams per TuSorimi trie J6
505 .
503 -
508
600
125 ..
132
441
345
358
4C7
384
529
645 .
384
691" »
594
601" »
606
661
120
424 ..
445 ...
135
404
223
11 i«5) .
" (67)..
15'j 'i.61a;
Iwr
oved
hods
119 >(614) "(28)
121
» 14)
122 » (612, 614)
131 ' (621)
"(24)
133
(75+7S)
" (79)
(75+78)
« (79)
148
154 .
"(11)
" (31)
156
M621)
. ...
><606)
• (624)
1(023)
iuu* fur sampling (iinl preservaiion of samples according to parameter measured may be tound in
inieiii Analysis of Waie; nr.il Wastes, 1^74" U.S. Environmental Protection Agency, table 2, pp.
HDERAl REGISTER, VOL. 4), NO. 232—WEDNESDAY, DECEMBER 1, 1976
-------
RULES AND REGULATIONS
52785
' All page references for USGS methods, unless otlievivise noted, are to Brown, E., Skougstad, M. W., and Fish-
man, M. J., "Methods for Collection and Analysis of Water Samples for Dissolved Minerals and Gases," U.S. Geologi-
cal Survey Techniques of Water-Resources Inv., books, ch. Al, (1970). , , .
» EPA comparable method may be found on indicated page of "Official Methods of Analysis of the Association of
Official Analytical Chemists" methods manual, 12th ed. (1975).
' Manual distillation is not required if comparability data on representative effluent samples are on company me
to show that this preliminary distillation step is not necessary; however, manual distillation will be required to resolve
any controversies.
s'Tlie method used must be specified.
t The 5 tube MPN is used. ,... ..
' Slack, K. V. and others, "Mel hods for Collect iou and Analysis of Aquatic Biological and Mircobiological Samples:
U.S. G eological Survey Techniques of Water-Kesources Inv. book 5, ch. A4 (1973)."
• Since the membrane filter technique usually yields low and variable recovery from chlorinated wastowaters, the
M 1'N method will be required to resolve any controversies.
' Adequately tested methods for beiizidine are not available. Until approved methods are available, the following
interim method can be used for the estimation of beuzidine: (1) "Method for Beiizidine and Its Salts in Wastewaters,"
available from Environmental Monitoring and Support Laboratory, U.S. Environmental Protection Agency, Cin-
cinnati, Ohio 45268.
1° American National Standard on photographic Processing Effluents, Apr. 2, 1075. Available from ANSI, 1130
Broadway, New York, N.Y. 10018.
11 Fishman, M. J. and Brown, Eugene, "Selected Methods of the U.S. Geological Survey for Analysis of Waste-
waters," (1976) open-nle report 76-177.
12 Procedures lor pentachlorophcnol. chlorinated organic compounds ,and pesticides can be obtained from the En-
vironmental Monitoring and Support Lbaoratory, U.S. Environmental Protection Agency, Cincinnati, Ohio 45268.
"Color method (ADMI procedure) available from Environmental Monitoring and Support Lbaoratory, U.S.
Environmental Protection Agency, Cincinnati, Ohio 45268.
and cyanide amendable to chlorination measurements.
« For the determination of total metals the sample is not tillered before processing. Because vigorous digestion
procedures may result in a loss of certain metals through prcciptation, a less vigorous treatment is recommended as
given on p. 83 (4.1.4) of "Methods for Chemical Analysis of Water and Wastes" (1U74). In those instances where a
more vigorous digestion is desired the, procedure on p. 82 (4.1.3) should be followed. For the measurement of the noble
metal series (gold, iridium, osmium, palladium, platimum, rhodium and ruthenium), an aqua regia digestion is to be
substituted as follows: Transfer a rcpresentrtive aliquot of the well-mixed sample to a (irillin beaker and add 3 ml
of concentrated redistilled HNOj. Place the beaker on a steam bath and evaporate to dryness. Cool the beaker and
cautiously add a 5 ml portion of aqua regia. (Aqua regia is prepared immediately before use by carefully adding 3
volumes of concentrated HC1 to one volume of concentrated HNOs.) Cover the beaker with a watch glass and return
to the steam bath. Continue heating the covered beaker for 50 min. Remove cover and evaporate to dryness. Cool
and take up the residue in a small quantity of 1:1 HC1. Wash down the beaker walls and watch glass with distilled
water and filter the sample to remove silicates and other insoluble material that could clog the atomizer. Adjust the
vohn e to some predetermined value based on the expected metal concentration. Thf sample is now ready for analysis.
H As the various furnace devices (flamele&s A A) are essentially atomic absorption techniques, they are considered
to be approved test methods. Met hods of standard addition are to be followed as notrd in p. 7H of "Methods for Chem-
ical Analysis of Water and Wastes," 1974.
17 Dissolved metals are defined as those const it ill en Is which will pass though a 0.45 ^m membrane filter. A pre-
filtration is permissible to free the sample from larger suspended solids. Filler the sample as soon as practical
after collection using the first 50 to 100 ml to rinse tlie filter flask. (Glass or pliistie filtering apparatus arc recommended
to avoid possible contamination.) Discard the portion used to rinse the flask and collect the. required volume of
filtrate. Acidify the filtrate with !:1 redistilled HNO; toapH of '1. Normally, 3 ml of (1:1) acid p.-r liter should be
sufficient to preserve the samples.
18 See "Atomic Absorption Newsletter," vol. 13, 75 (1071). Available from I'erkiu-Elmer Corp., Main Avc., Xorwalk.
Conn. 06852.
18 Method available from Environmental Monitoring and Support Laboratory, U.S. Environmental Protection
Agency. Cincinnati, Ohio 45268.
20 Recommended methods for the analysis of silver in industrial wastewaters at concentrations of 1 mg/1 and
above are inadequate where silver exists as an inorganic balide. Silver halides such as the bromide and chloride
are relatively insoluble in reagents such as nitric acid but are readily soluble in an aqueous buffer of sodium thio-
sulfate and sodium hydroxide to a pH of 12. Therefore, for levels of silver above 1 mg/1 20 ml of Cample should be
diluted to 100 ml by adding 40 ml each of 2M NasSiOj and 2M NaOH. Standards should be prepared in the same
manner. For levels of silver below 1 mg/1 the recommended method is satisfactory.
21 An automated hydrazine reduction method is available from the Environmental Monitoring and Support
Laboratory, U.S. Environmental Protection Agency, Cincinnati, Ohio 45268.
22 A number of such systems manufactured by various companies are, considered to be comparable in llu-ir per-
formance, in addition, another technique, based on combnslion-methane detection is also acceptable.
23 Goerlitz. D., Brown, E.. "Methods for Analysis of Organic Substances in Water": U.S. Geological Survey Tech-
niques of Water-Resources Inv., book 5, ch. A3 (1972).
« R. F. Addison and R. G. Ackman, "Direct Determination of Elemental Phosphorus by fins-Liquid Chroma-
tography," "Journal of Chromatography," vol. 47, No. 3, pp. 421 420, 1070.
2s The method found on p. 75 measures only the dissolved portion while the uu-tlioil on p. 7* measures only sn--
pended. Therefore, the 2 results must be added together to obtain "total,"
2l> Stevens, H. H., Ficke. J. F., and Smoot, G. F., "Water Tempera! lire--Influential factors, Field Mi-a^inriiM'nt
and Data Presentation: U.S. (leolopicai Survey Techniques of Water Resources Inv., bunk 1 UU7,V."
4. In § 136.4, the second sentence of
paragraph (c) is amended by deleting
the word "subchapter" Immediately fol-
lowing the phrase "procedure under this"
and immediately preceding the word
"shall" and replaced with the phrase
"paragraph c;" and § 136.4 is amended
by adding a new paragraph (d> to read
as follows:
§ 136.4 Application for
procedures.
alternate
(c) ' * * Any application for an alter-
nate test procedure under this paragraph
(c) shall: * * *
(d) An application for approval of an
alternate test procedure for nationwide
use may be made by letter in triplicate
to the Director, Environmental Monitor-
ing and Support Laboratory, Cincinnati,
Ohio 45268. Any application for an alter-
nate test procedure under this paragraph
(d) shall:
(1) Provide the name and address of
the responsible person or firm making
the application.
(2) Identify the pollutant is amended by inserting the
phrase, "proposed by the responsible
person or firm making the discharge."
immediately after the phrase, "applica-
tion for an alternate test procedure."
and immediately before the comma.
The second sentence of paragraph
is amended by deleting the phrase,
"Methods Development and Quality As-
surance Research Laboratory," immedi-
ately after the phrase, "to the Regional
Administrator by the Director of the,"
and immediately preceding the period
ending the sentence and inserting in its
place the phrase, "Environmental Moni-
toring and Support Laboratory, Cin-
cinnati."
The third sentence of paragraph
is amended by deleting the phrase,
"Methods Development and Quality As-
surance Research Laboratory," immedi-
ately after the phrase, "forwarded to the
Director." and immediately before the
second comma and by inserting in its
place the phrase, "Environmental Moni-
toring and Support Laboratory, Cin-
cinnati."
9. Section 136.5 is amended by the
addition of a new paragraph (e) to read
as follows:
FEDERAt REGISTER, VOL. 41, NO. 232—WEDNESDAY, DECEMBER 1, 1976
-------
52786 RULES AND REGULATIONS
§ 136.5 Approval of alternate test pro-
cedures.
* * * * *
(e) Within ninety days of the receipt
by the Director of the Environmental
Monitoring and Support Laboratory,
Cincinnati of an application for an
alternate test procedure for nationwide
•use, the Director of the Environmental
Monitoring and Support Laboratory,
Cincinnati shall notify the applicant of
his recommendation to the Adminis-
trator to approve or reject the applica-
tion, or shall specify additional informa-
tion which is required to determine
whether to approve the proposed test
procedure. After such notification, an
alternate method determined by the Ad-
ministrator to satisfy the applicable re-
quirements of this part shall be approved
for nationwide use to satisfy the require-
ments of this subchapter; alternate test
procedures determined by the Adminis-
trator not to meet the applicable require-
ments of this part shall be rejected.
Notice of these determinations shall be
submitted for publication in the FEDERAL
REGISTER not later than 15 days after
such notification and determination is
made.
[PR Doc.76-35082 Piled ll-30-76;8:45 am]
•DUAL UGISTEI, VOL 41. Ha 931—WEDNESDAY, DECKMUI 1. 1974
-------
BACTERIOLOGICAL INDICATORS OF WATER POLLUTION
Part 1. General Concepts
I INTRODUCTION
A Bacterial Indication of Pollution
1 In the broadest sense, a bacterial
indicator of pollution is any organism
which, by its presence, would demon-
strate that pollution has occurred, and
often suggest the source of the pollution.
2 In a more restrictive sense, bacterial
indicators of pollution are associated
primarily with demonstration of con-
tamination of water, originating from
excreta of warm-blooded animals
(including man, domestic and wild
animals, and birds).
B Implications of Pollution of Intestinal
Origin
1 Intestinal wastes from warm-blooded
animals regularly include a wide
variety of genera and species of
bacteria. Among these the coliform
group may be listed, and species of
the genera Streptococcus, Lactobacillus,
Staphylococcus, Proteus, Pseudomonas,
certain spore-forming bacteria, and
others.
2 In addition, many kinds of pathogenic
bacteria and other microorganisms
may be released in wastes on an inter-
mittent basis, varying with the geo-
graphic area, state of community
health, nature and degree of waste
treatment, and other factors. These
may include the following:
a Bacteria: Species of Salmonella,
Shigella, Leptospira, Brucella,
Mycobacterium, and Vibrio comma.
b Viruses: A wide variety, including
that of infectious hepatitis. Polio-
viruses, Coxsackie virus, ECHO
viruses (enteric cytopathogenic
human orphan -- "viruses in search
of a disease"), and unspecified
viruses postulated to account for
outbreaks of diarrheal and upper
respiratory diseases of unknown
etiology, apparently infective by
the water-borne route.
c Protozoa: Endamoeba histolytica
As routinely practiced, bacterial
evidence of water pollution is a test
for the presence and numbers of
bacteria in wastes which, by their
presence, indicate that intestinal
pollution has occurred. In this con-
text, indicator groups discussed in
subsequent parts of this outline are
as follows:
a Coliform group and certain sub-
groupings
b Fecal streptococci and certain
sub groupings
c Miscellaneous indicators of water
quality
Evidence of water contamination by
intestinal wastes of warm-blooded
animals is regarded as evidence of
health hazard in the water being tested.
II PROPERTIES OF AN IDEAL INDICATOR
OF POLLUTION
A An "ideal" bacterial indicator of pollution
should:
1 Be applicable in all types of water
W.BA.481.8.77
2-1
-------
Bacteriological Indicators of Water Pollution
Always be present in water when
pathogenic bacterial constituents of
fecal contamination are present.
Ramifications of this include --
a Its density should have some direct
relationship to the degree of fecal
pollution.
b It should have greater survival time
in water than enteric pathogens,
throughout its course of natural
disappearance from the water body.
c It should disappear rapidly from
water following the disappearance
of pathogens, either through natural
or man-made processes.
d It always should be absent in a
bacteriologically safe water.
3 Lend itself to routine quantitative
testing procedures without interference
or confusion of results due to extra-
neous bacteria
4 Be harmless to man and other animals
B In all probability, an "ideal" bacterial
indicator does not exist. The discussion
of bacterial indicators of pollution in the
following parts of this outline include
consideration of the merits and limitations
of each group, with their applications in
evaluating bacterial quality of water.
in APPLICATIONS OF TESTS FOR
POLLUTION INDICATORS
A Tests for Compliance with Bacterial
Water Quality Standards
1 Potability tests on drinking water to
meet Interstate Quarantine or other
standards of regulatory agencies.
2 Determination of bacterial quality of
environmental water for which quality
standards may exist, such as shellfish
waters, recreational waters, water
resources for municipal or other
supplies.
3 Tests for compliance with established
standards in cases involving the pro-
tection or prosecution of municipalities,
industries, etc.
B Treatment Plant Process Control
1 Water treatment plants
2 Wastewater treatment plants
C Water Quality and Pollutant Source Monitoring
1 Determination of intestinal pollution
in surface water to determine type and
extent of treatment required for com-
pliance with standards
2 Tracing sources of pollution
3 Determination of effects on bacterial
flora, due to addition of organic or
other wastes
D Special Studies, such as
1 Tracing sources of intestinal pathogens
in epidemiological investigations
2 Investigations of problems due to the
Sphaerotilus group
3 Investigations of bacterial interference
to certain industrial processes, with
respect to such organisms as Pseudo-
monas, A chromobacter. or others
IV SANITARY SURVEY
The laboratory bacteriologist is not alone in
evaluation of indication of water pollution of
intestinal origin. On-site study (Sanitary
Survey) of the aquatic environment and
adjacent areas, by a qualified person, is a
necessary collateral study with the laboratory
work and frequently will reveal information
regarding potential bacteriological hazard
which may or may not be demonstrated
through laboratory findings from a single
sample or short series of samples.
2-2
-------
Bacteriological Indicators of Water Pollution
Part 2. The Coliform Group and Its Constituents
I ORIGINS AND DEFINITION
A Background
1 In 1885, Escherich, a pioneer bacteri-
ologist, recovered certain bacteria from
human feces, which he found in such
numbers and consistency as to lead him
to term these organisms "the charac-
teristic organism of human feces. "
He named these organisms Bacterium
cqli-commune and _B. lactis aerogenes.
In 1895, another bacteriologist,
Migula, renamed J3. coli commune as
Escherichia coli, which today is the
official name for the type species.
2 Later work has substantiated much of
the original concept of Escherich, but
has shown that the above species are
in fact a heterogeneous complex of
bacterial species and species variants.
a This heterogeneous group occurs not
only in human feces but representatives
also are to be found in many environ-
mental media, including sewage,
surface freshwaters of all categories,
in and on soils, vegetation, etc.
b The group may be subdivided into
various categories on the basis of
numerous biochemical and other
differential tests that may be applied.
B Composition of the Coliform Group
1 Current definition
As defined in "Standard Methods for the
Examination of Water and Wastewater"
(14th ed); "The coliform group includes
all of the aerobic and facultative
anaerobic, Gram-negative, nonspore-
forming rod-shaped bacteria which
ferment lactose with gas formation
within 48 hours at 35° C. "
2 The term "coliforms" or "coliform
group" is an inclusive one, including
the following bacteria which may
meet the definition above:
a Escherichia coli, E. aurescens,
E. freundii, E, intermedia
b Enterobacter aerogenes. E_.
c Biochemical intermediates between
the genera Escherichia andEntero-
bacter
There is no provision in the definition
of coliform bacteria for "atypical" or
"aberrant" coliform strains.
a An individual strain of any of the above
species may fail to meet one of the
criteria of the coliform group.
b Such an organism, by definition, is
not a member of the coliform group,
even though a taxonomic bacteriologist
may be perfectly correct in classifying
the strain in one of the above species.
H SUBDIVISION OF COLIFORMS INTO
"FECAL" AND "NONFECAL"
CATEGORIES
A Need
Single-test differentiations between
coliforms of "fecal" origin and those of
"nonfecal" origin are based on the
assumption that typical E_. coli and
closely related strains are of fecal
origin while JE. aerogenes and its close
relatives are not of direct fecal origin.
(The latter assumption is not fully borne
out by investigations at this Center.
See Table 1, IMViC Type --++). A
number of single differential tests have
been proposed to differentiate between
"fecal" and "nonfecal" coliforms.
2-3
-------
Bacteriological Indicators of Water Pollution
Without discussion of their relative merits,
several may be cited here:
B Types of Single-Test Differentiation
1 Determination of gas ratio
Fermentation of glucose by E_. coli
results in gas production, with
hydrogen and carbon dioxide being
produced in equal amounts.
Fermentation of glucose by E.
aerogenes results in generation of
twice as much carbon dioxide as
hydrogen.
Further studies suggested absolute
correlation between H /CO2 ratios
and the terminal pH resulting from
glucose fermentation. This led to the
substitution of the methyl red test.
2 Methyl red test
Glucose fermentation by IS. coli
typically results in a culture medium
having terminal pH in the range 4.2-
4. 6 (red color a positive test with the
addition of methyl red indicator).
E_. aerogenes typically results in a
culture medium having pH 5. 6 or
greater (yellow color, a negative test).
3 Indole
is
When tryptophane, an amino acid,
incorporated in a nutrient broth,
typical E. coli strains are capable of
producing indole (positive test) among
the end products, whereas E_. aerogenes
does not (negative test).
In reviewing technical literature, the
worker should be alert to the method
used to detect indole formation, as the
results may be greatly influenced by
the analytical procedure.
Voges-Proskauer test (acetylmethyl
carbinoltest)
The test is for detection of acetylmethyl
carbinol, a derivative of 2, 3, butylene-
glycol, as a result of glucose
fermentation in the presence of
peptone. E_. aerogenes produces
this end product (positive test)
whereas E_. coli gives a negative test.
a Experience with coliform cultures
giving a positive test has shown a
loss of this ability with storage on
laboratory media for 6 months to
2iyears, in 20 - 25% of cultures
(105 out of 458 cultures).
b Some workers consider that all
coliform bacteria produce acetyl-
methyl carbinol in glucose metab-
olism. These workers regard
acetylmethyl carbinol-negative
cultures as those which have
enzyme systems capable of further
degradation of acetylmethyl
carbinol to other end products
which do not give a positive test
with the analytical procedure.
Cultures giving a positive test for
acetylmethyl carbinol lack this
enzyme system.
c This reasoning leads to a hypothesis
(not experimentally proven) that the
change of reaction noted in certain
cultures in 4.a above is due to the
activation of a latent enzyme system.
5 Citrate utilization
Cultures of E. coli are unable to use
the carbon of citrates (negative test)
in their metabolism, whereas cultures
of E_. aerogenes are capable of using
the carbon of citrates in their metab-
olism (positive test).
Some workers (using Simmons Citrate
Agar) incorporate a pH indicator
(brom thymol blue) in the culture
medium in order to demonstrate the
typical alkaline reaction (pH 8.4 - 9.0)
resulting with citrate utilization.
6 Elevated temperature (Eijkman) test
a The test is based on evidence that
E. coli and other coliforms of fecal
2-4
-------
Bacteriological Indicators of Water Pollution
origin are capable of growing and
fermenting carbohydrates (glucose
or lactose) at temperatures signif-
icantly higher than the body tem-
perature of warm-blooded animals.
Organisms not associated with direct
fecal origin would give a negative
test result, through their inability
to grow at the elevated temperature.
While many media and techniques
have been proposed, EC Broth, a
medium developed by Perry and
Hajna, used as a confirmatory
medium for 24 hours at 44. 5 ±
0. 2°C are the current standard
medium and method.
While the "EC" terminology of the
medium suggests "j2. cpli" the
worker should not regard this as a
specific procedure for isolation of
_E. coli.
A similar medium. Boric Acid
Lactose Broth, has developed
by Levine and his associates. This
medium gives results virtually
identical with those obtained from
EC Broth, but requires 48 hours of
incubation.
Elevated temperature tests require
incubation in a water bath. Standard
Methods 14th Ed. requires this
temperature to be 44. 5 + 0. 2°C.
Various workers have urged use of
temperatures ranging between
43. GOC and 46. DOC. Most of these
recommendations have provided a
tolerance of + 0. 5° C from the rec-
ommended levels. However, some
workers, notably in the Shellfish
Program of the Public Health Service,
stipulate a temperature of 44. 5 +
0. 2°C. This requires use of a water
bath with forced circulation to main-
tain this close tolerance. This
tolerance range was instituted
in the 13th Edition of Standard Methods
and the laboratory worker should
conform to these new limits.
e The reliability of elevated temper-
ature tests is influenced by the
time required for the newly-
inoculated cultures to reach the
designated incubation temperature.
Critical workers insist on place-
ment of the cultures in the water
bath within 30 minutes, at most,
after inoculation.
7 Other tests
Numerous other tests for differentiation
between coliforms of fecal vs. nonfecal
origin have been proposed. Current
studies suggest little promise for the
following tests in this application:
uric acid test, cellobiose fermentation,
gelatin liquefaction, production of
hydrogen sulfide, sucrose fermentation,
and others.
C IMViC Classification
1 In 1938, Parr reported on a review of
a literature survey on biochemical tests
used to differentiate between coliforms
of fecal vs. nonfecal origin. A summary
follows:
Test
No. of times
used for dif-
ferentiation
Voges-Proskauer 22
reaction
Methyl red test 20
Citrate utilization 20
Indole test 15
Uric acid test 6
Cellobiose fermentation 4
Gelatin liquefaction 3
Eijkman test 2
Hydrogen sulfide 1
production
Sucrose fermentation 1
a-Methyl-d~glucoside 1
fermentation
2-5
-------
Bacteriological Indicators of Water Pollution
2 Based on this summary and on his own
studies. Parr recommended utilization
of a combination of tests, the indole,
methyl red, Voges-Proskauer, and the
citrate utilization tests for this differ-
entiation. This series of reactions is
designated by the mnemonic "IMViC".
Using this scheme, any coliform culture
can be described by an "IMViC Code"
according to the reactions for each
culture. Thus, a typical culture of
_E. coli would have a code ++--, and a
typical E_. aerogenes culture would
have a code —H-.
3 Groupings of coliforms into fecal,
non-fecal, and intermediate groups,
as shown in "Standard Methods for the
Examination of Water and Wastewater"
are shown at the bottom of this page.
D Need for Study of Multiple Cultures
All the systems used for differentiation
between coliforms of fecal vs. those of
nonfecal origin require isolation and study
of numerous pure cultures. Many workers
prefer to study at least 100 cultures from
any environmental source before attempting
to categorize the probable source of the
coliforms.
in NATURAL DISTRIBUTION OF COLIFORM
BACTERIA
A Sources of Background Information
Details of the voluminous background of
technical information on coliform bacteria
recovered from one or more environ-
mental media are beyond the scope of
this discussion. References of this
outline are suggested routes of entry
for workers seeking to explore this
topic.
B Studies on Coliform Distribution
1 Since 1960 numerous workers
have engaged in a continuing study of
the natural distribution of coliform
bacteria and an evaluation of pro-
cedures for differentiation between
coliforms of fecal vs. probable non-
fecal origin. Results of this work
have special significance because:
a Rigid uniformity of laboratory
methods have been applied through-
out the series of studies
b Studies are based on massive
numbers of cultures, far beyond
any similar studies heretofore
reported
INTERPRETATION OF IMVIC REACTIONS
Organism
Indole
foe
^roskauer
Citrate
Escherichia coli + or - +
Citrobacter freundii - +
Klebsiella-Enterobacter group + or - ^_
2-6
-------
Bacteriological Indicators of Water Pollution
c A wider variety of environmental and
biological sources are being studied
than in any previous series of reports.
d All studies are based on freshly
recovered pure culture isolates
from the designated sources.
e All studies are based on cultures
recovered from the widest feasible
geographic range, collected at all
seasons of the year.
2 Distribution of coliform types
Table 1 shows the consolidated results
of coliform distributions from various
biological and environmental sources.
a The results of these studies show a
high order of correlation between
known or probable fecal origin and
the typical _E. coli IMViC code
(++--). On the other hand,
human faces also includes
numbers ofJE_. aerogenes and other
IMViC types, which some regard as
"nonfecal" segments of the coliform
group. (Figure 1)
b The majority of coliforms attributable
to excretal origin tend to be limited
to a relatively small number of the
possible IMViC codes; on the other
hand, coliform bacteria recovered
from undisturbed soil, vegetation,
and insect life represent a wider
range of IMViC codes than fecal
sources, without clear dominance of
any one type. (Figure 2)
c The most prominant IMViC code
from nonfecal sources is the inter-
mediate type, -+-+, which accounts
for almost half the coliform cultures
recovered from soils, and a high
percentage of those recovered from
vegetation and from insects. It
would appear that if any coliform
segment could be termed a "soil
type" it would be IMViC code -+-+.
d It should not be surprising that
cultures of typical E. coli are
recovered in relatively smaller
numbers from sources judged,
on the basis of sanitary survey,
to be unpolluted. There is no
known way to exclude the influence
of limited fecal pollution from small
animals and birds in such environ-
ments.
e The distribution of coliform types
from human sources should be
regarded as a representative value
for large numbers of sources.
Investigations have shown that there
can be large differences in the
distribution of IMViC types from
person to person, or even from an
individual.
Differentiation between coliforms of
fecal vs. nonfecal origin
Table 2 is a summary of findings
based on a number of different criteria
for differentiating between coliforms
of fecal origin and those from other
sources.
a IMViC type ++-- is a measurement
of E_. coli, Variety I, and appears
to give reasonably good correlation
between known or highly probable
fecal origin and doubtful fecal origin.
b The combination of IMViC types,
-H—< + ( and -+--, gives
improved identification of probable
fecal origin, and appears also to
exclude most of the coliforms not
found in excreta of warm-blooded
animals in large numbers.
c While the indole, methyl red,
Voges Proskauer, and citrate
utilization tests, each used alone,
appear to give useful answers when
applied only to samples of known
pollution from fecal sources, the
interpretation is not as clear when
applied to coliforms from sources
believed to be remote from direct
fecal pollution.
2-7
-------
tsj
I
oo
Table 1. COLIFORM DISTRIBUTION BY IMViC TYPES AND ELEVATED TEMPERATURE
TEST FROM ENVIRONMENTAL AND BIOLOGICAL SOURCES
*120 of these
were ++—,
15 —H-,
11 -+--
*129 of these
were ++--,
27 -+-+,
5 ++-+
IMViC
type
Vegetation
No.
strains
i
H~-
--++
-+--
+++-
-+- +
++-+
-+++
-H-++
+-++
+
--H--
--+-
+-+-
+ — +
+
Total
No. EC +
% EC +
128
237
23
2
168
116
32
291
88
87
5
19
2
5
0
1203
169*
14.1*
%of
total
10. 6
19. 7
1.9
0.2
14.0
9.6
2.7
24.2
7.3
7. 2
0.4
1.6
0.2
0.4
<0. 1
Insects
No.
strains
134 !
113
0 !
0
332
118
28
254
46
42
0
0
0
8
9
1084
162*
14.9*
%o£
total
12.4
10.4
<0. 1
<0. 1
Soil
Undisturbed
No.
strains
131
443
78
7
!
30.6 : 1131
1
10.9 | 87
2.6
23.4
4. 2
3.9
<0. 1
<0. 1
<0. 1
0.7
0.8
181
159
67
4
1
53
6
0
0
2348
216
9.2
% of
total
5.6
18. 8
3. 3
0.3
48. 1
3. 7
7. 7
6. 8
2. 9
0.2
<0. 1
2.3
0. 3
<0. 1
<0. 1
Polluted
No.
strains
% of
total
536 j 80. 6
13 2.0
1
1
0.2
0 <0. 1
87
22
5
0
0
1
0
0
0
0
0
665
551
82.9
13. 0
3. 3
0. 7
<0. 1
<0. 1
0.2
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
Fecal sources
Hum
No.
strains
3932
245
99
106
50
35
21
6
14
2
0
0
0
0
2
4512
4349
96.4
an
%of
total
87.2
5.4
2.2
2.4
1. 1
0.8
Lives
No.
strains
2237
0
14
59
1
27
0. 5 j 0
0. 1 0
0.2
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
0
0
0
0
0
0
0
2339
2309
98.7
tock
%of
total
95. 6
Poultry
No.
strains
1857
%of
total
97. 9
i
<0. 1 1
0.6
2.5
<0. 1
1.2
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
20
0
5
11
0
0
0
0
0
0
0
0
2
1896
1765
93.0
0. 1
1. 1
<0. 1
0. 3
0. 6
<0. 1
i
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
<0. 1
a
a.
p
I
CO
o
>-»l
»
ef-
CD
o
3
-------
Bacteriological Indicators of Water Pollution
HUMAN
EC ® 96.4
BALB® 94.7
SUMMARY
Type
n- —
-+ —
f
__-M
EC®
BALB
Percent posifive
91.8
1.5
O.I
2.8
96.3
95.3
LIVESTOCK
EC® 98.7
BALB ® 98.6
POULTRY
EC® 93.0
8ALB® 92.5
FIGURE 1
CpLIFORMS
67 Soil Samples
(GeUireich. et. cil.)
Undisturbed
Soil
Polluted
Soil
FIGURE 2
2-9
-------
Bacteriological Indicators of Water Pollution
Table 2. COMPARISON OF COLIFORM STRAINS ISOLATED FROM WARM-BLOODED ANIMAL
FECES, FROM UNPOLLUTED SOILS AND POLLUTED SOILS WITH USE OF THE
IMViC REACTIONS AND THE ELEVATED TEMPERATURE TEST IN EC MEDIUM
AT 44.50 c ( + 0.50) (12th ed. 1965; Standard Methods for the Examination of Water
and Wastewater)
Test
+ + - -
+ + - -,
+ - - - and
- + - -
Indole positive
Methyl red positive
Voges-Proskauer positive
Citrate utilizers
Elevated temperature (EC)
positive .
Number of cultures
studied
Warm-blooded
animal feces
91. 8%
93. 3%
94. 0%
96. 9%
5. 1%
3. 6%
96. 4%
8, 747
Soil:
Unpolluted
5.6%
8.9%
19. 4%
75. 6%
40. 7%
88. 2%
9.2%
2,348
Soil:
Polluted
80. 6%
80. 7%
82. 7%
97. 9%
97. 3%
19. 2%
82. 9%
665
Vege-
tation
10. 6%
12.5%
52. 5%
63. 6%
56. 3%
85. 1%
14. 1%
1,203
Insects
12. 4%
13. 2%
52.4%
79.. 9%
40. 6%
86. 7%
14. 9%
1,084
Total Pure Cultures Studied: 14, 047
d The elevated temperature test gives
excellent correlation with samples
of known or highly probable fecal
origin. The presence of smaller,
but demonstrable, percentages of
such organisms in environmental
sources not interpreted as being
polluted could be attributed largely
to the warm-blooded wildlife in the
area, including birds, rodents, and
other small mammals.
e The elevated temperature test yields
results equal to those obtained from
the total IMViC code. It has marked
advantages in speed, ease and
simplicity of performance, and yields
quantitative results for each water
sample. Therefore, it is
the official standard method for
differentiation coliforms of
probable direct fecal origin from those
which may have become established
in the bacterial flora of the aquatic
or terrestrial habitat.
IV EVALUATION OF COLIFORMS AS
POLLUTION INDICATORS
A The Coliform Group as a Whole
1 Merits
a The absence of coliform bacteria is
evidence of a bacteriologically safe
water.
b The density of coliforms is roughly
proportional to the amount of
excretal pollution present.
c If pathogenic bacteria of intestinal
origin are present, coliform
bacteria also are present, in much
greater numbers.
d Coliforms are always present m the
intestines of humans and other warm-
blooded animals, and are eliminated
in large numbers in fecal wastes.
2-10
-------
Bacteriological Indicators of Water Pollution
e Coliforms are more persistent in
the aquatic environment than are
pathogenic bacteria of intestinal
origin.
f Coliforms are generally harmless
to humans and can be determined
quantitatively by routine laboratory
procedures.
2 Limitations
a Some of the constituents of the
coliform group have a wide environ-
mental distribution in addition to
their occurrence in the intestines
of warm-blooded animals.
b Some strains of the coliform group
may multiply in certain polluted
waters ("aftergrowth"), of high
nutritive values thereby adding to
the difficulty of evaluating a pollution
situation in the aquatic environment.
Members of the IS. aerogenes section
of the coliform are commonly
involved in this kind of problem.
c Because of occasional aftergrowth
problems, the age of the pollution
may be difficult to evaluate under
some circumstances.
d Tests for coliforms are subject to
interferences due to other kinds of
bacteria. False negative results
sometimes occur when species of
Pseudomonas are present. False
positive results sometimes occur
when two or more kinds of non-
coliforms produce gas from lactose,
when neither can do so alone
(synergism).
The Fecal Coliform Component of the
Coliform Group (as determined by elevated
temperature test)
1 Merits
a The majority (over 95% of the coli-
form bacteria from intestines of
warm-blooded animals grow at the
elevated temperature.
These organisms are of relatively
infrequent occurrence except in
association with fecal pollution.
Survival of the fecal coliform group
is shorter in environmental waters
than for the coliform group as
whole. It follows, then, that high
densities of fecal coliforms is
indicative of relatively recent
pollution.
Fecal coliforms generally do not
multiply outside the intestines of
warm-blooded animals. In certain
high-carbohydrate wastes, such as
from the sugar beet refineries,
exceptions have been noted.
In some wastes, notably those from
pulp and paper mills, Klebsiella has
been found in large numbers
utilizing the elevated temperature
test. There has been much contro-
versy about whether the occurrence
of Klebsiella is due to aftergrowth
due to soluble carbohydrates in such
wastes. The significance of
Klebsiella as an indicator of direct
discharge of intestinal wastes thus
is under challenge. The issue is
still further complicated by questions
over whether Klebsiella is in and of
itself a pathogenic organism or is
potentially pathogenic. This is a
serious problem which is the subject
of intensive research efforts.
2 Limitations
Feces from warm-blooded animals
include some (though proportionately
low) numbers of coliforms which do
not yield a positive fecal coliform
test when the elevated temperature
test is used as the criterion of
differentiation. These organisms
are E. coli varieties by present
taxonomic classification.
There is at present no established
and consistent correlation between
2-11
-------
Bacteriological Indicators of Water Pollution
ratios of total coliforms/fecal
coliforms in interpreting sanitary
quality of environmental waters.
In domestic sewage, the fecal
coliform density commonly is
greater than 90% of the total
coliform density. In environmental
waters relatively free from recent
pollution, the fecal coliform density
may range from 10-30% of the total
coliforms. There are, however,
too many variables relating to
water-borne wastes and surface
water runoff to permit sweeping
generalization on the numerical
relationships between fecal- and
total coliforms.
c Studies have been made
regarding the survival of fecal
coliforms in polluted waters
compared with that of enteric
pathogenic bacteria. In recent
pollution studies, species of
Salmonella have been found in the
presence of 220 fecal coliforms per
100 ml (Spino), and 110 fecal
coliforms per 100 ml (Brezenski,
Raritan Bay Project).
The issue of the Klebsiella problem
described in an earlier paragraph
may ultimately be resolved as a
merit or as a limitation of the value
of the fecal coliform test.
V APPLICATIONS OF COLIFORM TESTS
A Current Status in Official Tests
1 The coliform group is designated, in
"Standard Methods for the Examination
of Water and Wastewater" (14th ed.,
1975), through the Completed Test
MPN procedure as the official test
for bacteriological potability of water.
The Confirmed Test MPN procedure
is accepted where it has been demon-
strated, through comparative tests,
to yield results equivalent to the
Completed Test. The membrane filter
method also is accepted for examination
of waters subject to interstate regulation.
2 The 12th edition of Standard Methods
introduced a standard test for fecal
coliform bacteria. It is emphasized
that this is to be used in pollution
studies, and does not apply to the
evaluation of water for potability.
This procedure has been continued in
the 13th and 14th Editions.
B Applications
1 Tests for the coliform group as a
whole are used in official tests to
comply with interstate drinking water
standards, state standards for shell-
fish waters, and in most, if not all,
cases where bacterial standards of
water quality have been established
for such use as in recreational or
bathing waters, water supplies, or
industrial supplies. Laboratory
personnel should be aware of possible
implementation of the fecal coliform
group as the official test for recreational
and bathing waters.
2 The fecal coliform test has application
in water quality surveys, as an adjunct
to determination of total coliform
density. The fecal coliform test is
being used increasingly in all water
quality surveys.
3 It is emphasized that no responsible
worker advocates substitution of a
fecal coliform test for total coliforms
in evaluating drinking water quality.
2-12
-------
Bacteriological Indicators of Water Pollution
Part 3. The Fecal Streptococci
I INTRODUCTION
Investigations regarding streptococci
progressed from the streptococci of medical
concern to those which were distributed in
differing environmental conditions which,
again, related to the welfare of man. The
streptococci were originally reported by Laws
and Andrews (1894), and Houston (1899, 1900)
considered those streptococci, which we now
call "fecal streptococci, "as ... "indicative
of dangerous pollution, since they are readily
demonstrable in waters recently polluted and
seemingly altogether absent from waters above
suspicion of contamination.
From their discovery to the present time the
fecal streptococci appear characteristic of
fecal pollution, being consistently present in
both the feces of all warm-blooded animals
and in the environment associated with animal
discharges. As early as 1910 fecal strepto-
cocci were proposed as indicators to the
Metropolitan Water Board of London.
However, little progress resulted in the
United States until improved methods of
detection and enumeration appeared after
World War II.
Renewed interest in the group as indicators
began with the introduction of azide dextrose
broth in 1950, (Mallmann & Seligmann, 1950).
The method which is in the current edition
of Standard Methods appeared soon after.
(Litsky, et al. 1955).
With the advent of improved methods for
detection and enumeration of fecal strep-
tococci, significant body of technical
literature has appeared.
This outline will consider the findings of
various investigators regarding the fecal
streptococci and the significance of discharges
of these organisms into the aquatic environment.
II FECAL MATERIALS
A Definition
The terms "enterococci, " "fecal
streptococci, " "Group D streptococci, "
"Streptococcus fecalis, " and even
' streptococci' have been used in a loose
and interchangeable manner to indicate
the streptococci present in the enteric
tract of warm-blooded animals or of the
fresh fecal material excreted therefrom.
Enterococci are characterized by specific
taxonomic biochemistry. Serological
procedures differentiate the Group D
streptococci from the various groups.
Although they overlap, the three groups,
fecal streptococcus, enterococcus, and
Group D streptococcus, are not synonymous.
Because our emphasis is on indicators of
unsanitary origin, fecal streptococcus is
the more appropriate term and will include
the enterococcus as well as other groups.
Increasing attention is being paid to certain
streptococci found in humans and certain
birds whic were, at one time, considered
to be biotypes of Str. faecalis or
Str. faecium and therefore legitimate
fecal streptococci. These are now con-
sidered to be in a separate group in their
own right, the Group Q streptococci.
A rigid definition of the fecal streptococcus
group is not possible with our present
knowledge. The British Ministry of Health
(1956) defines the organisms as "Gram-
positive" cocci, generally occurring in
pairs or short chains, growing in the
presence of bile salt, usually capable of
development at 45° C, producing acid but
not gas in mannitol and lactose, failing to
attack raffinose, failing to reduce nitrate
to nitrite, producing acid in litmus milk
and precipitating the casein in the form of
a loose, but solid curd, and exhibiting a
greater resistance to heat, to alkaline
conditions and to high concentrations of
salt than most vegetative bacteria. "
However, it is pointed out that "streptococci
departing in one or more particulars from
the type species cannot be disregarded
in water. "
2-13
-------
Baterlological Indicators of Water Pollution
Standard Methods (14th ed., 1975)
describes the fecal streptococci as pure
culture selective medium organisms
which are Catalase negative and capable
of originating growth in BHI broth
(45°C. for 48 hours) and Bile broth
medium (35° C. for 3 days).
For the proposes of this outline, and in line
with the consensus of most water micro-
biologists in this country, the general
definition of the fecal streptococci is-.
. . . "The group composed of Group Dand Q
species consistently present in
significant numbers in fresh fecal
excreta of warm-blooded animals,
which includes all of the enterococcus
group in addition to other groups of
streptococci. "
B Species Isolated
1 Findings
a Human feces
Examination of human fecal specimens
yields a high percentage of the
enterococcus group and usually
demonstration of the S. salivarius
which is generally considered a
member of the human throat flora
and to be surviving in human fecal
materials rather than actively
multiplying in the enteric tract.
Also present would be a small
percentage of variants or biotypes
of the enterococcus group.
b Nonhuman Feces
1) Fecal material which are from
nonhuman and not from fowl will
yield high percentages of the
S. bovis and/or S. equinus
organisms with a concomitantly
reduced percentage of the
enterococcus group.
2) Fowl excreta
Excrement from fowl characteris-
tically yields a large percentage
of enterococcal biotypes
(Group Q) as well as a significant
percentage of enterococcus
group.
2 Significance
Species associations with particular
animal hosts is an established fact and
leads to the important laboratory
technique of partition counting of colonies
from, the membrane filter or agar
pour plates in order to establish or
confirm the source of excretal
pollution in certain aquatic investi-
gations .
It is important to realize that a suitable
medium is necessary in order to
allow all of the streptococci which
we consider to be fecal streptococci
to grow in order to give credence to
the derived opinions, Use of liquid
growth media into which direct
inoculations from the sample are
made have not proven to be successful
for partition counting due to the differing
growth rates of the various species of
streptococci altering the original
percentage relationships. Due to the
limited survival capabilities of some
of the fecal streptococci it is necessary
to sample fresh fecal material or water
samples in close proximity to the
pollution source especially when
multiple sources are contributing to a
reach of water. Also the pH range
must be within the range of 4. 0-9. 0.
Standard Methods (14th ed., 1975) now
includes a schematic allowing for the
identification of fecal streptococci types
present within a given sample.
HI FECAL STREPTOCOCCI IN THE
AQUATIC ENVIRONMENT
A General
From the foregoing it is appears that
the preponderant human fecal streptococci
are composed of the enterococcus group
and, as this is the case, several media are
presently available which will detect only
the enterococcal group will be suitable
for use with aquatic samples which are
known to be contaminated or potentially
contaminated with purely domestic
2-14
-------
Bacteriological Indicators of Water Pollution
(human) wastes. On the other hand,
when it is known or suspected that other-
than-human wastes have potential egress
to the aquatic environment under investi-
gation, it is necessary to utilize those
media which are capable of quantitating
the whole of the fecal streptococci group.
B Stormwaters and Combined Sewers
1 General
Storm sewers are a series of pipes
and conduits which receive surface
runoffs from the action of rainstorms
and do not include sewage which are
borne by a system of sanitary sewers.
Combined sewers receive both the storm
runoff and the water-borne
wastes of the sanitary system.
Both storm water and combined
sewer flows have been found to
usually contain large quantities of
fecal streptococci in numbers which
generally are larger than those of
the fecal coliform indicator organisms.
2 Bacteriological Findings *
Table 1 represents, in a modified form,
some of the findings of Geldreich and
Kenner (1969) with respect to the
densities of fecal streptococci when
considering Domestic sewage in contrast
to Stormwaters:
The Ratio FC/FS is that of the
Fecal coliform and Fecal streptococci
and it will be noted that in each case,
when considering the Domestic
Sewage, it is 4. 0 or greater while
it is less than 0.7 for Stormwaters.
The use of this ratio is useful to
identify the source of pollution as
being human or nonhuman warm-
blooded animal polluted. When the ratio
is greater than 4. 0 it is considered to be
human waste contaminated while a ratio
of less than 0. 7 is considered to be
nonhuman. It is evident that the storm-
waters have been primarily polluted by
excreta of rats and other rodents and
possibly domestic and/or farm animals.
Species differences are the main cause
of different fecal coliform-fecal
streptococci ratios. Table 2 compares
fecal streptococcus and fecal coliform
counts for different species. Even-
though individuals vary widely, masses
of individuals in a species have charac-
teristic proportion of indicators.
C Surface Waters
In general, the occurrence of fecal
streptococci indicates fecal pollution and
its absence indicates that little or no
warm-blooded fecal contribution. In
studies of remote surface waters the fecal
streptococci are infrequently isolated and
occurrences of small numbers can be
attributed to wild life and/or snow melts
and resultant drainage flows.
Various examples of fecal streptococcal
occurrences are shown in Table 3 in
relation to surface waters of widely varying
quality. (Geldreich and Kenner 1969)
IV FECAL STREPTOCOCCI:
AND LIMITATIONS
ADVANTAGES
A General
Serious studies concerning the streptococci
were instituted when it became apparent
that they were the agents responsible or
suspected for a wide variety of human
diseases. Natural priority then focused
itself to the taxonomy of these organisms
and this study is still causing consternation
as more and more microbiological techniques
have been brought to bear on these questions.
The sanitary microbiologist is concerned
with those streptococci which inhabit the
enteric tract of warm-blooded animals,
their detection, and utilization in develop-
ing a criterium for water quality standards.
2-15
-------
Bacteriological Indicators of Water Pollution
Table 1
DISTRIBUTION OF FECAL STREPTOCOCCI
IN DOMESTIC SEWAGES AND STORMWATER
RUNOFFS
Fecal Streptococci
per 100 ml Ratio
Water Source median values FC/FS
Domestic Sewage
Preston, ID
Fargo, ND
Moorehead, MN
Cincinnati, OH
Lawrence, MA
Monroe, MI
Denver, CO
Stormwater
64,000
290, 000
330,000
2,470,000
4, 500, 000
700, 000
2,900,000
Business District 51,000
Residential 150,000
Rural 58, 000
5.3
4.5
4.9
4.4
4.0
27.9
16.9
0.26
0.04
0.05
Table 3
INDICATOR ORGANISMS IN SURFACE
WATERS
Densities/100 ml
Fecal Fecal
Water Source coliform streptococci
Prairie Watersheds
Cherry Creek, WY 90 83
Saline River, KS 95 180
Cub River, ID 110 160
Clear Creek, CO 170 110
Recreational Waters
Lake Mead 2 444
Lake Moovalaya 9 170
Colorado River 4 256
Whitman River 32 88
Merrimack River 100 96
Public Water Intakes
Missouri River (1959)
Mile 470.5 11,500 39,500
Mile 434.5 22,000 79, 000
Mile 408.8 14, 000 59,000
Tablet. ESTIMATED
PER CAPITA CONTRIBUTION OF INDICATOR MICROORGANISMS
FROM SOME ANIMALS*
Average indicator
density per gram
of feces
Animals
Man
.Duck
Sheep
Chicken
Cow
Turkey
Pig
Avg wt of
Feces/ 24 hr,
wet wt, g
150
336
1, 130
182
23,600
448
2,700
Fecal
coliform,
million
13.0
33.0
16.0
1.3
0.23
0.29
3.3
Fecal
streptococci,
million
3.0
54.0
38.0
3.4
1.3
2.8
84.0
Average contribution
per capita per 24 hr
Fecal
coliform,
million
2,000
11,000
18, 000
240
5,400
130
8,900
Fecal
streptococci,
million
450
18.000
43, 000
620
31,000
1,300
230, 000
Ratio
FC/FS
4.4
0.6
0.4
0.4
0.2
0.1
0.04
^Publication WP-20-3, P. 102
2-16
-------
Bacteriological Indicators of Water Pollution
Kabler (1962) discussed the slow acceptance
of the fecal streptococci as indicators of
pollution resulting from:
1 Multiplicity and difficulty of laboratory
procedures
2 Poor agreement between methods of
quantitative enumeration
3 Lack of systematic studies of ....
a sources
b survival, and
c interpretations, and
4 Undue attention to the S. faecalls group.
Increased attention to the fecal streptococci,
especially during the last decade, have
clarified many of the earlier cloudy issues
and have elevated the stature of these
organisms as indicators of pollution.
Court precedents establishing legal status
and recommendations of various technical
advisory boards have placed the fecal
coliform group in a position of primacy
in many water quality applications. The
fecal streptococci have evolved from a
position of a theoretically useful indicator
to one which was ancillary to the coliforms
to one which was useful when discrepancies
or questions evolved as to the validity of
the coliform data to one where an equality
status was achieved in certain applications.
In the future it is anticipated that, for
certain applications, the fecal streptococci
will achieve a position of primacy for
useful data, and, as indicated by Litsky
(1955) "be taken out of the realm of step-
children and given their legitimate place
in the field of santiary bacteriology as
indicators of sewage pollution. "
B Advantages and Limitations
1 Survival
In general, the fecal streptococci have
been observed to have a more limited
survival time in the aquatic environment
when compared to the coliform group.
They are rivaled in this respect only
by the fecal coliforms. Except for cases
of persistence in waters of high electro-
lytic content, as may be common to
irrigation waters, the fecal streptococci
have not been observed to multiply in
polluted waters as may sometimes be
observed for some of the coliforms.
Fecal streptococci usually require a
greater abundance of nutrients for sur-
vival as compared to the coliforms and
the coliforms are more dependent upon
the oxygen tension in the waterbody.
In a number of situations it was concluded
that the fecal streptococci reached an
extinction point more rapidly in warmer
waters while the reverse was true in the
colder situations as the coliforms now
were totally eliminated sooner.
2 Resistance to Disinfection
In artificial pools the source of
contamination by the bathers is
usually limited to throat and skin
flora and thus increasing attention
has been paid to indicators other
than those traditionally from the
enteric tract. Thus, one of the
organisms considered to be a fecal
streptococci, namely, S. salivarius.
can be a more reliable indicator
when detected along with the other
fecal streptococci especially since
studies have confirmed the greater
resistance of the fecal streptococci
to chlorination. This greater
resistance to chlorination, when
compared to the fecal coliforms, is
important since the dieoff curve
differences are insignificant when
the curves of the fecal coliforms
are compared to various Gram
negative pathogenic bacteria which
reduces their effectiveness as
indicators.
3 Ubiquitous Strains
Among the fecal streptococcus are
two organisms, one a biotype and
the other a variety of the S. faecalis,
which, being ubiquitous (omnipresent)
have limited sanitary significance.
2-17
-------
Bacteriological Indicators of Water Pollution
The biotype, or atypical, S. faecalis
is characterized by its ability to
hydrolyze starch while the varietal
form, liquefaciens, is nonbeta
haemolytic and capable of liquefying
gelatin. Quantitation of these organisms
in anomalous conditions is due to their
capability of survival in soil or high
electrolytic waters and in waters with
a temperature of less than 12 Degrees C.
Samples have been encountered which
have been devoid of fecal coliforms
and yet contain a substantial number of
"fecal streptococci" of which these
ubiquitous strains constitute the majority
or all of the isolations when analyzed
b io che mi cally.
V STANDARDS AND CRITERIA
Acceptance and utilization of Total Coliform
criteria, which must now be considered a
pioneering effort, has largely been supplanted
in concept and in fact by the fecal coliforms
in establishing standards for recreational
waters.
The first significant approach to the utiliza-
tion of the fecal streptococci as a criterium
for recreational water standards occurred in
1966 when a technical committee recommended
the utilization of the fecal streptococci with the
total coliforms as criteria for standards
pertaining to the Calumet River and lower
Lake Michigan waters. Several sets of
criteria were established to fit the intended
uses for this area. The use of the fecal
streptococci as a criterium is indicated to
be tentative pending the accumulation of
existing densities and could be modified in
future standards.
With the existing state-of-the-art knowledge
of the presence of the fecal streptococci in
waters containing low numbers of fecal
coliforms it is difficult to establish a specific
fecal streptococcus density limit of below
100 organisms/100 ml when used alone or
in conjunction with the total coliforms.
The most useful application of the fecal
streptococcus test is in the development
of the fecal coliform: fecal streptococcus
ratio as previously described.
2-18
-------
Bacteriological Indicators of Water Pollution
Part 4. Other Bacterial Indicators of Pollution
I TOT A L BA CTERIA L COUNTS
A Historical
1 The early studies of Robert Koch led
him to develop tentative standards of
water quality based on a limitation of
not more than 100 bacterial colonies
per ml on a gelatin plating medium
incubated 3 days at 20° C.
2 Later developments led to inoculation
of samples on duplicate plating media,
with one set incubated at 37° C and the
other at 2QOC.
a Results were used to develop a ratio
between the 37° C counts and the
counts.
Waters having a predominant
count at 37° C were regarded as
being of probable sanitary signifi-
cance, while those giving
predominant counts at 20° C were
considered to be of probable soil
origin, or natural inhabitants of
the water being examined.
B Groups Tested
There is no such thing as "total" bacterial
count in terms of a laboratory determination.
1 Direct microscopic counts do not
differentiate between living and dead
cells.
2 Plate counting methods enumerate only
the bacteria which are capable of using
the culture medium provided, under the
temperature and other growth conditions
used as a standard procedure. No one
culture medium and set of growth
conditions can provide, simultaneously,
an acceptable environment for all the
heterogeneous, often conflicting,
requirements of the total range of
bacteria which may be recovered from
waters.
C Utilization of Total Counts
1 Total bacterial counts, using plating
methods, are useful for:
a Detection of changes in the bacterial
composition of a water source
b Process control procedures in
treatment plant operations
c Determination of sanitary conditions
in plant equipment or distributional
systems
2 Serious limitations in total bacterial
counts exist because:
a No information is given regarding
possible or probable fecal origin
of bacterial changes. Large numbers
of bacteria can sometimes be
cultivated from waters known to be
free of fecal pollution.
b No information of any kind is given
about the species of bacteria
cultivated.
c There is no differentiation between
harmless or potentially dangerous
forms,
3 Status of total counts
Methodology for the determination of
the Standard Plate Count has been
retained in the 14th Edition of Standard
Methods for the stated reason:
... "total counts may yield useful
information about the quality of
water and supporting data on the
significance of coliform results ...
also, useful in judging the efficiency
in operation of various water treat-
ment processes and may have sig-
nificant application as an in-plant
control test. It is also valuable for
periodic checking of finished
distribution water"
(abridged for this inclusion)
2-19
-------
Bacteriological Indicators of Water Pollution
Technique for the Standard Plate
Count is necessary for the
performance of the Distilled Water
Suitability Test as outlined in
Standard Methods and elsewhere within
this manual.
B Spore-Forming Bacteria (Clostridium
perfringens, or C. welchii)
1 Distribution
This is one of the most widely distributed
species of bacteria. It is regularly
present in the intestinal tract of warm-
blooded animals.
2 Nature of organism
C. perfringens is a Gram-positive,
spore-forming rod. The spores cause
a distinct swelling of the cell when
formed. The organism is extremely
active in fermentation of carbohydrates,
and produces the well-known "stormy
fermentation" of milk.
3 Status
The organism, when present, indicates
that pollution has occurred at some
time. However, because of the ex-
tremely extended viability of the spores,
it is impossible to obtain even an '.
approximation of the recency of pollution
based only on the presence of
C_. perfringens.
The presence of the organism does not
necessarily indicate an unsafe water.
C Tests for Pathogenic Bacteria of Intestinal
Origin
1 Groups considered include Salmonella
sp, Shigella sp, Vibrio comma,
Mycobacterium sp, Pasteurella sp,
Leptospira sp, and others.
2 Merits of direct tests:
Demonstration of any pathogenic
species would demonstrate an
unsatisfactory water quality, hazardous
to persons consuming or coming into
contact with that water.
3 Limitations
a There is no available routine pro-
cedure for detection of the full
range of pathogenic bacteria cited
above.
b Quantitative methods are not avail-
able for routine application to any
of the above.
c The intermittent release of these
pathogens makes it impossible to
regard water as safe, even in the
absence of pathogens.
d After detection, the public already
would have been exposed to the
organism; thus, there is no built-in
margin of safety, as exists with
tests for the coliform group.
4 Applications
a In tracing the source of pathogenic
bacteria in epidemiological investi-
gations
b In special research projects
c In water quality studies concerned
with enforcement actions against
pollution, increasing attention is
being given to the demonstration of
enteric pathogenic bacteria in the
presence of the bacterial indicators
of pollution.
D Miscellaneous Indicators
It is beyond this discussion to explore the
total range of microbiological indicators
of pollution that have been proposed and
2-20
-------
Bacteriological Indicators of Water Pollution
investigated to some extent. Mention can
be made, however, of consideration of
tests for the following.
1 Bacteriophages specific for any of a
number of kinds of bacteria
2 Tests for Enterovirus
3 Serological procedures for detection
of coliforms and other indicators: a
certain amount of recent attention has
been given to applications of fluorescent
antibodies in such tests
4 Tests for Klebsiella
5 Tests for Pseudomonas aeruginosa
6 Tests for Salmonella
7 Tests for Fungi
8 Tests for Staphylococcus
REFERENCES
1 Standard Methods for the Examination of
Water and Wastewater, 14th ed.,
APHA, AWWA, WPCF. Published by
American Public Health Association,
1790 Broadway, New York, N. Y. 1975
2 Prescott, S. C., Winslow, C.E.A., and
McCrady, M. Water Bacteriology.
John Wiley & Sons, Inc. 1946.
3 Parr, L.W. Coliform Intermediates in
Human Feces. Jour. Bact. 36:1.
1938.
4 Clark, H. F. and Kabler, P.W. The
Physiology of the Coliform Group.
Proceedings of the Rudolfs Research
Conference on Principles and Appli-
cations in Aquatic Microbiology. 1963.
5 Geldreich, E. E., Bordner, R.H., Huff,
C.B., Clark, H. F. , and Kabler, P.W.
Type Distribution of Coliform Bacteria
in the Feces of Warm-Blooded Animals.
JWPCF. 34:295-301. 1962.
6 Geldreich et al. The Fecal Coli-Aerogenes
Flora of Soils from Various Geographic
Areas. Journal of Applied Bacteriology
25:87-93. 1962.
7 Geldreich, E.E., Kenner, B.A., and
Kabler, P.W. Occurrence of
Coliforms, Fecal Coliforms, and
Streptococci on Vegetation and Insects.
Applied Microbiology. 12:63-69. 1964.
8 Kabler, P.W., Clark, H.F., and
Geldreich, E.E. Sanitary Significance
of Coliform and Fecal Coliform
Organisms in Surface Water. Public
Health Reports. 79:58-60. 1964.
9 Clark, H.F. and Kabler, P.W.
Re-evaluation of the Significance of the
Coliform Bacteria. Journal AWWA.
56:931-936. 1964.
10 Kenner, B.S., Clark, H.F., and
Kabler, P.W. Fecal Streptococci.
II. Quantification in Feces. Am. J.
Public Health. 50:1553-59. 1960.
11 Litsky, W., Mailman, W.L., and Fifield,
C.W. Comparison of MPN of
Escherichia coli and Enterococci in
River Water. Am. Jour. Public Health.
45:1949. 1955.
12 Medrek, T. F. and Litsky, W.
Comparative Incidence of Coliform
Bacteria and Enterococci in
Undisturbed Soil. Applied Micro-
biology. 8:60-63. 1960.
13 Mailman, W.L., and Litsky, W.
Survival of Selected Enteric Organisms
in Various Types of Soil. Am. J.
Public Health. 41:38-44. 1950.
14 ' Mailman, W.L., and Seligman, E.B., Jr.
A Comparative Study of Media for
Detection of Streptococci in Water and
Sewage. Am. J. Public Health.
40:286-89. 1950.
15 Ministry of Health (London). The
Bacterial Examination of Water Supplies.
Reports on Public Health and Medical
Subjects. 71:34.
2-21
-------
Bacteriological Indicators of Water Pollution
16 Morris, W. and Weaver, R.H.
Streptococci as Indices of Pollution
in Well Water. Applied Microbiology,
2:282-285. 1954.
17 Mundt, J.O., Coggin, J.H., Jr., and
Johnson, L. F. Growth of
Streptococcus fecalis var. liquefaciens
on Plants. AppliedMicrobiology.
10:-552-555. 1962.
18 Geldreich, E.E. Sanitary Significance
of Fecal Coliforms in the Environment.
U. S. Department of the Interior.
FWPCA Publ. WP-20-3. 1966.
19 Geldreich, E.E. and Kenner, B.A.
Concepts of Fecal Streptococci in
Stream Pollution. J. WPCF. 41:R336.
1969.
20 Kabler, P.W. Purification and Sanitary
Control of Water (Potable and Waste)
Ann. Rev. of Microbiol. 16:127. 1962.
21 Litsky, W., Mailman, W.L., and Fifield,
C. W. Comparison of the Most Probable
Numbers of Escherichia coli and
Enterococci in River Waters. A. J. P. H.
45:1049. 1955.
22 Geldreich, E.E. Applying Bacteriological
Parameters to Recreational Water
Quality. J. AWWA. 62:113. 1970.
23 Geldreich, E.E., Best, L. C., Kenner, B.A.
and Van Donsel, D. J. The Bacteriolog-
ical Aspects of Stormwater Pollution.
J. WPCF. 40:1860. 1968.
24 FWPCA Report of Water Quality Criteria
Calumet Area - Lower Lake Michigan,
Chicago, IL. Jan. 1966.
This outline was prepared by H. L. Jeter,
Chief, Program Support Training Branch,
and revised by R. Russomanno, Microbiologist,
USEPA, Cincinnati, Ohio 45268
Descriptors: Coliforms, Escherichia coli,
Fecal Coliforms, Fecal Streptococci,
Indicator Bacteria, Microbiology, Sewage
Bacteria, Water Pollution
2-22
-------
EXAMINATION OF WATER FOR COLIFORM AND
FECAL STREPTOCOCCUS CROUPS
(Multiple Dilution Tube [MPN] Methods)
I INTRODUCTION
The subject matter of this outline is contained
in three parts, as follows:
A Part 1
1 Fundamental aspects of multiple dilution
tube ("most probable numbers") tests,
both from a qualitative and a quantitative
viewpoint.
2 Laboratory bench records.
3 Useful techniques in multiple dilution
tube methods.
4 Standard supplies, equipment, and
media in multiple dilution tube tests.
B Part 2
Detailed, day-by-day, procedures in tests
for the coliform group and subgroups
within the coliform group.
C Part 3
Detailed, day-by-day, procedures in tests
for members of the fecal streptococci.
D Application of Tests to Routine Examinations
The following considerations (Table 1) apply
to the selection of the Presumptive Test,
the Confirmed Test, and the Completed
Test. Termination of testing at the
Presumptive Test level is not practiced
by laboratories of this agency. It must
be realized that the Presumptive Test alone
has limited use when water quality is to
be determined.
TABLE 1
Examination Terminated at -
Type of Receiving
Water
Sewage Receiving
Treatment Plant - Raw
Chlorinated
Bathing
Drinking
Other Information
Presumptive
Test
Applicable
Applicable
Not Done
Not Done
Not Done
Confirmed Test
Applicable
Applicable
Applicable
Applicable
Applicable
Applicable in all
cases where Pre-
sumptive Test alone
is unreliable.
| Completed Test
Important where results
are to be used for control
of raw or finished water.
Application to a statis-
tically valid number of
samples from the
Confirmed Test to estab-
lish its validity in
determining the sanitary
quality .
NOTE: Mention of commercial products and manufacturers does not imply endorsement by the
Environmental Protection Agency.
W.BA. 3m. 8a. 78
3-1
-------
MPN Methods
II BASIS OF MULTIPLE TUBE TESTS
A Qualitative Aspects
1 For purely qualitative aspects of testing
for indicator organisms, it is convenient
to consider the tests applied to one
sample portion, inoculated into a tube
of culture medium, and the follow-up
examinations and tests on results of the
original inoculation. Results of testing
procedures are definite: positive
(presence of the organism-group is
demonstrated) or negative (presence of
the organism-group is not demonstrated.)
2 Test procedures are based on certain
fundamental assumptions:
a First, even if only one living cell of
the test organism is present in the
sample, it will be able to grow when
introduced into the primary inoculation
medium;
b Second, growth of the test organism
in the culture medium will produce
a result which indicates presence of
the test organism; and,
c Third, extraneous organisms will
not grow, or if they do grow, they
will not limit growth of the test
organism; nor will they produce
growth effects that will be confused
with those of the bacterial group for
which the test is designed.
3 Meeting these assumptions usually
makes it necessary to conduct the tests
in a series of stages (for example, the
Presumptive, Confirmed, and Completed
Test stages, respectively, of standard
tests for the coliform group).
4 Features of a full, multi-stage test
a First stage: The culture medium
usually serves primarily as an
enrichment medium for the group
tested. A good first-stage growth
medium should support growth of all
the living cells of the group tested,
and it should include provision for
indicating the presence of the test
organism being studied. A first-
stage medium may include some
component which inhibits growth
of extraneous bacteria, but this
feature never should be included
if it also inhibits growth of any
cells of the group for which the
test is designed. The Presumptive
Test for the coliform group is a
good example. The medium
supports growth, presumably, of
all living cells of the coliform
group; the culture container has a
fermentation vial for demonstration
of gas production resulting from
lactose fermentation by coliform
bacteria, if present; and sodium
lauryl sulfate may be included in
one of the approved media for
suppression of growth of certain
noncoliform bacteria. This
additive apparently has no adverse
effect on growth of members of the
coliform group in the concentration
used. If the result of the first-stage
test is negative, the study of the
culture is terminated, and the result
is recorded as a negative test. No
further study is made of negative
tests. If the result of the first-
stage test is positive, the culture
may be subjected to further study
to verify the findings of the first
stage.
Second stage: A transfer is made
from positive cultures of the first-
stage test to a second culture medium.
This test stage emphasizes provision
to reduce confusion of results due to
growth effects of extraneous bacteria,
commonly achieved by addition of
selective inhibitory agents. (The
Confirmed Test for coliforms meets
these requirements. Lactose and
fermentation vials are provided for
demonstration of coliforms in the
medium. Brilliant green dye and
bile salts are included as inhibitory
agents which tend to suppress growth
of practically all kinds of noncoliform
bacteria, but do not suppress growth
of coliform bacteria when used as
directed).
3-2
-------
MPN Methods
If result of the second- stange test is
negative, the study of the culture is
terminated, and the result is recorded
as a negative test. A negative test here
means that the positive results of the
first-stage test were "false positive, "
due to one or more kinds of extraneous
bacteria. A positive second-stage test
is partial confirmation of the positive
results obtained in the first-stage test;
the culture may be subjected to final
identification through application of still
further testing procedures. In routine
practice, most sample examinations
are terminated at the end of the second
stage, on the assumption that the result
would be positive if carried to the third,
and final stage. This practice should be
followed only if adequate testing is done
to demonstrate that the assumption is
valid. Some workers recommend contin-
uing at least 5% of all sample examina-
tions to the third stage to demonstrate
the reliability of the second- stage result;
B Quantitative Aspects of Tests
1 These methods for determining bacterial
numbers are based on the assumption
that the bacteria can be separated from
one another (by shaking or other means)
resulting in a suspension of individual
bacterial cells, uniformly distributed
through the original sample when the
primary inoculation is made.
2 Multiple dilution tube tests for quantita-
tive determinations apply a Most Probable
Number (MPN) technique. In this pro-
cedure one or more measured portions
of each of a stipulated series of de-
creasing sample volumes is inoculated
into the first-stage culture medium.
Through decreasing the sample incre-
ments, eventually a volume is reached
where only one cell is introduced into
some tubes, and no cells are introduced
into other tubes. Each of the several
tubes of sample-inoculated first-stage
medium is tested independently,
according to the principles previously
described, in the qualitative aspects
of testing procedures.
The combination of positive and
negative results is used in an application
of probability mathematics to secure
a single MPN value for the sample.
To obtain MPN values, the following
conditions must be met:
a The testing procedure must result
in one or more tubes in which the
test organism _is demonstrated to
be present; and
b The testing procedure must result
in one or more tubes in which the
test organism is not demonstrated
to be present.
The MPN value for a given sample is
obtained through the use of MPN Tables.
It is emphasized that the precision of
an individual MPN value is not great
when compared with most physical or
chemical determinations.
Standard practice in water pollution
surveys conducted by this organization,
is to plant five tubes in each of a series
of sample increments, in sample
volumes decreasing at decimal intervals.
For example, in testing known polluted
waters, the initial sample inoculations
might consist of 5 tubes each in volumes
of 0. 1, 0.01,0.001, and 0.0001 ml,
respectively. This series of sample
volumes will yield determinate results
from a low of 200 to a high of 1, 600, 000
organisms per 100 ml.
3-3
-------
MPN Methods
IE LABORATORY BENCH RECORDS IV
A Features of a Good Bench Record Sheet
A
1 Provides complete identification of the
sample.
2 Provides for full, day-by-day informa-
tion about all tests performed on the
sample.
3 Provides easy step-by-step record
applicable to any portion of the sample.
4 Provides for recording of the quantitative B
result which will be transcribed to sub-
sequent reports.
5 Minimizes the amount of writing by the
analyst.
t> Identifies the analyst(s).
B There is no such thing as "standard"
bench sheet for multiple tube tests; there
are many versions of bench sheets. Some
are prescribed by administrative authority
(such as the Office of a State Sanitary
Engineer); others are devised by laboratory
or project personnel to meet specific needs.
It is not the purpose of this discussion to
recommend an "ideal" bench form; however,
the form used in this training course
manual is essentially similar to that used
in certain research laboratories of this
organization. The student enrolled in the
course for which this manual is written
should make himself thoroughly familiar
with the bench sheet and its proper use.
See Figure 1.
NOTES ABOUT WORKING PROCEDURES
IN THE LABORATORY
Each bacteriological examination of water
by multiple dilution tube methods requires
a considerable amount of manipulation;
much is quite repetitious. Laboratory
workers must develop and maintain good
routine working habits, with constant
alertness to guard against lapses into
careless, slip-shod laboratory procedures
and "short cuts" which only can lead to
lowered quality of laboratory work.
Specific attention is brought to the following
by no means exhaustive, critical aspects of
laboratory procedures in multiple dilution
tube tests:
1 Original sample
a Follow prescribed care and handling
procedures before testing.
b Maintain absolute identification of
sample at all stages in testing.
c Vigorously shake samples (and
sample dilutions) before planting
in culture media.
2 Sample measurement into primary
culture medium
a Sample portions must be measured
accurately into the culture medium
for reliable quantitative tests to be
made. Standard Methods prescribe^
that calibration errors should not
exceed + 2.5%.
3-4
-------
BACTERIOLOGY BENCH SHEET
Project
- J7
V£AJ -J&4ASl4t£.
Multiple Dilution Tube Tests
Sample Station
Collection Data
Date
T
Analytical Record
Temperature f
Other Observations
Time f.\
f°C
Bench Number of Sample
Analyst jffn^!- fa
Test started at ff/:
Coliform MPN/100 ml
Confirmed:
Completed:
Fecal Coliform MPN:
Figure 1. SAMPLE BENCH SHEET
Fecal Streptococcus MPN/100 ml
A-D - EVA:
3-5
-------
MPN Methods
Suggested sample measuring practices
are as follows: Mohr measuring
pipets are recommended. 10 ml
samples are delivered at the top of
the culture tube, using 10 ml pipets.
1. 0 ml samples are delivered down
into the culture tube, near the sur-
face of the medium, and "touched
off" at the side of the tube when the
desired amount of sample has been
delivered. 1. 0 ml or 2. 0 ml pipets
are used for measurement of this
volume. 0.1 ml samples are
delivered in the same manner as 1.0
ml samples, using great care that
the sample actually gets into the
culture medium. Only 1.0 ml pipete
are used xor this sample volume.
After delivery of all sample incre-
ments into the culture tubes, the
entire rack of culture tubes may be
shaken gently to carry down any of
the sample adhering to the wall of
the tube above the medium.
Workers should demonstrate by actual
tests that the pipets and the technique
in use actually deliversthe ratedvolumes
within the prescribed limits of error.
Volumes as small as 0.1 ml routinely
can be delivered directly from the
sample with suitable pipets. Lesser
sample volumes first should be diluted,
with subsequent delivery of suitable
volumes of diluted sample into the
culture medium. A diagrammatic
sqheme for making dilutions is shown
in Figure 2.
Reading of culture tubes for gas
production
a On removal from the incubator,
snake culture rack gently, to
encourage release of gas which
may be supersaturated in the culture
medium.
b Gas in any quantity is a positive test.
It is necessary to work in conditions
of suitable lighting for easy recog-
nition of the extremely small amounts
of gas inside the tops of some
fermentation vials.
: Reading of liquid culture tubes for
growth as indication of a positive test
requires good lighting. Growth is
shown by any amount of increased
turbidity or opalescence in the culture
medium, with or without deposit of
sediment at the bottom of the tube.
Transfer of cultures with inoculation
loops and needlgg__.-
a Always sterilize inoculation loops
and needles to glowing (white hot)
in flame immediately before transfer
of culture; do not lay it down or
touch it to any non sterile object
before making the transfer.
b After sterilization, allow sufficient
time for cooling, in the air, to avoid
heat-killing bacterial cells which will
be gathered on the wire.
c Loops should be at least 3mm in inside
diameter, with a capability of holding
a drop of water or culture.
For routine standard transfers
requiring transfer of 3 loopsful of
culture, (Fecal Steptococci) many
workers form three- vi-mm loops on the
sance length of wire.
As an alternative to use of standard
inoculation loops, the use of
"applicator sticks" is described in the
14th Edition of Standard Methods.
3-6
-------
MPN Methods
Figure 2. PREPARATION OF DILUTIONS
Dilution Ratios:
1:100
1:10000
Water
Sample
. J ml.
Delivery volume
1ml
1ml O.lml 1ml O.lml
Tubes
Petri Dishes or Culture Tubes
Actual volume
of sample in tube
1ml
0. 1 ml 0. 01 ml 0. 001ml
0. 0001 ml 0. 00001 ml
The applicator sticks are dry heat
sterilized (autoclave sterilization is
not acceptable because of possible
release of phenols if the wood is
steamed) and are used on a single-
service basis. Thus, for every
positive culture tube transferred, a
new applicator stick is used.
This use of applicator sticks is
particularly attractive in field
situations where it is inconvenient or
impossible to provide a gas burner
suitable for sterilization of the
inoculation loop. In addition, use of
applicator sticks is favored in
laboratories where room temperatures
are significantly elevated by use of
gas burners.
7 Streaking cultures on agar surfaces
a All streak-inoculations should be
made without breaking the surface
of t'he agar. Learn to use a light
touch with the needle; however,
many inoculation needles are so
sharp that they are virtually useless
in this respect. When the needle is
platinum or platinum-iridium wire,
it sometimes is beneficial to fuse
the working tip into a small sphere.
This can be done by momentary
insertion of a well-insulated (against
electricity) wire into a carbon arc,
or some other extremely hot environ-
ment. The sphere should not be more
than twice the diameter of the wire
from which it is formed, otherwise
it will be entirely too heat-retentive
to be useful.
3-7
-------
MPN Methods
When the needle is nichrome
resistance wire, it cannot be heat-
fused; the writer prefers to bend
the terminal 1/16 - 1/8" of the wire
at a slight angle to the overall axis
of the needle. The side of the
terminal bent portion of the needle
then is used for inoculation of agar
surfaces.
b When streaking for colony isolation,
avoid using too much inoculum. The
streaking pattern is somewhat
variable according to individual
preference. The procedure favored
by the writer is shown in the
accompanying figure. Note
particularly that when going from
any one stage of the streaking to the
next, the inoculation needle is heat-
sterilized.
Preparation of cultures for Gram
stain
a The Gram stain always should be
made from a culture grown on a
nutrient agar surface (nutrient agar
slants are used here) or from nutrient
broth.
The culture should be young, and
should be actively growing. Many
workers doubt the validity of the
Gram stain made on a culture more
than 24 hours old.
Prepare a thin smear for the staining
procedure. Most beginning workers
tend to use too much bacterial sus-
pension in preparing the dried smear
for staining. The amount of bacteria
should be so small that the dried film
is barely visible to the naked eye.
V EQUIPMENT AND SUPPLIES
Consolidated lists of equipment, supplies,
and culture media required for all multiple
dilution tube tests described in this outline
are shown in Table 2.. Quantitative infor-
mation is not presented; this is variable -
according to the extent of the testing pro-
cedure, the number of dilutions used, and
the number of replicate tubes per dilution.
It is noted that requirements for alternate
procedures are fully listed and choices are
made in accordance to laboratory preference.
3-1
-------
MPN Methods
a Flame-sterilize an inoculation needle and air-cool,
b Dip the tip of the inoculation needle into the bac-
terial culture being studied.
c Streak the inoculation needle tip lightly back and
forth over half the agar surface, as in ( 1), avoid-
ing scratching or breaking the agar .surface.
d Flame-sterilize the inoculation needle _md air-cool.
a Turn the Petri dish one-quarter-turn and streak the
inoculation needle tip lightly back and forth over one-
half the agar surface, working from area (1) into one-
half the unstreaked area of the agar.
b Flame-sterilize the inoculation needle and air-cool.
a Turn the Petri dish one-quarter-turn and streak the
inoculation needle tip lightly back and forth over one-
half the agar surface, working from area (2) into
area (3), the remaining unstreaked area.
b Flame-sterilize the inoculation needle and set it aside.
c Close the culture container and incubate as prescribed.
Figure 3. A SUGGESTED PROCEDURE FOR COLONY TSOl ATION BY A
STREAK-PLATE TECHNIQUE
AREA } (Heavy inoculum)
AREA 3 (Isolated colonies)
AREA 2
(Moderate growth)
APPEARANCE OF STREAK PLATE
AFTER INCUBATION INTERVAL
3-9
-------
MPN Methods
TABLE 2. APPARATUS AND SUPPLIES FOR STANDARD
FERMENTATION TUBE TESTS
Description of Item
Lauryl tryptose broth or Lactose
broth. 20 ml amounts of 1. 5 X
concentration medium, in 25 X 150 mm
culture tubes with inverted fermen-
tation vials, suitable caps.
Lauryl tryptose broth or Lactose
Total Coliform Group
Presumptive j Confirmed
Test ; Test
X
X
broth. 10 ml amounts of single
strength medium in 20 X 150 mm
culture tubes with inverted fermen-
tation vials, suitable- caps.
Brilliant green lactose bile broth, 2%
in 10 ml amounts, single strength,
in 20 X 150 mm culture tubes with
inverted fermentation vials.
suitable caps.
Eosin methylene blue agar, poured
in 100 X 15 mm Petri dishes
Endo Agar, poured in 100 X 15 mm
dishes
Nutrient agar slant, screw cap tube
EC Broth, 10 ml amounts of single
strength medium in fermentation
tubes '.
Culture tube racks, 10X5 openings;
each opening to accept 25 mm dia-
meter tubes.
Pipettes, 10 ml, .Mohr type, sterile,
in suitable cans.
Pipettes, 2 ml (optional), Morh type.
sterile, in suitable cans
Pipettes, 1 ml, Mohr type, sterile
in metal suitable cans
Standard buffered dilution water.
sterile, 99-ml amounts in screw-
capped bottles.
Gas burner, Bunsen type
Inoculation loop, loop 3mm dia-
meter, of nichrome or platinum-
iridium wire, 26 B &. S gauge, in
suitable holder, (or sterile applicator'
stick)
Inoculation needle, nichrome, or
platinum-iridium wire, 26 B & S
gauge, in suitable holder.
Incubator, adjusted to 35 + 0. 50 c
Waterbath incubator, adjusted to
44.5+ 0.2°C.
Glass microscopic slides, l" X3"
Slide racks (optional)
Gram-stain solutions, complete set
Compound microscope, oil immer-
sion lens, Abbe' condenser
Basket for discarded cultures
Container for discarded pipettes
1
1
X
1
i x
X
1
X
Completed
Test
Fecal Coliforms
(EC Iproth)
X
\
X
X
X
1
1
X ' X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
J£
X
X
X
X
X
X
i
X
X
X
X
X
3-10
-------
Part 2
DETAILED TESTING PROCEDURES FOR MEMBERS OF THE
COLIFORM GROUP BY MULTIPLE DILUTION TUBE METHODS
I SCOPE
A Tests Described
1 Presumptive Test
2 Confirmed Test
3 Completed Test
4 Fecal Coliform Test
B Form of Presentation
The Presumptive, Confirmed, and
Completed Tests are presented as total,
independent procedures. It is recognized
that this form of presentation is somewhat
repetitious, inasmuch as the Presumptive
Test is preliminary to the Confirmed
Test, and both the Presumptive Test and
the Confirmed Test are preliminary to the
Completed Test for total coliforms.
In using these procedures, the worker
must know at the outset what is to be the
stage at which the test is to be ended, and
the details of the procedures throughout,
in order to prevent the possibility of
discarding gas-positive tubes before
proper transfer procedures have been
followed.
Thus, if the worker knows that the test will
be ended at the Confirmed Test, he will
turn at once to Section III, TESTING TO
THE CONFIRMED TEST STAGE, and will
ignore Sections II and IV.
The Fecal Coliform Test is described
separately, in Section V, as an
adjunct to the Confirmed Test and to the
Completed Test.
II TESTING TO PRESUMPTIVE TEST
STAGE
A First-Day Procedures
1 Prepare a laboratory data sheet for
the sample. Record the following
information: assigned laboratory
number, source of sample, date and
time of collection, temperature of the
source, name of sample collector,
date and time of receipt of sample in
the laboratory. Also show the date
and time of starting tests in the
laboratory, name(s) of worker(s) per-
forming the laboratory tests, and the
sample volumes planted.
2 Label the tubes of lauryl tryptose broth
required for the initial planting of the
sample (Table 3). The label should
bear three identifying marks. The
upper number is the identification of
the worker(s) performing the test
(applicable to personnel in training
courses), the number immediately
below is the assigned laboratory num-
ber, corresponding with the laboratory
record sheet. The lower number is the
code to designate the sample volume
and which tube of a replicate series is
indicated.
NOTE: Be sure to use tubes containing
the correct concentrations of culture medium
for the inoculum/tube volumes. (See the
chapter on media and solutions for multiple
dilution tube methods or refer to the current
edition of Standard Methods for Water and
Wastewater).
3-11
-------
MPN Methods
Table 3. SUGGESTED LABELING SCHEME FOR ORIGINAL CULTURES AND
SUBCULTURES IN MULTIPLE DILUTION TUBE TESTS
Bench number
Volume & tube
Bench number
Volume & tube
Bench number
Volume & tube
Bench number
Volume & tube
Bench number
Volume & tube
Tube
1
312
A
312
a
312
a
312
la
312
2a
Tube
2
312
B
312
b
312
b
312
Ib
312
2b
Tube
3
312
C
312
c
312
c
312
Ic
312
2c
Tube
4
312
D
312
d
312
d
312
Id
312
2d
Tube
5
312
E
312
e
312
e
312
le
312
2e
Sample volume
represented
Tubes with 10 ml
of sample
Tubes with 1 ml
of sample
Tubes with 0. 1 ml
of sample
Tubes with 0.01 ml
of sample
Tubes with 0. 001 ml
of sample
Typical Example
RB
312-
A .
101
Lab. Worker
Identification
-Bench Number
"Sample Volume
Tube of Culture Medium
The labeling of cultures can be reduced by labeling only the first tube of
each series of identical sample volumes in the initial planting of the sample.
All subcultures from initial plantings should be labeled completely.
Place the labeled culture tubes in an
orderly arrangement in a culture tube
rack, with the tubes intended for the
largest sample volumes in the front
row, and those intended for smaller
volumes in the succeeding rows.
Shake the sample vigorously, approxi-
mately 25 times, in an arc of one foot
within seven seconds and withdraw the
sample portion at once.
Measure the predetermined sample
volumes into the labeled tubes of lauryl
tryptose broth, using care to avoid
introduction of any bacteria into the
culture medium except those in the
sample.
a Use a 10 ml pipet for 10 ml sample
portions, and 1 ml pipets for portions
of 1 ml or less. Handle sterile pipets
only near the mouthpiece, and protect
the delivery end from external con-
tamination. Do not remove the cotton
plug in the mouthpiece as this is
intended to protect the user from
ingesting any sample.
b When using the pipet to withdraw
sample portions, do not dip the
pipet more than 1/2 inch into the
sample; otherwise sample running
down the outside of the pipet will
make measurements inaccurate.
6 After measuring all portions of the
sample into their respective tubes of
medium, gently shake the rack of
inoculated tubes to insure good mixing
of sample with the culture medium.
Avoid vigorous shaking, as air bubbles
may be shaken into the fermentation
vials and thereby invalidate the test.
7 Place the rack of inoculated tubes in the
• incubator at 35° + 0.5OC for 24 +
2 hours.
B 24-hour Procedures
1 Remove the rack of lauryl tryptose
broth cultures from the incubator, and
shake gently. If gas is about to appear
in the fermentation vials, the shaking
will speed the process.
3-12
-------
MPN Methods
2 Examine each tube carefully. Record,
in the column "24" under LST on the
laboratory data sheet, each tube showing
gas in the fermentation vial as a positive
(+) test and each tube not showing gas
as a negative (-) test. GAS IN ANY 2
QUANTITY IS A POSITIVE TEST.
3 Discard all gas-positive tubes of lauryl
tryptose broth, and return all the gas-
negative tubes to the 35°C incubator for
an additional 24 i 2 hours.
C 48-hour Procedures
1 Remove the rack of culture tubes from
the incubator, read and record gas
production for each tube.
2 Be sure to record all results under the
48-hour LTB column on the data sheet.
Discard all tubes. The Presumptive
Test is concluded at this point, and
Presumptive coliforms per 100 ml can
be computed according to the methods
described elsewhere in this manual.
Ill TESTING TO CONFIRMED TEST STAGE
Note that the description starts with the
sample inoculation and includes the
Presumptive Test stage. The Confirmed
Test preferred in Laboratories of this agency
is accomplished by means of the brilliant
green lactose bile broth (BGLB) and the
acceptable alternate tests are mentioned in
III F. In addition, the Fecal Coliform Test is
included as an optional adjunct to the procedure.
A First-Day Procedures
1 Prepare a laboratory data sheet for the
sample. Record the following infor-
mation: assigned laboratory number,
source of sample, date and time of
collection, temperature of the source,
name of sample collector, date and
time of receipt of sample in the
laboratory. Also show the date and
time of starting tests in the laboratory.
name(s) of worker(s) performing the
laboratory tests, and the sample
volumes planted.
Label the tubes of lauryl tryptose broth
required for the initial planting of the
sample. The label should bear three
identifying marks. The upper number
is the identification of the worker(s)
performing the test (applicable to
personnel in training courses), the
number immediately below is the
assigned laboratory number, corres-
ponding with the laboratory record
sheet. The lower number is the code
to designate the sample volume and
which tube of a replicate series is indicated.
NOTE: If 10-ml samples are being
planted, it is necessary to use tubes
containing the correct concentration
of culture medium.- This has previously
been noted in II A-2.
Place the labeled culture tubes in an
orderly arrangement in a culture tube
rack, with the tubes intended for the
largest sample volumes in the front
row, and those intended for smaller
volumes in the succeeding rows.
Shake the sample vigorously, approxi-
mately 25 times, in an up-and-down
motion.
Measure the predetermined sample
volumes into the labeled tubes of lauryl
tryptose broth, using care to avoid
introduction of any bacteria into the
culture medium except those in the sample.
a Use a 10-ml pipet for 10 ml sample
portions, and 1-ml pipets for portions
of 1 ml or less. Handle sterile pipets
only near the mouthpiece, and protect
the delivery end from external con-
tamination. Do not remove the cotton
plug in the mouthpiece as this is intended
to protect the user from ingesting any
sample.
3-13
-------
MPN Methods
b When using the pipet to withdraw
sample portions, do not dip the
pipet more than 1/2 inch into the
sample; otherwise sample running
down the outside of the pipet will
make measurements inaccurate.
c When delivering the sample into the
culture medium, deliver sample
portions of 1 ml or less down into
the culture tube near the surface of
the medium. Do not deliver small
sample volumes at the top of the tube
and allow them to run down inside
the tube; too much of the sample
will fail to reach the culture medium.
d Prepare preliminary dilutions of
samples for portions of 0. 01 ml or
less before delivery into the culture
medium. See Table 1 for preparation
of dilutions. NOTE: Always deliver
diluted sample portions into the
culture medium as soon as possible
after preparation. The interval
between preparation of dilution and
introduction of sample into the
medium never should be as much
as 30 minutes.
6 After measuring all portions of the
sample into their respective tubes of
medium, gently shake the rack of
inoculated tubes to insure good mixing
of sample with the culture medium.
Avoid vigorous shaking, as air bubbles
may be shaken into the fermentation
vials and thereby invalidate the test.
7 Place the rack of inoculated tubes in
the incubator at 35° + 0. 5° C for 24 +
2 hours.
B 24-hour Procedures
1 Remove the rack of lauryl tryptose
broth cultures from the incubator, and
shake gently. If gas is about to appear
in the fermentation vials, the shaking
will speed the process.
Examine each tube carefully. Record,
in the column "24" under LST on the
laboratory data sheet, each tube showing
gas in the fermentation vial as a
positive (+) test and each tube not
showing gas as a negative (-) test.
GAS IN ANY QUANTITY IS A POSITIVE
TEST.
Retain all gas-positive tubes of lauryl
tryptose broth culture in their place
in the rack, and proceed.
Select the gas-positive tubes of lauryl
tryptose broth culture for Confirmed
Test procedures. Confirmed Test
procedures may not be required for all
gas-positive cultures. If, after 24-hours
of incubation, all five replicate cultures
are gas-positive for two or more con-
secutive sample volumes, then select
the set of five cultures representing
the smallest volume of sample in which
all tubes were gas-positive. Apply
Confirmed Tast procedures to all these
cultures and to any other gas-positive
cultures representing smaller volumes
of sample, in which some tubes were
gas-positive and some were gas-negative.
Label one tube of brilliant green lactose
bile broth (BGLB) to correspond with
each tube of lauryl tryptose broth
selected for Confirmed Test procedures.
Gently shake the rack of Presumptive
Test cultures. With a flame-sterilized
inoculation loop transfer one loopful of
culture from each gas-positive tube to
the corresponding tube of BGLB. Place
each newly inoculated culture into BGLB
in the position of the original gas-positive
tube.
After making the transfers, the rack
should contain some 24-hour gas-
negative tubes of lauryl tryptose broth
and the newly inoculated BGLB.
If the Fecal Coliform Test is included
in the testing procedures, consult
Section V of this part of the outline of
testing procedures.
3-14
-------
MPN Methods^
9 Incubate the 24-hour gas-negative
BGLB tubes and any newly-inoculated
tubes of BGLB an additional 24 + 2
hours at 35° + 0. 5°C.
C 48-hour Procedures
1 Remove the rack of culture tubes from
the incubator, read and record gas
production for each tube.
2 Some tubes will be lauryl tryptose broth
and some will be brilliant green lactose
bile broth (BGLB). Be sure to record
results from LTB under the 48-hour
LTB column and the BGLB results under
the 24-hour column of the data sheet.
3 Label tubes of BGLB to-correspond with
all (if any) 48-hour gas-positive cultures
in lauryl tryptose broth. Transfer one
loopful of culture from each gas-positive
LTB culture to the correspondingly-
labeled tube of BGLB. NOTE: All
tubes of LTB culture which were
negative at 24 hours and became
positive at 48 hours are to be transferred
The option described above for 24-hour
cultures does not apply at 48 hours.
4 If the Fecal Coliform Test is included
in the testing procedure, consult
Section V of the part of the outline
of testing procedures.
5 Incubate the 24-hour gas-negative
BGLB tubes and any newly-inoculated
tubes of BGLB 24 + 2 hours at 35O +
0.50C.
6 Discard all tubes of LTB and all 24-hour
gas-positive BGLB cultures.
D 72-hour Procedures
1 If any cultures remain to be examined,
all will he BGLB. Some may be 24
hours old and some may be 48 hours
old. Remove such cultures from the
incubator, examine each tube for gas
production, and record results on the
data sheet.
2 Be sure to record the results of 24-hour
BGLB cultures in the "24" column under
BGLB and the 48-hour results under the
"48" column of the data sheet.
3 Return any 24-hour gas-negative cultures
for incubation 24 + 2 hours at 35 +
0.5°C.
4 Discard all gas-positive BGLB cultures
and all 48-hour gas-negative cultures
from BGLB.
5 It is possible that all cultural work and
results for the Confirmed Test have
been finished at this point. If so, codify
results and determine Confirmed Test
coliforms per 100 ml as described in
the outline on use of MPN Tables.
E 96-hour Procedures
At most only a few 48-hour cultures in
BGLB may be present. Read and record
gas production of such cultures in the "48"
column under BGLB on the data sheet.
Codify results and determine Confirmed
Test coliforms per 100 ml.
F Streak-plate methods for the Confirmed
Test, using eosin methylene blue agar or
Endo agar plates, are accepted procedures
in Standard Methods. The worker who
prefers to use one of these media in
preference to BGLB (also approved in
Standard Methods) is advised to refer to
the current edition of "Standard Methods^
for the Examination of Water and Waste-
water" for procedures.
3-15
-------
MFN Methods
IV TESTING TO COMPLETED TEST STAGE
(Note that this description starts with the
sample inoculation and proceeds through the
Presumptive and the Confirmed Test stages.
In addition, the Fecal Coliform Test is
referred to as an optional adjunct to the
procedure.)
A First-Day Procedures
1 Prepare a laboratory data sheet for the
sample. Record the following information;
assigned laboratory number, source of
sample, date and time of collection,
temperature of the source, name of
sample collector, date and time of
receipt of sample in the laboratory.
Also show the date and time of starting
tests in the laboratory, name(s) of
worker(s) performing the laboratory
tests, and the sample volumes planted.
2 Label the tubes of lauryl tryptose broth
required for the initial planting of the
sample. The label should bear three
identifying marks. The upper number
is the identification of the worker(s)
performing the test (applicable to
personnel in training courses),
the number immediately below Is ths-
assigned laboratory number, corres-
ponding with the laboratory record
sheet. The lower number is the code
to designate the sample volume and
which tube of a replicate series is
indicated. Guidance on labeling for
laboratory data number and identification
of individual tubes is described else-
where in this outline.
NOTE: If 10-ml samples are being
plated, it is necessary to use tubes
containing the correct concentration
of rulture medium. This has previously
been noted elsewhere in this outline
and referral is made to tables.
3 Place the labeled culture tubes in an
orderly arrangement in a culture tube
rack, with the tubes intended for the
largest sample volumes in the front
row, and those intended for smaller
volumes in the succeeding rows.
4 Shake the sample vigorously, approxi-
mately 25 times, in an up-and-down
motion.
5 Measure the predetermined sample
volumes into the labeled tubes of lauryl
tryptose broth, using care to avoid
introduction of any bacteria into the
culture medium except those in the
sample.
a Use a 10-ml pipet for 10 ml sample
portions, and 1-ml pipets for portions
of 1 ml or less. Handle sterile
pipets only near the mouthpiece,
and protect the delivery end from
external contamination. Do not move
the cotton plug in the mouthpiece
-s this is intended to protect the
user from ingesting any sample.
When using the pipet to withdraw
sample portions, do not dip the
pipet more than 1/2 inch into the
sample; otherwise sample running
down the outside of the pipet will
make measurements inaccurate.
When delivering the sample into the
culture medium, deliver sample
portions of 1 ml or less down into
3-16
-------
MPN Methods
the culture tube near the surface of
the medium. Do not deliver small
sample volumes at the top of the
tube and allow them to run down
inside the tube; too much of the
sample will fail to reach the culture
medium.
d Prepare preliminary dilutions of
samples for portions of 0. 01 ml or
less before delivery into the culture
medium. See-Table 2 for preparation
of dilutions. NOTE: Always deliver
diluted sample portions into the
culture medium as soon as possible
after preparation. The interval
between preparation of dilution and
introduction of sample into the
medium never should be as much as
30 minutes.
6 After measuring all portions of the
sample into their respective tubes of
medium, gently shake the rack of
inoculated tubes to insure good mixing
of sample with the culture medium.
Avoid vigorous shaking, as air bubbles
may be shaken into the fermentation
vials and thereby invalidate the test.
7 Place the rack of inoculated tubes in
the incubator at 35° + o. 5OC for 24 +
2 hours.
B 24-hour Procedures
1 Remove the rack of lauryl tryptose broth
cultures from the incubator, and shake
gently. If gas is about to appear in the
fermentation vials, the shaking will
speed the process.
2 Examine each tube carefully. Record,
in the column "24" under LST on the
laboratory data sheet, each tube showing
gas in the fermentation vial as a positive
(+) test and each tube not showing gas
as a negative (-) test. GAS IN ANY
QUANTITY IS A POSITIVE TEST.
3 Retain all gas-positive tubes of lauryl
tryptose broth culture in their place in
the rack, and proceed.
4 Select the gas-positive tubes of lauryl
tryptose broth culture for the Confirmed
Test procedures. Confirmed Test
procedures may not be required for
all gas-positive cultures. If, after
24-hours of incubation, all five
replicate cultures are gas-positive for
two or more consecutive sample
volumes, then select the set of five
cultures representing the smallest
volume of sample in which all tubes
were gas-positive. Apply Confirmed
Test procedures to all these cultures
and to any other gas-positive cultures
representing smaller volumes of
sample, in which some tubes were
gas-positive and some were gas-
negative .
5 Label one tube of brilliant green lactose
bile broth (BGLB) to correspond with
each tube of lauryl tryptose broth
selected for Confirmed Test procedures.
6 Gently shake the rack of Presumptive
Test cultures. With a flame-sterilized
inoculation loop transfer one loopful of
culture from each gas-positive tube to
the corresponding tube of BGLB. Place
each newly inoculated culture into
BGLB in the position of the original
gas-positive tube.
7 If the Fecal Coliform Test is included
in the testing procedure, consult
Section V of this outline for details of
the testing procedure.
8 After making the transfer, the rack
should contain some 24-hour gas-
negative tubes of lauryl tryptose borth
and the newly inoculated BGLB.
Incubate the rack of cultures at 35° C
+ 0.50C for 24 + 2 hours.
C 48-hour Procedures
1 Remove the rack of culture tubes from
the incubator, read and record gas
production for each tube.
2 Some tubes will be lauryl tryptose broth
and some will be brilliant green lactose
3-17
-------
MPN Methods
bile broth (BGLB). Be sure to record
results from LTB under the 48-hour
LTB column and the BGLB results
under the 2 4-hour column of the data
sheet.
3 Label tubes of BGLB to correspond with
all (if any) 48-hour gas-positive cultures
in lauryl tryptose broth. Transfer one
loopful of culture from each gas-positive
LTB culture to the correspondingly-
labeled tube of BGLB. NOTE: All tubes
of LTB culture which were negative at
24 hours and became positive at 48 hours
are to be transferred. The Option
described above for 24-hour LTB
cultures does not apply at 48 hours.
4 Incubate the 24-hour gas-negative BGLB
tubes and any newly-inoculated tubes of
BGLB 24 + 2 hours at 35O + 0. 5OC.
Retain all 24-hour gas-positive cultures
in BGLB for further test procedures.
5 Label a Petri dish preparation of eosin
methylene blue agar (EMB agar) to
correspond with each gas-positive
culture in BGLB.
6 Prepare a streak plate for colony
isolation from each gas-positive culture
in BGLB on the correspondingly-labeled
EMB agar plate.
Incubate the EMB agar plates 24 + 2
hours at 35° + 0.5°C.
D 72-hour Procedures
1 Remove the cultures from the incubator.
Some may be on BGLB; several EMB
agar plates also can be expected.
2 Examine and record gas production
results for any cultures in BGLB.
3 Retain any gas-positive BGLB cultures
and prepare streak plate inoculations
for colony isolation in EMB agar.
Incubate the EMB agar plates 24 +
2 hours at 35 + 0. 50 C. Discard the
gas-positive BGLB cultures after
transfer.
4 Reincubate any gas-negative BGLB
cultures 24 + 2 hours at 35° + 0. 5OC.
5 Discard all 48-hour gas-negative BGLB
cultures.
6 Examine the EMB agar plates for the
type of colonies developed thereon.
Well-isolated colonies having a dark
center (when viewed from the lower
side, held toward a light) are termed
"nucleated or fisheye" colonies, and
are regarded as "typical" coliform
colonies. A surface sheen may or may
not be present on "typical" colonies.
Colonies which are pink or opaque but
are not nucleated are regarded as
"atypical colonies. " Other colony
types are considered "noncoliform. "
Read and record results as + for
"typical" (nucleated) colonies + for
"atypical" (non-nucleated pink or
opaque colonies), and - for other types
of colonies which might develop.
7 With plates bearing "typical" colonies,
select at least one well-isolated colony
and transfer it to a correspondingly-
labeled tube of lactose broth and to an
agar slant. As a second choice, select
at least two "atypical" colonies (if
typical colonies are not present) and
transfer them to labeled tubes of
lactose broth and to agar slants. As a
third choice, in the absence of typical
or atypical coliform-like colonies,
select at least two well-isolated
colonies representative of those
appearing on the EMB plate, and trans-
fer them to lactose broth and to agar
slants.
8 Incubate all cultures transfered from
EMB agar plates 24+2 hours at 35 +
0.50C.
E 96-hour Procedures
1 Subcultures from the samples being
studied may include: 48-hour tubes
of BGLB, EMB agar plates, lactose
broth tubes, and agar slant cultures.
3-18
-------
MPN Methods
If any 48-hour tubes of BGLB are
present, read and record gas production
in the "48" column under BGLB. From
any gas-positive BGLB cultures pre-
pare streak plate inoculations for colony
isolation on EMB agar. Discard all
tubes of BGLB, and incubate EMB agar
plates 24 + 2 hours at 35 + 0. 5° C.
If any EMB plates are present, examine
and record results in the "EMB" column
of the data sheet. Make transfers to
agar slants and to lactose broth from
all EMB agar plate cultures. In
decreasing order of preference, transfer
at least one typical colony, or at least
two atypical colonies, or at least two
colonies representative of those on the
plate.
Examine and record results from the
lactose broth cultures.
Prepare a Gram-stained smear from
each of the agar slant cultures, as
follows:
NOTE: Always prepare Gram stain
from an actively growing culture,
preferably about 18 hours old, and
never more than 24 hours old. Failure
to observe this precaution often results
in irregular staining reactions.
a Thoroughly clean a glass slide to
free it of any trace of oily film.
b Place one drop of distilled water on
the slide.
c Use the inoculation needle to suspend
a tiny amount of growth from the
nutrient agar slant culture in the
drop of water.
d Mix the thin suspension of cells with
the tip of the inoculation needle, and
allow the water to evaporate.
e "Fix" the smear by gently warming
the slide over a flame.
g Flush the excess dye solution
off in gently running water.
h Flood the smear with Lugol's
iodine for 1 minute.
i Wash the slide in gently running
water.
j Decolorize the smear with acetone
alcohol solution with gentle agitation
for 10-30 seconds, depending upon
extent of removal of crystal violet dye.
k Counterstain for 10 seconds with
safranin solution, then wash in running
water and gently blot dry with bibulous
paper.
1 Examine the slide under the microscope,
using the oil immersion lens. Coliform
bacteria are Gram-negative (pink to red
color and nonspore-forming, rod-shaped
cells, occurring singly, in pairs, or
rarely in short chains.
m If typical coliform staining reaction and
morphology are observed, record + in
in the appropriate space under the "Gram
Stain" column of the data sheet. If typical
morphology and staining reaction are not
observed, then mark it + or -, and make
suitable comment in the "remarks" column
at the right-hand side of the data sheet.
n If spore-forming bacteria are observed, it
will be necessary to repurify the culture
from which the observations were made.
Consult the instructor, or refer to Standard
Methods, for procedures.
At this point, it is possible that all cultural
work for the Completed Test has been finished.
If so, codify results and determine Completed
Test coliforms per 100 ml.
f Stain the smear by flooding it for 1
minute with ammonium oxylate-crystal
violet solution.
3-19
-------
MPN Methods
F 120-hour Procedures and following:
1 Any procedures to be undertaken from
this point are "straggler" cultures on
media already described, and requiring
step-by-step methodology already given
in detail. Such cultures may be on:
EMB plates, agar slants, or lactose
broth. The same time-and-temperature
of incubation required for earlier studies
applies to the "stragglers" as do the
observations, staining reactions, and
interpretation of results. On con-
clusion of all cultural procedures,
codify results and determine Completed
Test coliforms per 100 ml.
V FECAL COLIFORM TEST
A General Information
1 The procedure described is an elevated
temperature test for fecal coliform
bacteria.
2 Equipment required for the tests are
those required for the Presumptive
Test of Standard Methods, a water-bath
incubator, and the appropriate culture
media.
B Fecal Coliform Test with EC Broth
1 Sample; The test is applied to gas-
positive tubes from the Standard
Methods Presumptive Test (lauryl
tryptose broth), in parallel with
Confirmed Test procedures.
2 24-hour Operations. Initial procedures
are the planting procedures described
for the Standard Methods Presumptive
Coliform test.
a After reading and recording gas-
production on lauryl tryptose broth,
temporarily retain all gas-positive
tubes.
b Label a tube of EC broth to corre-
spond with each gas-positive tube
of lauryl tryptose broth. The option
regarding transfer of only a limited
number.-of tubes to the Confirmed
Test sometimes can be applied here.
However, the worker is urged to
avoid exercise of this option until
he has assured the applicability of
the option by preliminary tests on
the sample source.
c Transfer one loopful of culture from
each gas-positive culture in lauryl
tryptose broth to the correspondingly
labeled tube of EC broth.
d Incubate EC broth tubes 24 i 2 hours
at 44. 5 .£ 0. 2°C in a waterbath
with water depth sufficient to come
up at least as high as the top of the
culture medium in the tubes. Place
in waterbath as soon as possible
after inoculation and always within
30 minutes after inoculation.
3 48-hour operations
a Remove the rack of EC cultures
from the waterbath, shake gently,
and record gas production for each
tube. Gas in any quantity is a
positive test.
b As soon as results are recorded,
discard all tubes. (This is a 24-
hour test for EC broth inoculations
and not a 48-hour test.)
c Transfer any additional 48-hour
gas-positive tubes of lauryl tryptose
broth to correspondingly labeled
tubes of EC broth. Incubate 24 +
2 hours at 44. 5 i 0.2°C.
4 72-hour operations
a Read and record gas production for
each tube. Discard all cultures.
b Codify results and determine fecal
coliform count per 100 ml of sample.
3-20
-------
Examination of Water for Coliform and Fecal Streptococcus Groups
TESTS FOR COLIFORM GROUP
LACTOSE OR LAURYL TRYPTOSE BROTH
FERMENTATION TUBES (SERIAL DILUTION)
GAS POSITIVE GAS NEGATIVE
COLIFORM CROUP COLIFORM CROUP
CONFIRMED NOT CONFIRMED
LACTOSE & LAURYL TRYPTOSE
BROTH ARE INTERCHANCEABLE MEDIA
AND ARE INCUBATED AT 35 DEC C +
OS DEO C.
GAS POSITIVES TUBES {ANY AMOUNT
OF GAS) CONSTITUTE A POSITIVE
PRESUMPTIVE TEST
TOTAL INCUBATION TIME FOR LACTOSE OR
LIB IS 48 MRS t 3 HRS
INCUBATE BGLB TUBES FOR 48 HRS
1 3 HRS. AT 35 DEG. C± 0.5 DEG. C.
INCUBATE EMB OR ENDO AGAR
PLATES FOR 24 HRS. ± 2 HRS. AT
35 DEG. C± 0.5 DEG. C.
CAS NEGATIVE
COLIFORM CROUP
ABSENT
TRANSFER TO EMB PLATE
AND REPEAT PROCESS
COLIFORM GROUP ABSENT
3-21
-------
THIS PAGE INTENTIONALLY
BLANK
-------
Part 3
LABORATORY METHODS FOR FECAL STREPTOCOCCUS
(Day-By-Day Procedures)
I GENERA L INFORMA TION
A The same sampling and holding procedures
apply as for the coliform test.
B The number of fecal streptococci in water
generally is lower than the number of
coliform bacteria. It is good practice
in multiple dilution tube tests to start the
sample planting series with one sample
increment larger than for the coliform
test. For example: If a sample planting
series of 1.0, 0.1, 0.01, and 0.001 ml
is planned for the coliform test, it is
suggested that a series of 10, 1.0, 0. 1,
and 0.01 ml be planted for the fecal
streptococcus test.
C Equipment required for the test is the same
as required for the Standard Methods
Presumptive and Confirmed Tests, except
for the differences in culture media.
H STANDARD METHODS (Tentative)
PROCEDURES
A First-Day Operations
1 Prepare the sample data sheet and
labeled tubes of azide dextrose broth
in the same manner as for the
Presumptive Test. NOTE: If 10-ml
samples are included in the series, be
sure to use a special concentration
(ordinarily double-strength) of azide
dextrose broth for these sample
portions.
2 Shake the sample vigorously, approxi-
mately 25 times, in an up-and-down
motion.
3 Measure the predetermined sample
volumes into the labeled tubes of azide
dextrose broth, using the sample
measurement and delivery techniques
used for the Presumptive Test.
4 Shake the rack of tubes of inoculated
culture media, to insure good mixing
of sample with medium.
5 Place the rack of inoculated tubes in
the incubator at 35° + 0. 50 C for 24 +
2 hours.
B 2 4-hour Operations
1 Remove the rack of tubes from the
incubator. Read and record the results
from each tube. Growth is a positive
test with this test. Evidence of growth
consists either of turbidity of the
medium, a "button" of sediment at the
bottom of the culture tube, or both.
2 Label a tube of ethyl violet azide broth
to correspond with each positive culture
of azide dextrose broth. It may be
permissible to use the same confirmatory
transfer option as described for the
coliform Confirmed Test, in this outline.
3 Shake the rack of cultures gently, to
resuspend cells which have settled
out to the bottom of the culture tubes.
4 Transfer three loopfuls or use a
wood applicator to transfer culture
from each growth-positive tube of
azide dextrose broth to the correspond-
ingly labeled tube of ethyl violet azide
broth.
5 As transfers are made, plsce the newly
inoculated tubes of ethyl violet azide
broth in a separate rack while returning
the AD tubes to their former positions
in the rack.
6 Return the rack, all azide dextrose
broth tubes and newly-inoculated tubes
of ethyl violet azide broth, to the in-
cubator. Incubate 24 _+ 2 hours at 35°
+ 0. 5°C. ~
3-23
-------
MPN Methods
C 48-hour Operations
1 Remove the rack of tubes from the
incubator. Read and report results.
Growth, either in azide dextrose broth
or in ethyl violet azide broth, is a
positive test. Be sure to report the
results of the azide dextrose broth
medium under the "48" column for that
medium and the results of the ethyl
violet azide broth cultures under the
"24" column for that medium.
2 Any 48-hour growth-positive cultures
of azide dextrose broth are to be
transferred (as before) to ethyl
violet azide broth. Discard all 48-hour
growth-negative tubes of azide dextrose
broth and all 24-hour growth-positive
tubes of ethyl violet azide broth.
3 Re-incubate the 24-hour growth-negative
Ethyl Violet azide tubes after again re-
inoculating with their respective positive
Azide Dextrose tubes and the newly-
inoculated tubes of ethyl violet azide
broth 24 -f 2 hours at 35° +_ 0. 5°C.
D 7 2-hour Operations
1 Read and report growth results of all
tubes of ethyl violet azide broth.
2 Discard all growth-positive cultures
and all 48-hour growth-negative
cultures.
3 Reincubate any 24-hour growth-negative
cultures in ethyl violet azide broth after
reinoculating with their respective
positive azide dextrose; tubes for an
additional 24 +_ 2 hours at 35° + 0. 5°C.
E 9 6-hour Operations
1 Read and report growth results of any
remaining tubes of ethyl violet azide
broth.
Codify results and determine fecal
streptococci per 100 ml.
REFERENCES
1 Standard Methods for the Examination of
Water and Wastewater (14th Ed).
Prepared and published jointly by
American Public Health Association,
American Water Works Association,
and Water Pollution Control
Federation. 1975.
\
2 Geldreich, E.E., Clark, H.F., Kabler,
P.W., Huff, C.B. and Bordner, R.H.
The Coliform Group. II. Reactions
in EC Medium at 45° C. Appl.
Microbiol. 8:347-348. 1958.
3 Geldreich, E. E., Bordner, R.H., Huff,
C.B., Clark, H.F. and Kabler, P.W.
Type Distribution of Coliform. Bacteria
in the Feces of Warm-Blooded Animals.
J. Water Pollution Control Federation.
34:295-301. 1962.
This outline was prepared by H. L.
Jeter, Chief, Program Support
Training Branch, USEPA, Cincinnati,
Ohio 45268
Descriptors: Coliforms, Fecal Coliforms
Fecal Streptococci, Indicator Bacteria,
Laboratory Equipment, Laboratory
Tests, Microbiology, Most Probable
Number, MPN, Sewage Bacteria,
Water Analysis
3-24
-------
MEDIA AND SOLUTIONS FOR MULTIPLE DILUTION TUBE METHODS
I
I INTRODUCTION
A This chapter is intended to present detailed
information on preparation and management
of media and solutions needed with the tests
and observations described elsewhere in
this course manual.
B The preparation and management of
supplies of culture media and solutions
is one of the most critical aspects of a
bacteriological water quality testing
program.
1 In the same manner that the chemist
relies on correctly prepared and
standardized reagents for his analytical
work, the bacteriologist must depend
on satisfactory culture media for the
type of analysis with which he is con-
cerned.
2 In many laboratories preparation of
media is entrusted to subprofessional
personnel. Most such personnel,
properly trained and guided, are able
to perform the required tasks efficiently
and reliably.
3 The professional supervisor should
maintain close attention to all details,
however, to guard against gradual
introduction of bad habits in preparing
and preserving media and other liquid
supplies.
II GENERA L INFORMA TION
A Use of Commercially Available
Dehydrated Media
1 The preparation of all media described
in this chapter is given in terms of the
individual components, and preparation
of the finished medium. This is done,
even through commercially available
dehydrated media are widely used, to
acquaint the worker with the compo-
sition of the media and to indicate the
required specifications of each medium.
2 The use of commercially available
dehydrated media, requiring only
careful weighing and dissolving of the
powder in the proper quantity and
quality of distilled water, is strongly
recommended. Such media are much
more likely to have uniformity at an
acceptably high level of quality than
are media compounded in the laboratory
from the individual constituents.
3 It is recommended that the worker,
when using commercially prepared
dehydrated media, keep a careful
record of the lot numbers of media
being used. With first use of each
new lot number of a given medium, it
is suggested that the medium be checked
for stability, pH after sterilization,
and to see that performance is satis-
factory. While rare, an occasional
lot of medium will have some unforeseen
fault which reduces or destroys its
effectiveness. Maintenance of lot
number records on medium gives
opportunity for communication with
the manufacturer to determine whether
similar problems are being encountered
in other laboratories.
B Quality of General Materials
1 Distilled water
Distilled water, or demineralized water,
is required. It must be free from
NOTE: Mention of commercial products and manufacturers is for illustration and does not imply
endorsement by the Environmental Protection Agency.
W. HA. met. 19e. 1. 78
4-1
-------
Media and Solutions for Multiple Dilution Tube Methods
C
dissolved metals or chlorine. Freedom
from bactericidal constituents or growth
promoting substances should be dem-
onstrated through laboratory tests.
A procedure for this test is described
elsewhere in this course manual.
2 Beef extract
Any brand of beef extract is acceptable,
provided that it is known to give results
acceptable to the user. Meat infusion
is not acceptable.
3 Peptones
Peptones are sold under a wide variety
of trade names. Any peptone shown
satisfactory by comparative tests with
an acceptable peptone, may be accepted.
4 Sugars
All sugars must be chemically pure,
and suitable for bacteriological media.
5 A gar
Any form of bacteriologic grade of
agar can be used.
6 General chemicals must be reagent
grade or ACS if used in culture media.
Chemicals used in the distilled water
quality test must be of the highest purity
available.
7 Dyes
All dyes used in culture media must be
certified by the Biological Stain
Commission; they will be so labeled
on the container.
Quality of Equipment and Supplies Used
for Preparation of Media
1 Glassware
It is recommended that all glassware
be of borosilicate glass. Such glass
is not subject to release of soluble
products into the culture medium, as
with some of the so-called "soft glass. "
2 Balance
A balance with sensitivity of + 2 grams
with a load of 150 grams is the minimum
acceptable standard for weighing of
culture media in dehydrated form.
3 pH meter
An electrometric meter is recommended.
While a comparator block with pH
indicator solutions is useful for such
media as lauryl tryptose broth, it
cannot be used satisfactorily with dye-
containing media such as brilliant
green lactose bile broth. Therefore
it is suggested that all pH control
work on bacteriological media be done
with an electrometric type of pH meter.
Accuracy of the meter should be estab-
lished through calibration against a
standard buffer.
4 Autoclave
The autoclave should be of sufficient
size to permit loose packing of tubed
media when normal load is being
sterilized. This is to permit free
access of steam to all surfaces.
Operation should be such that sterilizing
temperature is reached in not more than
30 minutes.
A pressure gauge should be present.
More important, the autoclave should
be equipped with at least 1 thermometer,
which should be located properly in the
exhaust line.
Pressure regulation should permit
operation up to and including 121°C.
When media containing carbohydrates
are present, sterilization should be
continued 12 - 15 minutes; in media
not containing carbohydrates, normal
sterilization time should be a standard
15 minutes.
After sterilization, media should be
removed from the autoclave as soon
as possible. In no case should an
autoclave simply be turned off after
4-2
-------
Media and Solutions for Multiple Dilution Tube Methods
the usual exposure to steam under
pressure, and allowed to stand until the
following morning before removing media.
5 Utensils for mixing and preparing media
Borosilicate glass is suggested, but
other materials, such as stainless
steel, porcelain (unchipped) containers,
or other containers free of soluble
bactericidal or bacteriostatic materials,
are acceptable. In any case, the con-
tainers must be thoroughly clean.
IE CONCENTRATION OF MEDIA
A Basic formulas of all media described in
Section IV are presented as single-strength
media. Most media are used in the single-
strength concentration.
B The concentration of primary inoculation
media (media into which the measured
portions of the original sample are
delivered) requires special consideration.
1 When the amount of medium is 10 ml or
greater, and the volume of sample or
sample dilution is 1 ml or less, then
single-strength medium is satisfactory.
2 When the sample volume introduced
into the primary inoculation medium
is greater than 1 ml, then it is necessary
to compensate for the diluting effect
of the sample on the culture medium.
In such cases, it is necessary to
increase the initial concentration of
the medium so that after sample
inoculation the concentration of nutrients
in medium-plus-sample is equivalent
to the concentration of nutrients in the
single strength medium.
IV PREPARATION OF MEDIA AND SOLUTIONS
A Lauryl Tryptose Broth (Lauryl Sulfate Broth)
1 Use: Primary inoculation medium in
Presumptive Test
2 Composition:
Tryptose (or Trypticase
or equivalent)
Lactose
Dibasic Potassium
Phosphate (K2HPO4)
Monobasic Potassium
Phosphate (KH2PO4)
Sodium Chloride
Sodium Lauryl Sulfate
(Total Dry Constituents
Distilled Water
20.0 g
5.0 g
2.75 g
2.75 g
5.0
0. 1
35.60 g)
1000 ml
Sterilization: 12 - 15 minutes at 121QC
Reaction after sterilization: pH 6. 8
approximately)
3 Compensation for diluting effect of
samples
No. ml
medium
in tube
10
10
20
35
Ml of
sample or
dilution
0. 1 - 1. 0
10
10
100
Nominal
concentra-
tion before
inoculation
Ix
2x
1. 5x
4x
No. grams
dehydrated
medium per
liter
35.6
71.2
53.4
137.3
B Brilliant Green Lactose Bile Broth
1 Use: Confirmed Test
2 Composition
Peptone (Bacto or equivalent) 10.0 g
Lactose 10.0 g
Oxgall (dehydrated) 20.0 g
Brilliant Green 0.0133 g
(Total weight dry constituents 40. 0133 g)
Distilled Water
1000
ml
Sterilization: 12 - 15 minutes at 121OC
Reaction after sterilization: pH 7.1 to 7.4
4-3
-------
Media and Solutions for Multiple Dilution Tube Methods
C: Eosin Methylene Blue Agar (Levine's
Modification)
1 Confirmed Test
Use: Isolation of colitorm-like
colonies as a preliminary to
Completed Test procedures.
2 Composition
Peptone (Bacto or equivalent)
Lactose
Dipotassium Phosphate (K HPO ) 2
Agar 20
Eosin Y
Methylene Blue
(Total weight dry constituents
Distilled Water
1000 ml
Sterilization: 12 - 15 minutes at 121OC
3 Special suggestions on preparation:
a This medium can be prepared and
dispensed into bottles or flasks in
portions of 100 ml or 200 ml each.
The sterile medium may be stored
for extended periods in cool places
out of the light.
b When ready for use of such medium,
the medium should be melted by
immersion of the bottle of prepared
medium in a boiling water bath,
after which it is dispensed into
sterile Petri dishes in portions of
approximately 15 ml. After cooling
and solidifying in the Petri dish, the
medium is ready for use. It should
be used preferably on the day it is
poured into Petri dishes, but can be
stored for a day or two in the
refrigerator.
c An alternate method of preparing
this medium requires preparation
the agar base medium which
includes all the constituents
of the medium except the dyes.
When ready to use such a preparation,
the agar base medium is melted in
a water bath, and to each 100 ml of
the melted agar base medium, 2 ml
of 2% of aqueous solution of eosin Y
and 1.3 ml of 0. 5% methylene blue
solution is delivered with a pipet.
The medium is mixed thoroughly,
poured into Petri dishes, and used
as previously described.
D Agar Slants
Use: This medium is used in the
Completed Test, to cultivate pure
cultures of strains of bacteria being
cultivated in preparation of a Gram-
stained smear.
Composition:
agar
The medium is nutrient
Peptone
Beef extract
Agar
(Total weight dry
constituents
Distilled Water
5.0 g
3.0 g
15.0 g
23.0 g)
1000 ml
Sterilization: 15 minutes at 121°C
Reaction after sterilization: pH 6.8
approximately
3 Special instructions: Dissolve the con-
stituents, using heat as needed; dispense
in amounts of approximately 8 ml per
tube. Screw-capped tubes extend shelf
life of the medium. After sterilization,
remove the melted medium from the
autoclave and place in a slanting
position until the medium has become
solidified. A routine procedure should
be established so that a uniform volume
of medium and a uniform surface of
slanted medium be present in each tube.
While this has no particular bearing on
Standard Methods procedures, certain
other laboratory procedures do require
uniform exposed surface area of the
slanted medium.
E Plate Count Agar
1 Use: This medium is used in the
distilled water test. It is not used in
other Standard Methods procedures
described in this course manual.
4-4
-------
Media and Solutions for Multiple Dilution Tube Methods
2 Composition: (Tryptone Glucose
Yeast Agar)
Peptone-tryptone (or equivalent) 5.0 g
Yeast extract 2 . 5 g
Glucose (dextrose) 1-Og
Agar 15. 0 g
(Total weight dry constituents 23. 5 g
Distilled Water 1000 ml
Sterilization: 15 minutes at 121° C
Reaction after sterilization: pH 7.0 +
0. 1
3 Special instructions in preparation:
Use heat as needed to dissolve and
melt the constituents. Dispense the
medium in flasks o.r bottles in portions
of 100 or 200 ml each and sterilize. In
this state it can be preserved for many
months, provided that it is protected
from evaporation of the water.
When ready to use, melt the medium
by heating, and cool to 45° C. At this
temperature the medium still should be
melted, and will be satisfacotry for
preparation of pour plates for plate
counts.
F EC Broth
1 Use: Test for fecal coliform bacteria
2 Composition:
Tryptose (Bacto or equivalent) 20. 0 g
Lactose 5.0 g
Bile Salts (Bacto #3 or equivalent) 1.5 g
Dipotassium phosphate (KgHPO ) 4.0 g
Monopotassium phosphate (KHPO) 1. 5 g
£1 ~L
Sodium chloride 5.0 g
(Total weight dry constituents 37. 0 g)
Distilled Water 1000 ml
Sterilization: 12 - 15 minutes at 121OC
Reaction after sterilization: pH 6.9
3 This medium is dispensed into culture
tubes with inverted fermentation, vials
and suitable caps.
G Azide Dextrose Broth
1 Use: Primary inoculation medium for
fecal streptococcal presumptive test.
2 Composition:
Beef extract 4.5 g
Tryptone or Polypeptone 15. g
Glucose 7.5 g
Sodium chloride 7.5 g
Sodium azide 0. 2 g
(Total dry constituents 34.7 g)
Distilled Water 1000 ml
Sterilization: 12 - 15 minutes at 121°C
Reaction after sterilization: about pH 7.2
3 Fermentation vials are not used with
azide dextrose broth.
H Ethyl Violet Azide Broth
1 Use: Confirmed test for fecal
streptococci
2 Composition:
Tryptone or Biosate
Glucose
Sodium chloride
Potassium phosphate,
diabasic (KgHPC^)
Potassium phosphate,
monobasic (KH PO )
Sodium azide
Ethyl violet (certified dye
if available)
(Total dry constituents 35. 8 g)
Distilled Water 1000 ml
Sterilization: 12 - 15 minutes at 121° C
Reaction after sterilization: about pH 7
20 g
5 g
5 g
2.7 g
2.7 g
0.4 g
. 00083 g
4-5
-------
Media and Solutions for Multiple Dilution Tube Methods
3 Fermentation vials are not used with
ethyl violet azide broth.
I Buffered Dilution Water
1 Use: Preparation of sample dilutions
preliminary to primary inoculation, in
membrane filter work, and in plate
counts.
2 Composition
a Stock phosphate buffer solution
Monobasic Potassium 34. 0 g
Phosphate (KH PO )
Distilled Water 500 ml
IN NaOH solution (about 175 ml)
to give pH 7. 2
Distilled water sufficient to bring final
volume to 1000 ml
b Working solution of phosphate buffered
distilled water
Stock phosphate buffer solution 1.25 ml
Distilled water 1000 ml
3 Preparation and handling:
a Stock solution: After preparation the
stock solution should be stored in the
refrigerator until use. If at any time
evidence of mold or other contam-
ination appears, the stock solution
should be discarded and a fresh
solution prepared.
b Working solution: Dispense the
required amount into distilled water,
and deliver into screw-capped bottles
for dilution water. The amount added
should be such that, after sterilization,
the bottles will contain 99 + 2 ml of
the dilution water. Ordinarily this
requires initial addition of approxi-
mately 102 ml of the solution prior
to sterilization.
c Sterilization is 20 minutes at
121oc.
d Tightly stoppered bottles of the
dilution water, protected against
evaporation, in suitable containers,
appear to last indefinitely.
J Solutions for Gram Stain
1 Ammonium oxalate crystal violet
solution:
a Dissolve 2 g crystal violet
(approximately 85% dye content) in
20 ml of 95% ethyl alcohol.
b Dissolve 0. 8 grams ammonium
oxalate in 80 nil distilled water.
c Mix solutions a and b.
d Filter through cheesecloth or coarse
filter paper.
e Problems with the gram stain
technique frequently are traceable
to the ammonium oxalate crystal
violet solution. In the event that
decolorization does not seem satis-
factory, the amount of crystal violet
in the solution can be reduced to as
little as 10% of the recommended
amount.
2 Lugol's iodine: Dissolve 1 g iodine
crystals and 2 g potassium iodide in
the least amount (usually about 5 ml)
of distilled water in which they are
soluble. After all crystals are in
solution, add sufficient distilled water
to bring the final solution to a volume
of 300 ml.
3 Counterstain: Dissolve 2.5 grams of
safranin in 100 ml of 95% ethyl alcohol.
For the working solution of counterstain,
add 10 ml of this solution of safranin to
100 ml of distilled water.
4-6
-------
Media and Solutions for Multiple Dilution Tube Methods
This outline was prepared by H. L. Jeter,
,., . „ _. ,-, , _. . . _ ,
Chief, Program Support Training Branch,
, c. , , ,. A, , . , USEPA, Cincinnati, Ohio 45268
1 Standard Methods for the Examination of
Water and Wastewater. 14th ed. 1975.
Descriptors: Cultures, Bacteria,
Enteric Bacteria, MPN, Multiple
Dilution Method, Most Probable Number,
Coliform, Fecal Coliform
4-7
-------
USE OF TABLES OB1 MOST PROBABLE NUMBERS
Part 1
I INTRODUCTION
A Using probability mathematics, it is
possible to estimate the number of bacteria
producing the observed result for any com-
bination of positive and negative results
in'dilution tube tests. Because the
computations are so repetitious and time-
consuming, it is common laboratory
practice to use Tables of Most Probable
Numbers. These tables are orderly
arrangements of the possible cultural
results obtainable from inoculating various
sample increments in differential culture
media. Each possible combination of
positive and negative tube results is
accompanied by the result (MPN) of the
calculated estimate and the 95% confidence
limits of the MPN.
B The Tables of Most Probable Numbers
used in the current (14) edition of Standard
Methods for the examination of Water and
Wastewater were developed by Swaroop. (1)
Previous editions of Standard Methods have
used the tables prepared by Hoskins.^2)
1 Most of the tables are based on using
3 sample volumes in decreasing decimal
increments. Thus, the systems are
based on using volumes of 10 ml, 1. 0 ml,
and 0. 1 ml, etc. Other quantity
relationships can be used, such as
50 ml, 10 ml, and 1. 0 ml in a table.
Tables of Most Probable Numbers can
be prepared for any desired series of
sample increments.
2 In addition, tables can be devised for
different numbers of replicate
inoculations of individual sample
volumes. For example, the MPN
Table most commonly used in the
laboratories of this agency is based
on five replicate 10 ml portions, five
1. 0 ml portions, and five 0. 1 ml portions.
A separate table is required for another
combination of sample volumes, con-
sisting of five replicate 10 ml portions,
II
one 1. 0 ml portion, and one 0. 1 ml
portion. This is popular in bacteri-
ological potability tests on water.
MPN Tables can be prepared for any
desired combinations of replicates of
the sample increments used in a
dilution tube series.
3 An approximation of the MPN values
shown in the Tables can be obtained
by a simple calculation, developed by
Thomas.'3) The formula and
application of this calculation is shown
on a later page of this chapter.
The method of using a Table of Most
Probable Numbers is described here,
based on the table for five 10 ml portions,
five 1. 0 portions, and five 0. 1 portions.
The principles apply equally to the other
tables presented in the current edition of
Standard Methods for the Examination of
Water and Wastewater.
DETERMINING THE MOST PROBABLE
NUMBER
Codifying Results of the Dilution Tube
Series
If five 10 ml portions, five 1. 0 ml portions,
and five 0. 1 ml portions are inoculated
initially, and positive results are secured
from five of the 10 ml portions, three of
the 1. 0 ml portions and none of the 0. 1 ml
portions, then the coded result of the test
is 5-3-0. The code can be looked up in
the MPN Table, and the MPN per 100 ml
is recorded directly. If more than the
above three sample volumes are to be
considered, then the determination of the
coded result may be more complex. The
examples described in Table 1 are useful
guides for selection of the significant series
of three sample volumes.
VV. BA. 42g. 1.78
5-1
-------
Use of Tables of Most Probable Numbers
Table 1. EXAMPLES OF CODED RESULTS
No. ml sample per tube — >• 100
No. tubes per sample vol. -* 5
No. tubes in sample giving
positive results in test _
0
5
10
5
5
5
4
5
5
5
0
0
1
1.0
5
4
4
1
4
5
5
0
1
0
0. 1
5
1
0
0
1
5
5
0
0
0
0.01
5
0
0
1
4
5
0
0
0
0.001
5
0
0
0
Code
5-4-1
5-4-0
4-1-0
5-4-2
5-5-4
5-5-5
0-0-0
0-1-0
1-0-0
See Below
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
Discussion of examples:
1 When all the inoculated tubes of more
than one of the decimal series give
positive results, then it is customary
to select the smallest sample volume
(here, 10 ml) in which all tubes gave
positive results. The results of this
volume and the next lesser volumes
are used to determine the coded result.
2 When none of the sample volumes give
positive results in all increments of
the series, then the results obtained
are used to designate the code. Note
that it is not permissible to assume
that if the next larger increment had
been inoculated, all tubes probably
would have given positive results and
therefore assign a 5-4-1 code to the
results.
3 Here the results are spread through
four of the sample volumes. In such
cases, the number of positive tubes in
the smallest sample volume is added
to the number of tubes in the third
sample volume (counting down from the
smallest sample volume in which all
tubes gave positive results).
4 Here it is necessary to use the 5-5-4
code, because inoculations were not
made of 0. 001 ml sample volumes;
and it is not permissible to assume
that if such sample volumes had been
inoculated, they would have given
negative results, or any other arbitrarily-
designated result.
5 This is an indeterminate result. Many
MPN tables do not give a value for such
a result. If the table used does not
have the code, then look up the result
for code 5-5-4, and report the result
"greater than" the value shown for the
5-5-4 code. The first number of the
5-5-4 code is based on the 1. 0 ml
sample volume.
6 Like (5), this is an indeterminate
result. If the code does not appear in
the table being used, then look up the
result for code 1-0-0, and report the
MPN as "less than" the value shown
for the 1-0-0 code.
7 The current edition of Standard
Methods stipulates this type of code
designation when unusual results
such as this occur.
-------
Use of Tables of Most Probable Numbers
8 Note the difference from (7) above.
Inoculations of 100 ml portions were
not made, and it cannot be assumed
that the result would have called for
code 0-1-0.
B Computing and Recording the MPN
When the dilution tube results have been
codified, they are read and recorded from
the appropriate MPN Table.
1 If, as in the first four of the examples
shown under (A) the first number
in the coded result represents a 10 ml
sample volume, then the MPN per 100 ml
is read and recorded directly from the
appropriate column in the table.
2 On the other hand, if the first number
in the coded result represents a sample
volume other than 10 ml, then a
calculation is required to give the
corrected MPN. For example (4) under (A)
above, the first "5" of the 5-5-4 code
represents a sample volume of 1. 0 ml.
Look up the 5-5-4 code as if the 1. 0 ml
volume actually were 10 ml, as if the
0. 1 ml volume actually were 1. 0 ml
and as if the 0. 01 ml volume actually
were 0. 1 ml. The MPN obtained (1600)
then is multiplied by a factor of 10 to
give the corrected value. A simple
formula for this type of correction is
shown on a later page of this chapter.
IE PRECISION OF THE MPN VALUE
A The current edition of Standard Methods
shows for each MPN value, the 95%
confidence limits for that value. This
draws attention to the fact that a given
MPN value is not a precise measurement,
but an estimate. The 95% confidence
limits means that the observer will be
correct 95% of the time when he considers
that the actual number of cells producing
the observed combination of positive and
negative tubes was somewhere between
the stated upper limit and the stated
lower limit.
B The greater the number of replicates of
each sample volume in a dilution series,
the greater the precision (in other words,
the narrower the limits of the 95%
confidence range) of the test. The
precision of results, based on numbers
of tubes inoculated per sample volume,
is shown in Table 2.
(4)
C Woodward and other workers have
" studied the precision of the MPN in
detail. Such reports should be studied
by those desiring further information
regarding the precision of the MPN test.
Table 2. Approximate Confidence Limits for Bacterial Densities as
Per Cent of MPN as Determined from Various Numbers of Tubes
in Three Decimal Dilutions*
Number of tubes 50% 75% 80% 90% 95%
in each dilution Lower Upper Lower Upper Lower Upper Lower Upper Lower Upper
1
2
3
5
10
33
47
53
64
76
186
160
150
139
127
18
31
38
49
63
340
246
215
182
152
15
27
34
46
60
402
276
237
196
160
10
20
26
37
52
637
383
311
241
184
6. 5
15
21
31
46
955
511
395
289
208
*The interpretation of these figures is as follows: When MPN estimates are
made on the basis of dilution tests using one tube in each of three decimal
dilutions, you will be right 50% of the time if you say that the true bacterial
density is between 33% and 186% of the MPN. If you had used 5 tubes in each
dilution you could reduce this intervalto from 64% to 139% of the MPN and still
be right 50% of the time. If a greater certainty were desired, say 95%, you
would have to widen this interval to from 31% to 289%.
5-3
-------
Use of Tables of Most Probable Numbers
IV OCCURRENCE OF IMPROBABLE TUBE
RESULTS
A Many of the theoretically possible tube
results are omitted from the MPN Table.
For example, codes 0-0-3, 0-0-4, and
0-0-5 are not included as well as many
others. These are omitted, because, in
the opinion of the authors of the tables,
the probability of occurrence of such
results is so low as to exclude them from
practical consideration.
B The frequency of occurrence of various
code results is shown in the Table 2 both
on a theoretical basis and on the basis of
actual laboratory experience.
C From the MPN tables, it can be inferred
that the codes omitted from the MPN
Table can be expected to occur up to 1%
of the time. If, in reviewing laboratory
data, the theoretically unlikely codes
occur appreciably more than 1% of the
time, there is an indication for inquiry
into the causes. Such results can occur (1)
as a consequence of faulty laboratory
procedures, or (2) as a result of
extraneous influences in the samples.
D The current edition of Standard Methods
does not include MPN values for many
rare combinations listed in previous
editions. By pruning out those codes listed
as Group IV in Table 3, the table has been
considerably condensed. Table 4 suggests
maximum permissible numbers of samples
for various numbers of samples tested.
Table 3
FIVE-TUBE AND THREE-TUBE CODES THAT
INCLUDE 99 PEE CENT OF ALL RESULTS
Group
Class 1 codes
550, 551, 552, 553,
554, 500, 510, 520,
530, 540, 100, 200,
300, 400.
Class 2 codes
511, 521, 531, 541,
542, 110, 210, 310,
410, 420.
Class 3 codes
501, 010, 532, 320,
522, 220, 543, 430,
120, 533, 330, 502,
020, 544, 440, 301,
401, 431, 201, 411,
101, 311, 421, 211,
001.
Improbable codes
Class 1 codes
330, 331, 332, 300,
310, 320, 100, 200.
Class 2 codes
321, 311, 301, 210,
110, 010.
Class 3 codes
322, 220, 201, 101
312, 120.
Improbable codes
Theoretically Ex-
pected Percentage
of Results
Theoretically Ex-
pected Cumulative
Percentage
Observed Percentage
of 360 Samples
Five-Tube-Test
67.5
23. 6
7. 9
1.0
67.5
91. 1
99.0
100.0
68.0
23. 1
7. 5
1.4
Three-Tube Test
81.5
14.9
2. 7
0. 9
81.5
96.4
99. 1
100. 0
81. 7
14. 1
3.7
0.6
-------
Use of Tables of Most Probable Numbers
Table 4
MAXIMUM PERMISSIBLE NUMBERS
OF IMPROBABLE CODES FOR VARIOUS
NUMBERS OF SAMPLES TESTED
Number of Maximum Number
Samples of Improbable Codes
1 -
16 -
46 -
84 -
131 -
181 -
234 -
291 -
351 -
414 -
478 -
15
45
83
130
180
233
290
350
413
477
543
1
2
3
4
5
6
7
8
9
10
11
? Table 5 is from International standards
for Drinking-Water, published by the World
Health Organization, Geneva (1958). The
last three values, not shown in the WHO
publication, are from Woodward, "How
Probable is the Most Probable Number. "(4)
F Several theoretically possible combinations
of positive tube results are omitted in
Table 5. These combinations are omitted
because the statistical probability of
occurrence of any of the missing results
is less than 1%. If such theoretically
unlikely tube combinations occur in more
than 1% of samples, there is need for
review of the laboratory procedures and
of the nature of the samples being tested.
When the series of decimal dilutions is
other than 10, 1.0 and 0. 1 ml, use the
MPN in Table 5, according to the following
formula:
Example: From a sample of water, 5 out
of five 0. 01 - ml portions, 2 out of five
O.'OOl - ml portions, and 0 out of five
0. 0001 - ml portions, gave positive
reactions.
From the code 5-2-0 in the MPN table,
the MPN index is 49
49
X-
10
(from table) 0. 01
= 49,000
MPN Index = 49, 000
A simple approximation of the most
probable number may be obtained from
the following formula (after Thomas):
MPN/100 ml =
No. of Positive Tubes X 100
No. of ml in negative tubes) X(No. of ml
in all tubes)
Example: From a sample of water, 5 out
of five 10 - ml portions, 2 out of five
1. 0 ml portions, and 0 out of five 0. 1 ml
portions gave positive results.
MPN/100 ml =
7 X 100
•= 50.22
(3.5)X(55.5)
MPN/100 ml = 50
Note that the MPN obtained from the table
on the preceding pages with these tube
results is 49. "Most probable numbers
computed by the above fornula deviate
from values given by the usual methods
by amounts which ordinarily are
insignificant. The formula is not
restricted as to the number of tubes
and dilutions used " (Thomas)
MPN
(from table)
X
10
Largest quantity tested
= MPN/100 ml
-------
Use of Tables of Most Probable Numbers
Table 5. MPN INDEX AND 95% CONFIDENCE LIMITS FOR VARIOUS
COMBINATIONS OF POSITIVE AND NEGATIVE RESULTS WHEN FIVE
10-ML PORTIONS, FIVE 1-ML PORTIONS AND FIVE 0. 1 ML PORTIONS
ARE USED
No. of Tubes Giving
Positive Reaction out of
5 of 10
ml Each
0
0
0
0
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
3
4
4
4
4
4
4
5 of 1
ml Each
0
0
1
2
0
0
1
1
2
0
0
1
1
2
3
0
0
1
1
2
2
3
0
0
1
1
1
2
5 of 0. 1
ml Each
0
1
0
0
0
1
0
1
0
0
1
0
1
0
0
0
1
0
1
0
1
0
0
1
0
1
2
0
MPN
Index
per
100 ml
<2
2
2
4
2
4
4
6
6
5
7
7
9
9
12
8
11
11
14
14
17
17
13
17
17
21
26
22
95% Con-
fidence Limits
Lower
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
1
1
2
2
3
1
2
2
4
4
5
5
3
5
5
7
9
7
Upper
7
7
11
7
11
11
15
15
13
17
17
21
21
28
19
25
25
34
34
46
46
31
46
46
63
78
67
No. of Tubes Giving
Positive Reaction out of
5 of 10
ml Each
4
4
4
4
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5 of 1
ml Eac
2
3
3
4
0
0
0
1
1
1
2
2
2
3
3
3
3
4
4
4
4
4
5
5
5
5
5
5
5 of 0. 1
ml Each
1
0
1
0
0
1
2
0
1
2
0
1
2
0
1
2
3
0
1
2
3
4
0
1
2
3
4
5
MPN
Index
per
100 ml
26
27
33
34
23
31
43
33
46
63
'49
70
94
79
110
140
180
130
170
220
280
350
240
350
540
920
1600
=2400
35% Con-
fidence Limits
Lowe
9
9
11
12
7
11
15
11
16
21
17
23
28
25
31
37
44
35
43
57
90
120
68
120
180
300
640
Upper
78
80
93
93
70
89
110
93
120
150
130
170
220
190
250
340
500
300
490
700
850
1, 000
750
1, 000
1, 400
3, 200
5, 800
5-6
-------
Use of Tables of Most Probable Numbers
Table 6. MPN AND 95% CONFIDENCE LIMITS FOR VARIOUS
COMBINATIONS OF POSITIVE RESULTS IN A PLANTING
SERIES OF FIVE 10-ml PORTIONS OF SAMPLE
No. of Positive Tubes Out of:
Five 10 -ml Tubes
0
1
2
3
4
5
MPN per
100 ml
2.2
2. 2
5. 1
9. 2
16.0
> 16
Limits of MPN
Lower
0
0. 1
0.5
1.6
3. 3
8.0
Upper
6.0
12.6
19. 2
29.4
52.9
Table 7. MPN AND 95% CONFIDENCE LIMITS FOR VARIOUS
COMBINATIONS OF POSITIVE RESULTS IN A PLANTING SERIES OF
FIVE 10-ml, ONE 1-ml, AND ONE 0.1-ml PORTIONS OF SAMPLE
No. of Positive Tubes Out of:
Five 10 -ml
Tubes
0
0
1
1
2
2
3
3
4
4
4
5
5
5
5
One 1 -ml
Tube
0
1
0
1
0
1
0
1
0
0
1
0
0
1
1
One 0. 1-ml
Tube
0
0
0
0
0
0
0
0
0
1
0
0
1
0
1
MPN
per
100 ml
<2
2
2. 2
4.4
5
7. 6
8. 8
12
15
20
21
38
96
240
>240
Limits of MPN
Lower
0.050
0.050
0.52
0.54
1. 5
1. 6
3. 1
3. 3
5. 9
6.0
6.4
12
12
88
Upper
5.9
13
13
14
19
19
29
30
46
48
53
330
370
3700
5-7
-------
Use of Tables of Most Probable Numbers
IV TABLES OF MOST PROBABLE
NUMBERS
These tables consist of the MPN indices and
95% confidence limits, within which the
actual number of organisms can lie, for
various combinations of positive and
negative tubes. Three MPN tables are
presented. Table 5 is based on five 10 ml
five 1. 0 ml and five 0. 1 ml sample portions.
Table 6 is based on five 10 ml sample por-
tions; and Table 7 is based on five 10 ml,
one 1 ml ana one 0. 1 ml sample portion.
REFERENCES
1 Swaroop, S. Numerical Estimation of
JB_. coli by Dilution Method. Indian J.
Med. Research. 26:353. 1938.
2 Hoskins, J.K. Most Probable Numbers
for Evaluation of Coli-Aerogenes Tests
by Fermentation Tube Method. Public
Health Reports. 49:393. 1934.
3 Thomas, H. Al, Jr. Bacterial
Densities from Fermentation Tubes.
J.A.W.W.A. 34:572. 1942.
4 Woodward, R.L. How Probable is the
Most Probable Number? J.A.W.W.A.
49:1060. 1958.
5 Standard Methods for the Examination
of Water and Wastewater. 14th
Edition. Prepared and Published
Jointly by American Public Association.
American Water Works Association,
and Water Pollution Control Federation.
This outline was prepared by H. L. Jeter,
• -tor, National Training Center,
Water Programs Operations, EnvironmentaJ
Protection Agency, Cincinnati, OH 452H8.
-------
MPN THEORY
Part 2
I DERIVATION OF THE MPN
A Assumptions
The validity of the MPN procedure is based
upon two principal assumptions.
1 In statistical language, the first is that
the organisms are distributed randomly
throughout the liquid. This means that
an organism is equally likely to be
found in any part of the liquid, and that
there is no tendency for pairs or groups
of organisms either to cluster together
or to repel one another.
2 The second assumption is that each
sample from the liquid, when incubated
in the culture medium, is certain to
exhibit growth whenever the sample
contains one or more organisms.
B The Probability Equation
Based upon these assumptions, an equation
for the probability of the observed com-
bination of positive and negative tubes can
be derived as a function of the true density
6. By solving this equation for different
values of 6 a curve can be plotted as shown
in Figure 1.
Curves of this type always have a single
maximum or peak. The value of 6, say d,
which corresponds to the peak of the curve
is called the most probable number,
commonly designated as MPN.
The MPN is "most probable" in the sense
that it is the number which maximizes the
probability of the observed results. It is
interesting to note that although the
original derivation of the MPN predates
modern statistical estimation, the MPN
procedure corresponds to the currently
accepted estimation procedure known as
the "method of maximum likelihood. "
MAXIMUM
d(MPN)
BACTERIA PERIOD ML
FIGURE 1
C Indeterminant Solutions
The MPN provides a meaningful estimate
of 6 only if there are both positive and
negative tubes in at least one dilution.
If all tubes are negative, the maximum
of the probability curve occurs when 6 is
set equal to zero (see Figure 2) and thus
the MPN is zero. If all tubes are positive,
the maximum of the probability curve
occurs when 6 is set equal to infinity
(see Figure 3) and thus the MPN is infinity.
ST. 47."7". 70
5-9
-------
MPN Theory
1.00
ALL TUBES NEGATIVE
MAXIMUM AT 8=0
BACTERIA PER 100 ML
FIGURE 2
CQ
<
03
O
at
a.
SKEWED
MPN
FIGURE 4
1.00
ALL TUBES POSITIVE
CO
<
CQ
O
Qi
0.50-
MAXIMUM AT 8=
BACTERIA PER 100 ML
FIGURE 3
n DISTRIBUTION OF MPN VALUES
A Skewed Distribution
If a very large number of independent
MPN determinations were made on the
same water sample, the distribution of
the MPN values would be such that very
high values relative to the median value
would occur more frequently than very
low values. Thus the distribution of MPN
values is skewed to the right as shown in
Figure 4.
B Logarithmically Normal Distribution
Since it is mathematically inconvenient
to work with data distributed asymmetri-
cally, it is desirable to transform the
skewed data in such a way that the trans-
formed values have a symmetric distri-
bution resembling the normal. In the
case of MPN values the logarithms of
the MPN's are approximately normally
distributed as shown in Figure 5.
CO
<
CQ
O
c*
O.
SYMMETRIC
(NORMAL)
log(MPN)
FIGURE 5
5-10
-------
MPN Theory
C Precision of MPN Estimates
The lack of precision of MPN estimates
of bacterial densities is generally recog-
nized. A measure of the precision is
given by the confidence limits on the
estimate which can be computed on the
basis of the normal distribution of the
logarithms of the MPN values. It has
been verified that three-tube and five-
tube MPN estimates are approximately
logarithmically normal and the standard
deviation of the logarithms of the MPN's
is given by the formula;
log
where o is the standard deviation of
the logarrtnms of the MPN estimates and
n is the number of tubes in each dilution.
The upper and lower 95% confidence limits
of an MPN estimate are given by the
formulas:
UCL = antilog (log MPN + 1. 96 a )
o
= MPN • k,
LCL = antilog (log MPN - 1. 96 cr )
= MPN -H k,
where k = antilog (1.96
z
LU
o
u.
z
o
450-
400-
300-
200 -
o
t^
O
JE 100-
Q
35 10 12
NUMBER OF TUBES PER DILUTION
I
20
FIGURE 6
5-11
-------
MPN Theory
Similarly, the expected number or
organisms in the lowest sample volume
(highest dilution) v should not exceed
one, to avoid the risk that all tubes will
be positive.
B The Rule
The above line of reasoning leads to the
rule that a dilution series is capable of
estimating any density between 1/v and
1/v . In practice, we use the rule oy
firsi guessing two limits 6 and 6 between
which we are fairly certain that tne actual
density lies. The sample volumes are
then chosen to satisfy the rules
"«- i; "Li t
Table 1 displays the range of densities
covered by various decimal dilution
series.
TABLE 1
SAMPLE VOLUME
(ML)
RANGE COVERED
(COLIFORMS/100 ML)
10
10
10
1C
io
ID
io-
io-7j
10 ' - IO3
102-104
IO3 - TO5
104-106
105-107
io6-io8
io7-io9
This outline was prepared by J. H. Parker,
former Statistician, Analytical Quality
Control, Bureau of Water Hygiene, EPA,
Cincinnati, OH.
descriptors: Bacteria, Coiiforms, MPN,
Most Probable Number, Measurement,
Microbiology, Laboratory Tests
-------
THE MEMBRANE FILTER IN WATER BACTERIOLOGY
I HISTORICAL BACKGROUND
There is sometimes a tendency to look upon
membrane filters and their bacteriological
applications as new developments. Both the
filters and many of their present bacterio-
logical applications are derived from earlier
work in Europe.
A Some European developments prior to
1947 are as follows:
1 Pick is credited with application of
collodion membranes in biological
investigations in 1855.
2 Sanarelli, in 1891, reported develop-
ment of membrane filters impermeable
to bacteria but permeable to their
toxins.
3 Bechhold, in the early 1900's made
a systematic study of the physico-
chemical properties of a number of
varieties of these membranes. After
1911 numerous investigations were
made in several countries with respect
to the properties of collodion membranes.
4 Zsigmondy and Bachmann, 1916-1918,
developed improved production methods
which were applicable on a commercial
scale. Membrane filters have been
produced for many years at the
Membranfiltergesellschaft, Sartorious
Werke, in Goettingen, Germany. In
1919 Zsigmondy applied for a U.S.
patent on his production methods; it
was granted in 1922.
5 In the 1930's, W. J. Elford in England,
and P. Graber in France, made new
contributions in developing and teaching
methods for making collodion membranes
with controlled pore size.
6 Before World War II filtration
procedures using the Zsigmondy
membrane had been suggested for the
determination of bacterial counts,
coliform determinations, and isolation
of pathogenic bacteria from water and
other fluids. Most early interest in
developing these techniques seems to
have been in Germany and in Russia.
During World War II Dr. G. Mueller
applied membrane filter techniques
to the bacteriological examination of
water, following bomb destruction of
many of the laboratories.
B Developments in the United States
1 In 1947, Dr. A. Goetz reported on a
mission to Germany as a scientific
consultant to the Technical Industrial
Intelligence Branch, U. S. Department
of Commerce. He obtained detailed
information about the nature, method
of preparation, and specific bacterio-
logical applications of the Zsigmondy
Membranfilter being manufactured
by the Membranfiltergesellschaft in
Goettingen.
2 After his return to this country,
Dr. Goetz developed methods for
preparing and improved type of
membrane filter from domestic
materials. On a small scale he
manufactured filters under a
government contract; afterward
membrane filter manufacture was
continued by a commercial organization.
3 In 1950, bacteriologists of the Public
Health Service began intensive study
of the applications of membrane filters
in bacteriological examinations of
water. Their first report was
published in 1951, and was followed
by numerous reports of other similar
investigations. Such studies have
been widely expanded, as indicated in
references shown elsewhere in this
manual.
NOTE: Mention of commercial products and manufacturers does not imply endorsement by tht
OWP, Env."j.roimi.jnlai Protection Agency.
6-1
-------
The Membrane Filter in Water Bacteriology
In 1955 the 10th Edition of "Standard
Methods for the Examination of Water,
Sewage, and Industrial Wastes" included
a tentative method for coliforms by mem-
brane filter method. In the llth and 12th
editions, the membrane filter method
for coliforms has become official. In
addition, methods for enterococcus
(fecal streptococci) are included as
ive methods. The 13th edition
ud ^iven tne fecal streptococcus test
a standard designation and the tentative
method status reserved for the "pour
plate" technique of quant it at ion.
The membrane filter is an official
method for examination of potable waters
in interstate commerce. The Public
Health Service Drinking Water Standards
(1962) state "Organisms of the coliform
group. . . All the details of technique
. . . shall be in accordance with Standard
Msthods for Examination of Water and
Wastewater, current edition. .." Thus,
acceptance by Standard Methods as
official automatically validates a method
for use with interstate waters.
II PROPERTIES OF MEMBRANE FILTERS
Membrane filters used in water bacteriology
are flat, highly porous, flexible plastic discs
about 0. 15 millimeters in thickness and usually
47-50 millimeters in diameter.
A Principle of Manufacture
The procedures described below are from
FIAT Report 1312. While the methods in-
dicated by Goetz do not necessarily describe
the current manufacturing processes, it is
assumed that similar principles of manu-
facture still apply.
1 One or more cellulose esters, such as
cellulose nitrate, is dissolved in a
suitable solvent.
2 Water, or some other liquid insoluble
in the cellulose solution, is added and
mixed, to form an emulsion having great
uniformity in size and distribution of
droplets of the insoluble liquid.
3 The emulsion is cast on plates and dried
in an environment rigidly controlled as to
humidity and temperature. The droplets
of insoluble fluid retain their size and
identity in the dried film, eventually
becoming the pores of.the finished
membrane.
4 The dried porous film is cut into filter
discs of the desired size. Representa-
tive discs are subjected to control tests
for accurate determination of the pore:
size obtained.
5 Particle retention by membrane filters
is at or very near the filter surface, by
a mechanical, sieve-like action. (This
applies to hydrosols, not to aerosols.)
Through manufacturing control it is
possible to make membrane filters with
controlled pore size, within narrow limits
B Some Important Characteristics of
Membrane Filters
1 The membrane filters used in micro-
biology should be flat, circular, gridded,
of uniform thickness and porosity, non-
toxic to microorganisms, wettable,
able to withstand commonly employed
sterilizing conditions, and unaffected by
the fluid to be filtered.
2 Without reference to specific manufacturers,
some particulars of their products have
included:
a ... Average pore diameter ranging
from 5 millimicrons to 10 microns.
Thicknesses ranging from 70 to
150 microns. Can be sterilized
by autoclaving at 121°C for 10
minutes.
b ... mean flow pore size ranging from
7. 5 millimicrons to 5 microns,
The pore size used in water
bacteriology having a standard
diameter, has a_water flow rate
of 70cc/min/cm and must pass
100 ml of particle-free water
within 9 seconds.
-------
The Membrane Filter in Water Bacteriology
. .. currently produced in more
than twenty distinct pore sizes
from 14 microns to 10 milli-
microns in discs ranging from
13 mm to 293 mm in diameter.
The total range of pore size
distribution of the type used in
water microbiology of 0.45 microns
is plus or minus 0. 02 micron.
. .. membranes are offered in
graduated pore sizes ranging
from 12 microns to 5 milli-
microns. The types used in
water bacteriology have a dis-
tilled water Jlow rate of 65
ml/min/cm at 700 mm Hg
differential pressure or an air
flow rate of 0.4 liters/min/cm
at a differential pressure of
500mm water.
4 Membrane filters are wettable. Thus,
after sample filtration, when a filter
is placed on moist culture medium the
medium diffuses through the pores and
is available to organisms collected on
the opposite surface.
5 Membrane filters are free of soluble
chemical substances inhibitory to bac-
terial growth. Water soluble plasticizers
are included in one commercially pro-
duced filter (glycerol, 2.5%). The
cellulose esters themselves have some
absorbing tendency illustrated by some
dyes and heavy metals. Total ash is
v e ry low, le s s th an 0.0001 %.
6 Membrane filters have a uniform index
of refraction. With membrane filters,
this index is N : 1.5. When wetted
with a liquid having refractive index
within this range, the filters become
transparent. This property permits
direct microscopic examination of
particulate matter collected on the
filter surface.
up to 125 C in air. Membranes
of cellulose triacetate are advertised t, >
withstand dry heat to 266 C. In general
however, membranes in current use
must be sterilized cautiously. Con-
sult the laboratory equipment discussio
for details. Overheating of all types
interferes with filtration by blocking
pores.
C Nomenclature
Membrane filters used in bacteriological
tests on water are known under several
names. Though the names are different,
the filters are similar in form, propertie-.,
and method of use. Names commonly en-
countered are:
1 Membrane filter. This is the general
name for filters made according to
the general principles and having the
properties discussed above. The term
"membrane filter" is most used in
technical reports on filters of this typt .
2 Molecular filter. This name used by
Goetz for the improved type of filter
that he and his associates developed
after study of the manufacturing met hod t
at the Membranfiltergesellschaft in
Goettingen, Germany.
3 Millipore filter is a trade name for
membrane filters made by the Milliporr
Filter Corporation.
4 Bac-T-Flex filter is a trade name
applied to certain membrane filters
made by Carl Schleicher and Schuell
Company,,
Oxoid filter is a trade name applied to
filters made by Oxo, Ltd., London,
England.
7 Temperature resistance depends on
plastics used in the filter. The nitro-
••P! lulose membrane fi Iter is stable dry
-------
The Membrane Filter in Water Bacteriology
Micropore, Polypore and Metricel
have been trade names used by the
Gelman Instrument Company.
Ill APPLICATIONS IN WATER BACTERIOLOGY
A The basic cultural procedures for bacterio-
logical tests on membrane filters are:
1 A sample is filtered through a membrane
filter.
2 The filter is placed in a culture con-
tainer, on an agar medium or a paper
pad impregnated with moist culture
medium.
3 The inoculated filter is incubated under
prescribed conditions of time, tempera-
ture, and humidity.
4 After incubation, the resulting culture
is examined and necessary interpreta-
tions and/or additional tests are made.
B With variations in such factors as culture
media, incubation time, and combinations
with other cultural and biochemical tests,
several different kinds of tests are
available.
2) A verified membrane filter coli-
form test v,an be used when needed
as a supplement to the direct mem-
brane filter test. Pure cultures
are obtained from individual
colonies differentiated on the
membrane filter and subjected
to further cultural, biochemical,
and staining tests to establish
the identity of the colonies being
studied.
3) The delayed membrane filter
coliform test was developed to
overcome bacterial changes fre-
quently occurring when there is
a delay of one to several days
between sample collection and
the initiation of laboratory tests.
The test consists of sample fil-
tration at or shortly after the time
of sample collection. The inocu-
lated filter is placed on a preser-
vative medium and taken or sent
to a laboratory, where it is trans-
ferred to a growth medium for the
differentiation of coliform colonies.
After incubation the culture is ex-
amined and the results are evalu-
ated as for the direct membrane
filter coliform test.
1 Total bacterial counts are made by
cultivation of bacteria on membrane
filters using an enriched all-purpose
culture medium.
2 Tests for bacterial indicators of
pollution.
a Coliform tests
1) The direct membrane filter tests
for coliforms is one in which,
after sample filtration, the mem-
brane filter is incubated in con-
tact with one or more special
media. At least one of the media
is a selective, differential medium
including components which per-
mit coliform bacteria to develop
colonies easily recognizable by
form, color, sheen, or other
char acte ristic s.
4) A medium and technique for
detecting and counting fecal
coliform bacteria has been
developed and is called M-FC
Broth. This medium currently
is being used increasingly in
water pollution studies.
b Selective, differential, culture media
have been developed for direct cul-
tural tests for members of the enter-
ococcus group of bacteria.
3 Tests for pathogenic bacteria
a Workers are currently testing new
media for the differentiation of
members of the Salmonella-Shigella
group of enteric pathogens. Avail-
able information indicates potential
usefulness of a screening medium
6-4
-------
The Membrane Filter in Water Bacteriology
for differentiation of nonlactose-
fermenting, non urease-producing
bacteria.
b One medium has been used
for screening tests in detection of
Salmonella typhosa.
c Further confirmatory cultural, bio-
chemical, and serological tests are
necessary to establish the identity of
bacteria differentiated with these
screening media.
Membrane filter techniques can be applied
both in the laboratory and under field con-
ditions. Several varieties of portable mem-
brane filter field units have been developed
on a commercial basis.
IV ADVANTAGES AND LIMITATIONS
This evaluation is limited to tests for the c oil-
form group. Similar, but separate evaluations
would have to be made for any other bacterio-
logical test.
A Advantages
1 Results are obtained in approximately
24 hours, as compared with 48-96 hours
required for the standard fermentation
tube method.
2 Much larger, and hence more represen-
tative samples of water can be sampled
routinely with membrane filters.
3 Numerical results from membrane
filters have much greater precision
(reproducibility) than is expected with
the fermentation tube method.
4 The equipment and supplies required are
not bulky. A great many samples can be
examined with minimum requirements
for laboratory space, equipment, and
supplies.
B Limitations
1 Samples having high numbers of non-
coliform bacteria capable of growing on
Endo type culture media sometimes
give difficulty. In such cases a high
ratio of these noncoliform bacteria to
coliforms results in poor sheen pro-
duction, or even suppression, of the
coliform organisms.
2 In samples having low coliform counts
and relatively great amounts of sus-
pended solids, bacterial growth some-
times develops in a continuous film on
the membrane surface. In such cases
the typical coliform sheen sometimes
fails to develop.
3 Some samples containing as much as
1 milligram per liter of copper or zinc,
or both, show irregular coliform bac-
terial results.
4 Occasional strains of bacteria growing
on membrane filters producing sheen
colonies prove, on subsequent testing,
to be acid but not gas-producers from
lactose. Where this occurs it may
give a falsely-high indication of coliform
density.
Such limitations as these are not frequent,
but they do occur often enough to require
consideration. In samples where these dif-
ficulties often occur, the best course of
action often is to avoid use of membrane
filter methods and use the multiple fer-
mentation tube procedures.
V SUMMARY
The development of membrane filters and
their bacterial applications has been discussed
briefly, from their European origin to their
current status in this country. Membrane
filters currently available here have been
described, and their properties have been
considered. Applications of membrane filters
in water bacteriology are indicated in general
terms. Some of the advantages and limitations
of membrane filter methods are presented for
coliform tests.
-------
The Membrane Filter in Water Bacteriology
REFERENCES
1 Goetz, Alexander. Materials, Techniques,
and Testing Methods for the Sanitation
(Bacterial Decontamination) of Small-
Scale Water Supplies in the Field Used
in Germany During and After the War.
Technical Industrial Intelligence Branch.
U.S. Department of Commerce. FIAT
Final Report 1312. December 8, 1947.
2 Clark, Harold F., Geldreich, E.E.,
Jeter, H. L., and Kabler, P. W. The
Membrane Filter in Sanitary Bacteri-
ology. Public Health Reports. 66:
951-77. July 1951.
3 Goetz, A. and Tsuneishi, N. Application
of Molecular Filter Membranes to the
Bacteriological Analysis of Water.
Journal American Water Works Associ-
ation. 43:943-69. December 1951.
4 Clark, H. F., Kabler, P. W., and
Geldreich, E.E. Advantages and
Limitations of the Membrane Filter
Procedure. Water and Sewage Works.
104: 385-387. 1957.
) Public Health Service Publication 956.
Public Health Service Drinking Water
Standards. 1962,.
6 Windle Taylor, E., and Burman, N. P.
The Application of Membrane Filtratio
Techniques to the Bacteriological Ex-
amination of Water. Journal Applied
Bacteriology, 27:294-303. 1964.
This outline was prepared by H. L.
Jeter, Chief, Program Support
Training Branch, USEPA, Cincinnati,
Ohio 45268
Descriptors: Filters, Membrane,
Bacteria, Microorganisms, Laboratory
Tests
6-6
-------
MEMBRANE FILTER EQUIPMENT AND ITS
PREPARATION FOR LABORATORY USE
I Some equipment and supplies used in the
bacteriological examination of water with
membrane filters are specific for the method.
Other items are standard in most well-
equipped bacteriological laboratories and
are readily adapted to membrane filter work.
This chapter describes needed equipment and
methods for its preparation for laboratory
use. Where more than one kind of item is
available or acceptable for a given function,
sufficient descriptive information is pro-
vided to aid the worker in selecting the one
best suited to his own needs.
II EQUIPMENT FOR SAMPLE FILTRATION
AND INCUBATION
A Filter Holding Unit
1 The filter holding unit is a device for
supporting the membrane filter and for
holding the sample until it passes
through the filter. During filtration
the sample passes through a circular
area, usually about 35 mm in diameter,
in the center of the filter. The outer
part of the filter disk is clamped
between the two essential components
of the filter holding unit. (See Plate 1)
a The lower element, called the filter
base, or receptacle, supports the
membrane filter on a plate about
50 mm in diameter. The central
part of this plate is a porous disk
to allow free passage of liquids.
The outer part of the plate is a
smooth nonporous surface. The
lower eier ant includes fittings for
mounting .ne unit in a suction flask
or other container suitable for
filtration with vacuum.
b The upper element, usually called
the funnel, holds the sample until
it is drawn through the filter. Its
lower portion is a flat ring that rests
on the outer part of the membrane
filter disk, directly over the non-
porous part of the filter support
plate.
c The assembled filter holding unit is
joined by a locking ring or by one
or more clamps.
2 Characteristics of filter holding units
should include:
a The design of filter holding units
should provide for filtration with
vacuum.
(A) ASSEMBLED FILTER
HOLDING UNIT
(B) UPPER ELEMENT
LOCKING RING
(C) LOWER ELEMENT
PLATE 1
"OTF: Monfiun of commercial products and manufacturers does not imply endorsement by the
•'";' TI 'he Fnvir< inmentn I Prot 00 I. >t> Agency.
-------
Membrane Filter Equipment and its Preparation for Laboratory Use
b Filter holding units may be made
of glass, porcelain, plastic, non-
corrosive metal, or other impervious
material.
c Filter holding units should be made
of bacteriologically inert materials.
d All surfaces of the filter holding
assembly in contact with the water
sample prior to its passage through
the membrane filter should be
uniformly smooth and free from
corrugations, seams, or other sur-
face irregularities that could become
lodging places for bacteria.
e Filter holding units should be easily
sterilized by routine methods.
f The filter holding unit should be
easily and quickly assembled and
disassembled in routine operational
use.
g Filter holding units should be durable
and inexpensive. Maintenance should
be simple.
Several forms of filter holding units
have been developed for use with
aqueous suspensions.
a SS 47 Membrane Filter Holder
(Plate 2, Figure 1)
Conical-shape funnel with a 500 ml
capacity. The base section includes
a wirescreen membrane support.
Funnel and base section are evenly
joined by a locking ring mechanism.
This assembly is designed to hold a
47 mm diameter membrane firmly in
place allowing an effective filter area
of approximately 9. 6 square centi-
meters. The entire filter unit is
made of stainless steel with the
funnel interior having a mirror-like
finish.
b The Millipore Pyrex Filter Holder
(Plate 2, Figure 2)
The unit is made of pyrex glass with
coarse grade fitted support in base
for filter. The upper element of
early models of glass filter holders
had a capacity of 1 liter. Currently
available units are supplied with
upper elements having 300 ml
capacity. The assembled filter
holder is joined with a spring
clamp which engages on flat sur-
faces encircling the upper and
lower elements.
Millipore Standard Hydrosol Filter
Holder (Plate 2, Figure 3)
Most components of this unit are
made of stainless metal. The
porous membrane support plate is
fine-mesh stainless steel screening.
The upper element is a straight-
sided cylinder 4 to 5 inches in
diameter, constricted to a narrow
cylinder at the bottom, to fit the
lower element. Capacity of the
funnel element is about 1 liter.
The assembled filter holding unit is
joined by a bayonet joint and locking
ring. Accessories may be obtained
for collection of small amounts of
filtrate and for anhydrous sterilization
of the filter holding assembly.
Gelman "Parabella Vacuum Funnel"
(Plate 2, Figure 5)
The unit is made of spun stainless
steel. The locking ring is a bayonet-
type fitting, and is spring-loaded.
The funnel element has a 1-liter
capacity.
The Sabro Membrane Filter Holder
(Plate 2, Figure 4)
The unit is mostly of stainless steel
construction. The lower element is
a combination vacuum chamber,
filtrate receiver, and filter support-
ing element. It consists of a stainless
steel cup with a metal cover. The
cover is fitted with a rubber gasket
permitting airtight fit of the cover
into the top of the cup. A porous
sintered stainless steel membrane
support disk is mounted in the
center of the cover. At the side
7-2
-------
Membrane Filter Equipment and its Preparation for Laboratory Use
of the beaker is a valve to which a
pumping device can be fitted. The
upper element is a stainless steel
funnel with about 500 ml capacity.
The assembled filter holding unit is
joined by a locking ring at the base
of the upper element. This engages
on three spring clamps on the covering
plate of the lower element.
Millipore "Sterifil" filter unit
(Plate 2, Figure 6)
A funnel and flask unit of poly-
carbonate with filter base and
support of polypropylene. Manu-
facturers tables should be referred
to regarding chemicals which may
be present in the sample and their
effect on the holder and flask
elements. This unit can be safely
sterilized under steam pressure.
FILTER HOLDING UNITS
FOR AQUEOUS SUSPENSIONS
FIG. 2
B
FIG. 6
FIG. 5
PLATE 2
4 Care and maintenance of filter holding
units
a Filter holding units should be kept
clean and free of accumulated
foreign deposits.
b Metal filter holding units should be
protected from scratches or other
physical damage which could result
in formation of surface irregularities.
The surfaces in contact with mem-
brane filters should receive
particular care to avoid formation
of shreds of metal or other
irregularities which could cause
physical damage to the extremely
delicate filters.
c Some filter holding units have
rubber components. The rubber
parts may in time become worn,
hardened, or cracked, necessitating
replacement of the rubber part
involved.
d The locking rings used in some
kinds of filter holders have two or
more small wheels or rollers
which engage on parts of the filter
holding assembly. Occasional
adjustment or cleaning is necessary
to insure that the wheels turn freely
and function properly. On some
units, the wheels are plastic, and
are not intended to turn. When
worn flat, they should be loosened,
turned a partial turn, and tightened
again.
Membrane Filters and Absorbent
Pads
1 The desired properties of membrane
filters have been discussed elsewhere.
Typical examples, commercially
available include:
a Millipore Filters, Type HA, white,
grid-marked, 47 mm in diameter
b S & S Type B-9, white, black-grid
mark, 47 mm diameter
-------
Membrane Filter Equipment and its Preparation for Laboratory Use
c Oxoid cellulose acetate membrane
filters, 4.7 cm, grid-marked
2 An absorbent pad for nutrient is a
paper filter disk, usually the same
diameter as the membrane filter.
Absorbent pads must be free of
soluble chemical substances which
could interfere with bacterial
growth. They should be of such
thickness that they will retain 1.8-
2. 2 ml of liquid culture medium.
During incubation of cultures on
membrane filters an absorbent pad
saturated with liquid culture medium
is the substrate for each filter.
Absorbent pads are supplied with
the purchase of membrane filters.
Additional absorbent pads may be
purchased separately. Sterilization
in an autoclave is recommended for
absorbent pads.
C Vacuum
Water can be filtered through a membrane
filter by gravity alone, but the filtration
rate would be too slow to be practical.
For routine laboratory practice, two
convenient methods are available for
obtaining vacuum to hasten sample filtration.
1 An electric vacuum pump may be used
connected to a filtration apparatus
mounted in a suction flask. The pump
need not be a high-efficiency type. For
protection of the pump, a water trap
should be included in the system,
between the filtration apparatus and
the vacuum pump.
2 A water pump, the so-called "aspirator"
gives a satisfactory vacuum, provided
there is reasonably high water pressure.
3 In emergency, a rubber suction bulb, a
hand pump, or a syringe, may be used
for vacuum. It will be necessary to
include some form of valve system to
prevent return flow of air.
D Culture Containers (Plate 3)
Most membrane filter cultures are
incubated in individual containers.
Almost any form of culture container
is acceptable if it is made of impervious
bacteriologically inert material. The
culture container should, or course, be
large enough to permit the membrane
filters to lie perfectly flat. The following
are widely used:
-1 Glass petri dishes
Conventional borosilicate glass culture
dishes are widely used in laboratory
applications of membrane filters. For
routine work, 60 mm X 15 mm petri
dishes are recommended. The common
100 mm X 15 mm petri dishes are
acceptable, but are subject to difficulties.
HYDRATOR
SUITABLE TYPES OF FORCEPS
ALCOHOL JAR
WITH
FORCEPS
GLASS PLASTIC
TYPES OF CULTURE CONTAINERS
PLATE 3
-------
Membrane Filter Equipment and its Preparation for Laboratory Use
2 Plastic petri dishes
Plastic containers have been developed
for use with membrane filter cultures.
Their cost is fairly low, and single-
service use feasible. They cannot be
heat-sterilized, but are supplied
sterile. They must be free from
soluble toxic substances. They can
either be loose-fitting or of a tight lid
to base friction fit.
E Other Equipment and Supplies Associated
with Sample Filtration
1 Suction flask (Plate 4)
a Most types of filter holding apparatus
are fitted in a conventional suction
flask for sample filtration. While
other sizes may be used, the 1-liter
size is most satisfactory.
b The suction flask can be connected
to the vacuum facility with thick-
walled rubber tubing. Latex rubber
tubing, 3/16" inside diameter, with
wall thickness 3/32", is suggested.
This tubing does not collapse under
vacuum, yet it is readily closed with
a pinch clamp.
c A pinch clamp on the rubber tubing
is a convenient means of cutting off
the vacuum from the suction flask
during intervals when samples are
not actually being filtered. It is
most convenient to have the vacuum
facility in continuous operation during
sample filtration work.
d In laboratories conducting a high
volume of filtration work, the
suction flask may be dispensed.
Filter-holding manifolds are available
to receive up to three filtration units.
The filtrate water is collected in a
trap (in series with the vacuum
source) which is periodically emptied.
e Another arrangement can be made
for dispensing with the suction flask.
In this case, the receptacle element
of the filtration unit is mounted in
the bench top. Instead of using a
suction flask, the lower element of
the filter holding unit has a dual
connection with the vacuum source
and with the laboratory drain. A
solenoid-operated valve is used to
determine whether the vacuum system
or the drain line is in series with
the filtration unit.
Ring stand with split ring (Optional)
(See Plate 4)
SUCTION FLASK RING STAND WITH SPLIT RING
PLATE 4
When the filter holding unit is
disassembled after sample filtration,
the worker's hands must be free to
manipulate the membrane filter.
Upon disassembly of the filter holding
unit, many workers place the funnel
element, inverted, on the laboratory
bench. Some workers, to prevent
bacterial contamination, prefer a rack
or a support to keep the funnel element
from any possible source of contam-
ination. A split ring on a ring stand is
a convenient rack for this purpose.
3 Graduated cylinders
In laboratory practice, 100 ml graduated
borosilicate glass cylinders are
satisfactory for measurement of
samples greater than 20 ml.
-------
Membrane Filter Equipment and its Preparation for Laboratory Use
4 Pipettes and cans
a Graduated Mohr pipettes are needed
for many procedures, such as
measurement of small samples, and
for preparing and dispensing culture
media. Pipettes should be available
in 1 ml and 10 ml sizes.
b Holding cans may be round or square
but must not be made of copper.
Aluminum or stainless steel are
acceptable.
5 Alcohol jar with forceps (Plate 3)
a All manipulation of membrane filters
is with sterile forceps. For steri-
lization, forceps are kept with their
tips immersed in ethanol or methanol.
When forceps are to be used, they
are removed from the container and
the alcohol is burned off.
b Forceps may be straight or curved.
They should be designed to permit
easy handling of filters without
damage. Some forceps have corru-
gations on their gripping tips. It is
recommended that such corrugations
be filed off for membrane filter work.
6 A gas burner or alcohol burner is needed
to ignite the alcohol prior to use of
forceps.
7 Dilution water
The buffered distilled water described
in "Standard Methods for the Examination
of Water and Wastewater" for bacterio-
logical examination of water is used
in membrane filter methods. Dilution
water is conveniently used in 99 + 2 ml
amounts stored in standard dilution
bottles. Some workers prefer to use
9. 0 + 0. 2 ml dilution blanks.
8 Culture medium
Bacteriological culture media used with
membrane filter techniques are dis-
cussed at length in another part of this
manual.
F Incubation Facilities
1 Requirements
a Temperature
For cultivation of a given kind of
bacteria, the same temperature
requirements apply with membrane
filter methods as with any other
method for cultivating the bacteria
in question. For example, incu-
bation temperature for coliform
tests on membrane filters should
be 350C + 0.50C.
b Humidity
Membrane filter cultures must be
incubated in an atmosphere main-
tained at or very near to 100%
relative humidity. Failure to
maintain high humidity during
incubation results in growth failure,
or at best, in small or poorly
differentiated colonies.
2 The temperature and humidity require-
ments can be satisfied in any of
several types of equipment.
a A conventional incubator may be
used. With large walk-in
incubators, it is extremely difficult
to maintain satisfactory humidity.
With most conventional incubators,
membrane filter cultures can be
incubated in tightly closed con-
tainers, such as plastic petri dishes.
In such containers, required
humidity conditions are established
with evaporation of some of the
culture medium. Because the
volume of air in a tightly closed
container is small, this results in
negligible change in the culture
medium. If glass petri dishes or
other loosely fitting containers are
used, the containers should be
placed in a tightly closed container,
with wet paper or cloth inside to
obtain the required humidity con-
ditions. A vegetable crisper, such
as used in most home refrigerators,
7-6
-------
Membrane Filter Equipment and its Preparation for Laboratory Use
is useful for the purpose.
Plate 3^
(See
III
A covered water bath maintaining
44. 5°C + 0. 2°C is necessary for
the fecaTcoliform test and this
will necessitate the use of
a water-bath having forced
circulation of water.
STERILIZATION OF MEMBRANE FILTER
EQUIPMENT AND SUPPLIES
A Filter Holding Unit
1 When is sterilization necessary?
a The filter holding unit should be
sterile at the beginning of each
filtration series. A filtration series
is considered to be interrupted if
there is an interval of 30 minutes
or longer between sample filtrations.
After such interruption any further
sample filtration is treated as a
new filtration series and requires
a sterile filter holding unit.
b It is not necessary to sterilize the
filter holding unit between successive
filtrations, or between successive
samples, of a filtration series.
After each filtration the funnel walls
are flushed with sterile water to
free them of bacterial contamination.
If properly done, the flushing pro-
cedure will remove bacteria
remaining on the funnel walls and
prevent contamination of later
samples.
2 Methods for sterilization of filter
holding unit
a Sterilization in the autoclave is
preferred. Wrap the funnel and
receptacle separately in Kraft paper
and sterilize in the autoclave 15
minutes at 121°C. At the end of
the 15 minutes holding period in the
B
autoclave, release the steam
pressure rapidly, to encourage
drying of the filter holding unit.
b The unit may be sterilized by
holding it 30 minutes in a flowing
steam sterilizer.
c The unit may be immersed 2 to 10
minutes in boiling water. This
method is recommended for
emergency or field use.
d Some units (Millipore Stainless Unit)
are available with accessories per-
mitting anhydrous sterilization with
formaldehyde. The method consists
of introduction of methanol into a
wick or porous plate in the sterili-
zation accessory, assembly of the
filter holding unit for formaldehyde
sterilization, ignition of the
methanol, and closure of the unit.
The methanol is incompletely
oxidized in the closed container,
resulting in the generation of
formaldehyde, which is bactericidal.
The filter holding unit is kept closed
for at least 15 minutes before use.
e Ultraviolet lamp sterilizers are
convenient to use. A device now
commercially available for ultra-
violet sterilization of membrane
filter funnel units.
Sterilization of Membrane Filters and
Absorbent Pads
1 Membrane filters
a Membranes are supplied in units
of 10 in kraft envelopes, or in
packages of 100 membranes. They
may be sterilized conveniently in
the packets of 10, but should be
repackaged if supplied in units of
100. Large packages of filters can
be distributed in standard 100 mm
X 15 mm petri dishes, or they can
be wrapped in kraft paper packets
for sterilization.
7-7
-------
Membrane Filter Equipment and its Preparation for Laboratory Use
b Sterilization in the autoclave is
preferred. Ten minutes at 121° C
or, preferably, at 116°C is
recommended. After sterilization
the steam pressure is released as
rapidly as possible, and the filters
are removed from the autoclave and
dried at room temperature. Avoid
excessive exposure to steam.
c In emergency, membrane filters may
be sterilized by immersion in boiling
distilled water for 10 minutes. The
filters should first be separated from
absorbent pads and paper separators
which usually are included in the
package. The boiling water method
is not recommended for general
practice, as the membranes tend to
adhere to each other and must be
separated from one another with
forceps.
2 Absorbent pads for nutrient
a Unsterile absorbent pads can be
wrapped in kraft paper or stacked
loosely in petri dishes, and auto-
claved with membrane filters (ten
minutes or longer at 121°C or 116° C).
b After sterilization absorbent pads
for nutrient should be dried before
use.
C Glassware
1 Sterilization at 170° C for not less than
1 hour is preferred for most glassware
(pipettes, graduated cylinders, glass
petri dishes). Pipettes can be sterilized
in aluminum or stainless steel cans, or
they may be wrapped individually in
paper. The opening of graduated
cylinders should be covered with paper
or metal foil prior to sterilization.
Glassware with rubber fittings must not
be sterilized at 170° C, as the rubber
will be damaged.
2 Sterilization in the autoclave, 15
minutes at 121°C, is satisfactory,
and preferred by many workers. When
sterilizing pipettes it is important to
7-8
exhaust the steam pressure rapidly
and vent the containers momentarily.
This allows the vapor to leave the can
and prevents wet pipettes.
D Culture Containers
1 Glass petri dishes
a Petri dishes may be sterilized in
aluminum or stainless steel cans,
or wrapped in kraft paper or metal
foil. They can be wrapped
individually or, more conveniently,
in rolls of up to 10 dishes.
b Preferably, sterilize glass petri
dishes at 170° C for at least 1 hour.
c Alternately, they may be sterilized
in the autoclave, 15 minutes at
121° C. After sterilization steam
pressure should be released rapidly
to facilitate drying of the dishes.
Other suggested methods for
sterilization of plastic dishes
include exposure to ethylene oxide
vapor (0. 5 ml ethylene oxide per
liter of container volume), or
exposure to ultraviolet light.
Ethylene oxide is a dangerous
chemical being both toxic and
explosive, and it should be used
only when more convenient and
safer methods are not available.
2 Plastic culture containers
a Because of the thermo-labile
characteristics of the plastic,
these containers cannot be heat
sterilized. Manufacturers supply
these in a sterile condition.
b For practical purposes, plastic
dishes may be sterilized by
immersion in a 70% solution of
ethanol in water, for at least 30
minutes. Dishes must be allowed
to drain and dry before use, as
ethanol will influence the perform-
ance of culture media.
This outline was prepared by H. L. Jeter,
Chief, Program Support Training Branch,
USEPA Cincinnati, Ohio 45268
Descriptors- Laboratory Facilities,
Bacteriology, Microbiology, Enteric
Bacteria, Filters, Membrane
-------
MEMBRANE FILTER EQUIPMENT FOR FIELD USE
I INTRODUCTION
One of the most troublesome problems in
bacterial water analysis is the occurrence
of changes in the bacterial flora of water
samples between the time of sample collectior
and the time the actual bacterial analysis is
started. Numerous studies have b^^n made
on this problem. From these have come such
recommendations (Standard Methods, 10th ed)
as holding the sample at 0-10 C and starting
laboratory tests as soon as possible after
collection of the sample. Recommendations of
Standard Methods, 12th Edition, was to hold
the sample as close as possible to the temper-
ature of the source and to start the laboratory
tests preferably within 1 hour and always with-
in a maximum of 30 hours after collection.
Changes in the 13th edition of Standard Meth-
ods (1971) will again call for the icing of
samples, and further, that samples of envir-
onmental waters be held for not more than 8
hours total elapsed time before samples are
plated or used for microbiological testing.
Tbe 3C hour maximum elapsed sample holding
time will still be retained for potable water
samples.
The purpose of this discussion is to
introduce some of the portable equipment
which has been developed and to point out
noteworthy features of each. Actual
practice and experience with these units
reveal strong point and weaknesses in
each type.
The membrane filter method has been
accepted by the Federal Government for
the bacteriological examination of water
under its jurisdiction. This acceptance
was based on methods developed and
procedures applied in fixed laboratories.
While the use of field kits is not excluded,
no special concession has been made
regarding the standards of performance
of membrane filter field kits. Thus, in
planning to use a membrane filter field
kit for the bacteriological examination of
water, it is the responsibility of the
individual laboratory to establish beyond
reasonable doubt, by comparison with
Standard Methods fermentation tube tests
or established laboratory membrane
filter methods, the value of use of the
membrane filter field kit in determining
the sanitary quality of water supplies
examined.
A They would be useful in certain routine
water quality control operations. Examples
include such places as on board ships;
some airlines, particularly in overseas
operations; and some national parks. In
each example, it is seen that there is an
obvious difficulty in getting water samples
to the examining laboratory in time for
early examination.
B In addition, such units would be invaluable
in emergencies when existing laboratories
are overburdened or inoperative. Portable
kits already have proven extremely helpful
in testing many small water supplies in a
short period of time. Further there is a
predictable need for such equipment in the
event of a wartime civil defense disaster.
Experience of the Germans in the vicinity
of Hamburg during World War II lends
support to this concept.
II TYPES OF COMMERCIALLY AVAILABLE
MEMBRANE FILTER EQUIPMENT FOR
FIELD USE
A Sabro Water Laboratory
This unit represents a fixed membrane
filter laboratory in miniature, with
adaptations for special situations to be
encountered in the field. Notable features:
1 The funnel unit supplied on older units
is glass. A newer model has been
released with an all-metal funnel unit.
2 The vacuum source is a hand pump
(modified bicycle pump) or, optionally,
an all-metal syringe.
3 The manufacturer sells prepared
ampouled medium, in a liquid state.
The medium should be kept at a cool
MOTE- Vientiou of commercial products and manufacturers does not imply endorsement by the
(-•"ede^al Water Pollution Control Administration
W. BA. mern. 63g. 8. 77 Q ,
=> o-l
-------
Membrane Filter Equipment for Field Use
temperature, out of the light. Its
useful shelf life is uncertain, but
limited tests by this agency indicate
that the medium performs accept-
ably with storage up to one year.
4 Incubation of the cultures is in an
incubator drawer having a capacity
of 18 1-ounce culture containers, or
36 plastic containers, and operates
electrically at 110V, and with suit-
able converters, at 6V, 12V. A
battery also can be used with this
unit.
5 Sterilization of the funnel unit is
carried out by a "light flaming
technique" or, optionally, by
immersion of the funnel unit in
hot or boiling water.
6 Useful accessories provided include
thermometer, alcohol lamp, measur-
ing cup, and forceps.
B Millipore Field Monitor Units
These units differ radically from any
other field equipment that has appeared.
Significant reductions in bulk of equip-
ment have been brought about through
major changes in function and design of
the usual equipment. Although not re-
commended for valid data due to
differences in quantitation when compared
to standard test procedures, the unit is
useful for rapid field testing to establish
"ball park" fiqures for later testing with
approved test procedures. Notable features:
1 The funnel unit has been eliminated in
its usual form. This has been done by
development of a carefully fitted, single -
use combination filtration unit and culture
container. This feature eliminates most
of the handling and use of accessory
equipment.
2 The vacuum source is an all metal syringe
with a fitting providing for direct connec-
tion to the culture container.
container. Alternately, the manufacturer
makes available a delayed-incubation
medium in the ampoules. Other culture
media can be used at the discretion of the
user, but some difficulty can be anticipated
in introducing the medium without special
equipment.
4 Incubation of the cultures is provided in the
field through use of an associated portable
incubator and equipment carrying kit. This
incubator has room for about 25 cultures.
It is electrically operated, and through
selection of available switching positions,
operates at 6V, 12V, 110V, or at 220V.
5 Sterilization of components in the field is
unnecessary. The culture containers and
plastic tubes are single-use units supplied
in a sterile condition. Samples do not come
in contact with the syringe until after they
have passed through the filter.
C Millipore Field Unit for Military Use
1 A modified Millipore field unit based on the
case and incubator described in B, 4 above,
has been adopted by the U.S. Department
of Defense. This unit includes a miniaturized
stainless metal funnel unit instead of the
Monitors.
2 The vacuum source is an all-metal syringe.
3 Sterilization of funnel unit is by formaldehyde
generated through incomplete combustion of
methyl alcohol.
Ill
COMMON DIFFICULTIES ENCOUNTERED IN
COMMERCIALLY AVAILABLE FIELD
B
C
D
The culture medium provided by the manu-
facturer includes M-Endo Broth, MF, ready-
to-use, in glass ampoules. These ampoules
are so designed as to permit easy introduc-
tion of the culture medium into the culture
The most conspicuous problem arising with field
use of most units is their ultimate reliance on a
fixed laboratory for essential supplies.
These portable laboratories will permit simulta-
neous incubation of up to 30 membrane filters.
For protracted field work, a fairly large amount
of reserve supplies and equipment will be neces-
sary. Such a reserve would include culture media.
membrane filters, culture containers, fuel, and
other expendable supplies required in the field,
organized in a supplementary carrying case.
No currently available field unit provides illumi-
nation or optical assistance for interpretation
~>f results.
8-2
-------
Membrane Filter Equipment for Field Use
Some of the sterilization methods
recommended by manufacturers are
unacceptable. If field sterilization in
boiling water is needed, then there must
be a heat source and a metal can or beaker.
Such equipment could be carried in the
case suggested in C above.
IV IMPROVISED FIELD EQUIPMENT
The initial cost of most of the commercially
manufactured units has met some objection,
This factor, coupled with need for additional
accessory supplies and equipment, has
aroused interest in improvised units. Such
a unit could consist largely of equipment
normally used in a fixed laboratory, packagec
in one or two fiberboard cases.
A The funnel unit could be one of the
familiar stainless steel units used in
many laboratories; or it could be specially
designed, smaller than ordinarily used,
permitting use of up to a dozen or more
upper filter holding elements in the field.
B The vacuum source could be the modified
bicycle pump (leathers reversed), and
provided with a by-pass valve. The suction
flask could be the standard side-arm glass
flask, or a metal unit could be devised.
C M-Endo Broth or LES Endo agar are
suitable media. Both are available as
dehydrated medium which must be
reconstituted and boiled in the field.
M-Endo Broth MF is now available in
liquid form, sterile, in sealed ampules.
A shelf life of approximately one year is
stated when stored under moderate
temperatures in the dark.
LES MF Holding Medium - Coliform
requires merely dissolving in distilled
water. No heating is necessary. Such
medium would be an advantage where
D Sterilization of funnel units, graduated
cylinders, media, etc., would be through
immersion in boiling water for 2 minutes
or longer, as indicated for the material
being sterilized. Provision for boiling
water is easy through use of a small camp
stove or other simple burner.
Improvised equipment, such as discussed
above, would have great usefulness in
emergencies, where commercially available
membrane filter field units are not on hand.
V In a separate outline are detailed
descriptions of procedures for use of
commercially available membrane filter
field equipment. In some cases the
suggested methods are different from
those recommended by the manufacturers.
In each case such departures are based
on a series of experimental studies m^de
by this agency, which suggested need
for modification of existing recommend-
ations.
REFERENCE
1 Laubusch, E. J. What You Should Know
About the Membrane Filter, Public
Works, 89: 106-13, 162-68, 1958.
This outline was prepared by H. L.
Jeter, Chief, Program Support
Training Branch, USEPA, Cincinnati,
Ohio 45268
Descriptors: On-Site Laboratories,
Field Investigations, Filters,
membrane, Bacteria, Microorganisms,
Laboratory Tests
-------
PRINCIPLES OF CULTURE MEDIA FOR USE WITH MEMBRANE FILTERS
I INTRODUCTION
A Many kinds of membrane filter media
have been described for use in bacterio-
logical tests on water. This is noteworthy
in view of the relatively few years the
filters have been widely available in this
courtry. This discussion is to consider
several of these media in terms of their
purposes, composition, and the ways in
which they are used.
B Basic Considerations
1 Filtration of water sample through a
membrane filter results in deposition
of bacteria and particles of suspended
matter on the filter surface. The
bacteria can be cultivated in place if
suitable culture medium is made avail-
able for their growth.
2 The bacteria are cultivated by placing
the membrane on a pad of absorbent
paper saturated with liquid culture
medium, or on an agar medium. The
culture medium diffuses through the
pores of the filter, and is available to
the bacteria on the opposite surface.
Proper time, temperature, and humidity
of incubation results in development of
bacterial colonies. In principle, each
bacterial cell multiplies to become a
single bacterial colony.
3 Some culture media, satisfactory for
tube cultures or agar plate cultures,
do not perform well when used with
membrane filters due to a selective
adsorptive property of the filter itself.
In the process of diffusion through the
pores some components of the culture
medium may be removed completely,
or reduced in concentration. Thus, the
composition of a given culture medium
at the filter surface where it is avail-
able for bacterial growth may be dif-
ferent from its composition beneath the
membrane filter.
There is evidence that improved cul-
tural results sometimes are obtained
with increased concentration of certain
nutritive constituents of membrane filter
culture media.
Pure cultures may be recovered
from membrane filters and subject-
ed to supplementary biochemical,
cultural, and serological procedures
for identification studies or for veri-
fication of interpretations based on
direct observation of membrane
filter cultures.
The same use can be made of agar plat-
ing media; however the membrane filtei
offers advantages due to the ability to
concentrate organisms from a large
volume of sample in which the organ-
isms are present in low density.
C Applications of Membrane
Media
The composition of bacteriological culture
media designed for tube or plate cultures
should be subjected to critical study be-
fore they are applied to membrane filter
procedures. Media based on well-known
bacteriological media have been modified
for use with membrane filters for the
following purposes in testing water.
1 Bacterial plate counts
2 Media for bacterial indicators of
pollution
a Coliform organisms
b Fecal streptococcus gro:m
c Clostridium perfringens
3 Salmonella and other enteric bacterial
pathogens
NOTE: Mention of commercial rroducts and manufacturers does not imply endorsement by
che FWPCA ,: id the U.S. Department of the Inte
W. BA. IT.P —. 681.1. 7«
-------
Principles of Culture Media
D Constituents of Membrane Filter Culture
Media
Membrane filter media for the differentia-
tion and counting of special groups of
bacteria are based on the same principles
used in differential agar plate media. Thus
the components of a differential medium
for membrane filter cultures include:
1 Substances favoring growth of the
organisms for which the medium is
designed. Inclusion of special peptones,
fermentable carbohydrates, yeast or
meat extracts, water and chemicals to
adjust pH to a desired level are common
methods of favoring growth of desired
organisms.
2 Differential indicator system. The
purpose of the indicator system is to
produce characteristic colonies of the
desired bacterial groups for easy
recognition when present in a mixture
with extraneous types of colonies. This
is done through inclusion of (a), a com-
ponent which is chemically changed by
the organisms to be differentiated, and
(b), indicator substances, which give
visible evidence of an intermediate or
end product resulting from a chemical
change of substance (a).
J Selective inhibitors. Some bacterial
groups to be tested may be over-
whelmingly outnumbered by extraneous
types of bacteria. In such cases, it is
necessary that substances be included ir
the medium which (a), prevent growth
of a maximum number of kinds of ex-
traneous bacteria, and (b), have mini-
mum adverse affect on growth of the
kind of bacteria for which the medium
is designed.
i: Variety of Methods of Using Media Avail-
able with Membrane Filter Methods
1 Single-stage tests
After sample filtration, the membrane
filter is placed on a designated culture
medium, and left there throughout the
incubation period. The culture results
arc examined and interpreted directly.
2 Multi-stage tests
A membrane filter can be transferred
from one culture medium to another
without disturbance of bacteria or
colonies on the filter. This is unique
with membrane filter methods, and
lends itself to a variety of cultural and
testing procedures.
a The membrane filter, after sample
filtration, can be incubated for a
specified time on one medium, thi i,
transfered to a second medium. The
method permits initiation of grow;.':
on enrichment medium, after w^'^1-
the membrane filter can be trans-
ferred to a less productive medium.
With growth already begun, some
differential culture media give belter
quantitative production than would
be the case without preliminary
incubation.
b After incubation on one or more
media, colonies on the membrane
filter can be subjected to bio-
chemical tests with reagents too
toxic to include in the culture
medium. Such reagents may be
flooded over the growth on the
filter, or the filter may be placed
on an absorbent pad saturated with
the reagent, in order to make such
tests.
c A third type of multi-stage test is
one in which the membrane filter,
after sample filtration, is placed
temporarily on a medium containing
a bacteriostatic agent. In the
presence of such a substance,
bacterial growth is inhibited or
slowed greatly, but the organisms
are not killed. During a limited
period, the membrane filters may
be transported or stored at ambient
temperatures. The filter can be
transferred later to a suitable
medium and incubated for develop-
ment of colonies.
-------
Principles of Culture Media
II CULTURE MEDIA FOR TOTAL BACTE-
RIAL COUNTS ON MEMBRANE FILTERS
A Concepts
1 Strictly, a "total" bacterial count
medium is nonexistent. No single
medium and incubation procedure can
provide simultaneously the full range
of oxygen requirements, needs for
special growth substances, pH require-
ments, etc. , of all the kinds of bacteria
found in water.
2 Actually, "total" bacterial counts are
counts of the bacteria developing visible
colonies on a defined culture medium
at a known pH after incubation for a
set time and temperature under aerobic
conditions.
3 Within the foregoing limitations, the
following criteria offer a useful basis
for selection of membrane filter media
to be used for estimates of the bacterial
density in water.
a The medium and its method of use
should produce a maximum number
of colonies from all types of water.
The colony yield should compare
favorably with the bacterial counts
determined as described in "Stand-
ard Methods for the Examination of
Water, Sewage and Industrial
Wastes. " 13th Ed. (1971).
b The colonies should develop rapidly
to a sufficient size to be counted
after a minimum incubation period.
At present, best results with mem-
brane filter methods are obtained
after about 18 hours incubation.
c The medium should be one which is
reproducible and routinely available
in laboratories.
B Composition of Total Count Media for
Membrane Filters
Almost any rich, general growth promoting
culture medium is acceptable for total bac-
terial counts on membrane filters. Several
such media have been suggested especially
for membrane filter methods. These differ
only in minor aspects, and can be discuss-
ed as a group. For details of composition
and specific applications of each, see the
media formulations elsewhere in this
i Growth promoting substances: All the
substances included in these media are
included to encourage growth of a maxi-
mum number of kinds of bacteria. MOM.
workers agree that the peptone should he
used in twice the concentration usually
found in conventional tube or agar plat-
ing media.
2 Indicator substances are unnecessary
with total count media.
3 Substances for the selective inhibition
of certain bacterial groups are not in-
cluded in total count media.
Problems Encountered with Total Count
Media
Bacterial colonial growth habits on mem-
brane filters are similar to their surface
growth habits on similar agar plate media.
1 As with agar plate media, some species
of bacteria grow continuously, spreaa-
ing over the surface of a membrane
filter, tending to obscure nonspreading
colonies which otherwise could be
counted.
2 Some samples contain an appreciable
amount of particulate matter. In sam-
ple filtration, this is deposited on the
surface of the membrane filter with the
bacteria. When the culture medium
diffuses through the filter, a capillary
film of liquid culture medium accumu-
lates around the particles of extraneous
matter. Bacteria not ordinarily con-
sidered "spreaders" sometimes develop
confluent colonies due to the film of
liquid medium accumulating around such
particles.
-------
Principles of Culture Media
D What is the best total count medium for
use with membrane filters?
Because of the relative ease of preparation,
most workers prefer the commercially
prepared dehydrated media. Difco M-
Enrichment Broth (B 408) or Baltimore
Biological Laboratories' M-Enrichment
Broth (No. 331) are used interchangeably.
Total colony productivity of these media
is equivalent to that of media prepared
from the individual components.
Ill CULTURE MEDIA FOR TOTAL COLIFORM
TESTS ON MEMBRANE FILTERS
A Concepts
The nature of membrane filter culture
methods imposes a different definition of
coliform bacteria than the Standard Methods
definition.
1 Standard Methods fermentation tube
method. "The coliform group includes
all of the aerobic and facultative anaero-
bic Gram-negative nonsporeforming
rod-shaped bacteria which ferment
lactose with gas formation within 48
hours at 35°C. "
2 Membrane Filter Methods: "in the
membrane filter procedure, all organ-
isms that produce a colony with
a metallic sheen in 22-24 hours are
considered members of the coliform
group. The sheen may appear as a
small central focus or cover the
entire colony. " The guiding prin-
ciple is that any amount of sheen
is considered positive.
3 The Standard Methods definition of
coliforms requires demonstration of
the ability of organisms to produce gas
through the fermentation of lactose.
The membrane filter method does not
lend itself to the demonstration of gas
production. It relies instead on the
development of a particular type of
colony on an Endo type of culture
medium. The culture medium is one
in which lactose, basic fuchsin, and
sodium sulfite comprise an indicator
system to cause differentiation of
coliform colonies. While the bacterial
groups measured by membrane filter
methods are not identical with the gro if
measured by Standard Methods pro-
cedures, they are believed to be es-
sentially the same, and to have equal
sanitary significance.
B Composition of Coliform Media for
Membrane Filters
Several different media have been suggested
for coliform tests on membrane filters.
The components of these media can be
classed into three convenient groups for
general considerations.
1 Growth-promoting substances. Growth
of bacteria on all the media is favored by
the inclusion of such components as
peptones (as Neopeptone, Thiotone Casi-
tone, Trypticase, and other proprietary
peptones), yeast extract, dipotassium
phosphate (for adjustment of reaction of
the medium), and distilled water. Lac-
tose is included in all these media. It
serves doubly, to favor growth of coli-
form bacteria, and as an essential compo-
nent of the systems for differentiating
coliform colonies.
2 Two kinds of differential indicator
systems are available for demonstration
of lactose fermentation on membrane
filters.
a Lactose-basic fuchsin-sodium sulfite
system (Endo type media).
1) Media UGing this oy LI tern include
lactose and a suitable concentra-
tion of basic fuchsin which has been
partially decolorized with sodium
sulfite.
2) The basic fuchsin-sodium sulfite
complex requires very careful
standardization. An excess of
either component results in an
unsatisfactory culture medium.
3) The indicator system demonstrates
lactose fermentation as follows:
d-4
-------
Principles of Culture Media
a) The coliform bacteria produce
aldehyde as an intermediate
product of the fermentation of
lactose.
b) The aldehyde is "complexed"
by the sodium sulfite-basic
fuchsin indicator. In this pro-
cess a reaction occurs in which
red color is restored to the
basic fuchsin. Colonies of
bacteria fermenting lactose
assume the color of the re-
stored fuchsin. As the restored
dye accumulates,, it apparently
precipitates on the colony,
giving the colony a character-
istic green-gold surface sheen.
The reaction occurs best in an
alkaline medium. The culture
medium is adjusted to pH 7. 5.
4) Endo type media require very
careful standardization for suc-
cessful use in the laboratory.
Most workers prefer to use a
commercially prepared and stan-
dardized medium. M-Endo Broth
MF is the recommended coliform
medium for use with membrane
filters.
b pH indicator system
1) Media using this system rely on
detection of pH change due to the
accumulation of organic acids,
end products of lactose
fermentation.
2) Bromcresol purple, for example,
is a pH indicator, approaching
yellow at more acid pH. Colon-
ies fermenting lactose and ac-
cumulating organic acids there-
fore turn yellow.
3) Studies in England with r-u'r»
filters for coliform tests havt
been based on a modification of
MacConkey's Medium, using this
principle of colony differentiation
in coliform tests.
Inhibitory substances in membrane
filter coliform media
a Confusing and erroneous results in
coliform detection can bp caused by
1) the overgrowtn of the membrane
filter by extraneous nonlactose
fermenting bacteria, preventing
coliform colonies from developing
the characteristic color and
sheen; and
2) the development of sheen
colonies of lactose fermenting
bacteria which produce acid but
not gas in the fermentation of
lactose.
b These difficulties car ' < •
through incorporating ot substances
harmless to coliform bacteria but
which have inhibitory effect on growth
of extraneous forms. Attention must
be given to the concentration of such
substances, as excessive amounts
also will reduce the productivity of
the medium for coliform colonies.
The following components of various
culture media have proven useful in
suppressing growth of noncoliform
bacteria on membrane filters.
1) Basic fuchsin-sodium sulfite. Al-
though these compounds are included
in Endo-type media for their role
in differentiating coliforms from
other types of colonies, they are
effective in preventing the growth
of many of the noncoliform bacteria
occurring in water samples.
2) Ethanol (95%. . . NOT denatured) is
included in M-Endo Broth MF. In
the concentration used, ethanol
suppresses growth of some kinds
of noncoliform bacteria, and tends
to limit the colony size of others.
In addition, the ethanol seems to
increase the solubility of some of
the other components of the media.
-------
Principles of Culture Media
Sodium desoxycholate or bile salts
are used in such media as M-Endo
Broth MF, and in the modified
MacConkey's Medium for membrane
filters used in British studies. They
are included primarily for their in-
hibitory effect against Gram-positive
cocci and spore formers.
C Methods Available for Using C»jUi".:v.
Media with Membrane Filters
1 Single-stage coliform tests
a After sample filtration, the mem-
brane filter is incubated for the
desired time on a selective coliforrr
differentiating medium.
b The coliform colonies are counted
without further tests.
c M-Endo Broth MF and LES Endo
Agar Media are alternate standard
single-stage coliform media.
2 Two-stage coliform tests
a Immediate coliform test
1) After sample filtration, the mem
brane filter is incubated If - 2
hours on the enrichment medium
of lauryl tryptose broth.
2) The membrane is then transi'erivi
to a new absorbent pad saturated
with the standard differential
medium for coliform bacteria,
and incubated for 20-22 hours at
35 + 0.5C.
3) The coliform colonies are counted
without further tests.
4) This test procedure, based on
EHC Endo Medium, was described
in the 10th edition of Standard
Methods. With the 12th edition,
an official two-stage coliform test
has been adopted, based on LES
Fnda Aear Medium.
b Delayed incjbation coliform tes
1) After sample filtration, the mem-
brane filter is placed on an ab-
sorbent pad saturated with benzo
ated Endo Medium or with LES
Holding Medium. The filter ma.
be preserved up to 72 hours at
ambient temperatures. During
this time it can be transported o
stored, Growth is stopped or
greatly reduced.
2) The membrane filter can be
transferred to a fresh absorbent
pad, saturated with such a mediuai
as M-Endo Broth MF, or to LES
Endo Agar and incubated up to 24
hours.
3) The differentiated coliform
colonies are counted as with
other membrane filter coliform
media.
4) This test procedure makes it
possible to filter samples in the
field, place the filters on pre-
servative medium, then mail or
transport them to the laboratory
for completion of the bacteriologi-
cal examination. The procedure
is designed to eliminate the neec
for maintaining sample tempera-
ture in the interval between
sample collection and initiation
of the bacteriological examination.
In addition, the method should
produce results more nearly
reflecting the quality of the sour* e
water than is available with other
methods of collecting and testing
samples.
Verified membrane filter coliform
test
a This is used to verify the interpreta-
tion of differentiated colonies on an\-
type of membrane filter coliform
medium. The test is suggested for:
self-training of laboratory workers.
for evaluation of new or experiment
-------
Principles of Culture Media
IV
media, and in any water examinatior
in which the interpretation of results
is in doubt or likely to be involved in
legal controversy.
The test consists of obtaining pure
cultures from differentiated coliform-
like colonies on membrane filters,
and subjected them to further cultural
and biochemical tests to establish
their identity as Gram-negative non-
sporeforming bacilli which ferment
lactose with gas production. The
technical procedures are described
elsewhere in this manual.
MEDIUM FOR THE FECAL COLIFORM
TEST
A Concepts
The selective effect of elevated temperature
has been the most important development
in fecal coliform tests since 1904. In that
year, Eijkman discovered that coliform
bacteria from the gut of warm-blooded
animals produced gas from glucose at
46°C, while the majority of coliform
bacteria from other sources did not.
Media variations were of only
secondary importance.
Much medium variation has resulted from
attempts to select for Escherichia coli,
only, as the fecal coliform. While E. coli
is usually the predominant coliform in
human (and animal) feces, other types
are present, including the alleged soil
and plant coliform, Aerobacter
aerogenes in very large numbers.
All coliforms demonstrated by isola-
tion to have arisen in feces are called here
fecal coliforms and are measured empir-
ically by the fecal coliform tube test.
Membrane filter tests reflect divergence
of attitude on indicators of fecal origin.
Delnney et al. (1962) have published:
Measurement of E. coli Type I by the
Membrane Filter?" Geldreich et al. (1965)
have presented: Fecal-Coliform-Organism
Medium for the Membrane Filter Technique
Temperatures are the same but media are
different. Because the fecal coliform test
appears more convenient, it will be
emphasized.
B Composition of Fecal Coliform Medium
MFC
1 MFC medium is a rich growth medium
containing lactose, proteose peptone
no. 3, tryptone and yeast extract. A
level of .3% sodium chloride produces
favorable osmotic balance. Vigorous
growth results. A practical result is
shortening of test time to 24 hours.
The growth constituents are similar to
those of the tube test for fecal coliform.
Both have 0.15% bile salts to select for
coliforms but elevated temperature is
the more important selective factor.
2 The indicator system of aniline blue
results in blue fecal coliform colonies.
Nonfecal coliform colonies, generally
few, are gray to cream-colored.
C Special Problems with MFC Broth Medium
1 Temperature control must be accurate.
Current recommendations call for
44.5 + 0.2 C and the temperature to
be maintained in a water incubator
of forced circulation.
2 Temperature equilibration must be
rapid. Nonfecal coliforms may initiate
growth at lower temperatures and sub-
sequently give false positive blue
colonies when incubated at 44. 5° C.
No more than 20 minutes lapse of time
is recommended from filtration to
incubation. Submergence in waterproof
plastic bags reduces actual temperature
equilibration to 10 - 12 minutes.
3 Rosolic acid presents some problems
in preparation. It is practically
insoluble in water and of limited
stability in alkaline solution. A 1%
solution in 0. 2 N NaOH should be
prepared and this added to the medium
as recommended by the manufacturer.
9-7
-------
Principles of Culture Media
V MEDIA FOR FECAL STREPTOCOCCUS
TESTS
A Introduction
The selective component of KF Agar
is sodium azide, used in 0.04%
concentration.
1 The development of membrane filter
culture media for the fecal streptococci
reflects the continuing interest in this
group of bacterial indicators of pollution,
The productivity of enterococcus media
recently has been greatly increased.
2 Standards of performance of a good
fecal streptococcus medium correspond
with those of a good coliform medium
on membrane filters. Thus, the
requirements of productivity, specificit;
ease of use, and reproducibility of the
medium, are equally applicable to
medium for the detection and enumera-
tion of the fecal streptococci.
B KF Agar
1 This streptococcus medium was
developed at SEC by Kenner, et al.
and designated KF agar.
While KF medium is productive for
detection and enumeration of the
Fecal Streptococcus Group, its use
is hampered by nonspecificity. Studies
have demonstrated that the medium
supports growth of S>. bovis, and other
forms common in animals, but not
numerous in the fecal excreta of humans.
2 Composition of KF Agar
a Nutritive requirements of the fecal
streptococci are supplied by peptone,
yeast extract, sodium glycerophos-
phate, maltose, lactose, and dis-
tilled water.
b The indicator system is phenol red
and 2, 3, 5 triphenyl tetrazolium
chloride. On KF Agar used with
membrane filters, the fecal strepto-
coccus colonies develop as small
colonies, up to 2 mm in diameter,
colored various shades from pale
pink to a dark wine color.
REFERENCES
1 APHA, AWWA, FSIWA. Standard Methods
for the Examination of Water and Waste-
water. 12th Edition.
2 Clark, H. F., Geldreich, E. E., Jeter, H. L.
and Kabler, P. W. The Membrane Filter
in Sanitary Bacteriology. Public Health
Reports. 66:951-77. 1951.
3 Fifield, C.W. and Schaufus, C. P. Improved
Membrane Filter Medium for the Detection of
Coliform Organisms. J. Amer. Water Works
Assn. 50:2:193-6. 1958.
4 Geldreich, E. E., Clark, H.F., Huff, C. B.
and Best, L. C. Fecal-Coliform-Organism
Medium for the Membrane Filter Technique.
J. Amer. Water Works Assn. 57:2:209-14.
1956.
5 Kabler, P.W. and Clark, H.F. Use of
Differential Media with the Membrane Filter.
Am. Jour. Pub. Health. 42:390-92. 1952.
6 Kenner, Bernard A. , Clark, H. F. and
Kabler, P.W. Fecal Streptococci.
I. Cultivation and Enumeration of
Streptococci in Surface Waters.
Applied Microbiology. 9:15-20. 1961.
9-8
-------
Principles of Culture Media
7 McCarthy, J. A. , Delaney, J. E. and
Grasso, R. J. Measuring Coliforms
in Water. Water and Sewage Works
108:238. 1961.
8 Rose, R. E. and Litsky, W. Enrichment
Procedure for Use with the Membrane
Filter for the Isolation and Enumeration
of Fecal Streptococci in Waters
Microbiol. 13:1:106-9. 1965.
9 Slanetz, L.W., Bent and Bartley. Use
of the Membrane Filter Technique to
Enumerate Enterococci in Water.
Public Health Reports. 70:67. 1955.
10 Slanetz, L. W. , Bartley, C. H. and Ray,
V.A. Further Studies on Membrane
Filter Procedures for the Determination
of Numbers of Enterococci in Water and
Sewage. Proc. Soc. Am. Bacteriologists.
56th General Meeting. 1956.
This outline was prepared by H. L. Jeter,
Chief, Program Support Training Branch,
and R. Rossomanno, Microbiologist, USEPA
Cincinnati, Ohio 45268.
Descriptors: Cultures, Bacteria, Enteric
Bacteria, Coliforms, Fecal Coliforms,
Streptococcus, Membrane Filters.
9-9
-------
SELECTION OF SAMPLE FILTRATION VOLUMES
FOR MEMBRANE FILTER METHODS
I INTRODUCTION
A Wide Range of Filtration Volumes
The membrane filter permits testing
a wide range of sample volumes, from
several hundred milliliters to as little
as 0.0001 ml, or even less. Suitable
dilution of sample volumes smaller
than 1. 0 ml may be required for
accuracy of sample measurement.
2 While the method lends itself to a wide
range of sample volumes, the filter has
limitations in the number of isolated
(or countable) differentiated colonies
which can develop on the available sur-
face area. Figure 1 illustrates a com-
mon pattern of colony counts over a
wide range of sample filtration volumes;.
For both the total colonies and the
coliforrn colonies there is a pro-
portional relationship between colony
count and sample volume over much
of the range of sample volumes.
With increasing colony counts, there
are some levels above which the pro-
portional relationship fails, both for
total and for coliform colonies.
In the straight lines in Figure 1,
where there is proportionality be-
tween colony counts and filtration
volumes, it is possible to compute
density of bacteria in the sample,
based on the equation:
No. organisms _
per 100 ml
100 X
No. colonies counted
No. ml of sample
filtered
The equation is not quantitatively
reliable in the curved portions of
the lines.
Total Colonies
I I I 1 I
Coliform Colonies
i i i i i i i j
Sample Volume
Figure 1
The graph is based on coliform de-
terminations using M Endo Broth
MF. The line designated "Total
Colonies" includes both coliform and
noncoliform colonies. The "Coliform
Colonies" line refers only to differ-
entiated colonies having the typical
color and sheen of coliform colonies
on the medium.
B Scope of this Presentation
1 To explain limitations on quantitatively
reliable colony counts on membrane
filters.
2 To present numbers of colonies accept-
able for quantitative tests with availabl<
media.
3 To demonstrate the different quantitative
bacterial density ranges determined by
single-volume filtrations, with available
media.
4 To demonstrate quantitative ranges
covered by a series of filtration volurr
with currently-used differential media.
W. RA. mem. 75f. 8. 77
10-1
-------
Selection of Sample Filtration Volumes
H
5 To provide guidance for selection of
sample filtration volumes under the
following practical conditions:
a When there is need to determine
compliance with established bac-
teriological water quality standards.
b When there is needto determine
density of a specifiedbacterial group.
1) In the absence of prior bacterio-
logical data, and
2) When prior bacteriological data
are available.
LIMITATIONS ON COLONY COUNTS
USED FOR QUANTITATIVE WORK
A Bacterial Density
1 The minimum sample volume should
result in production of at least 20
colonies of the bacteria being counted.
Sample volumes yielding lesser numbers
of colonies are subject to unacceptably
large random variations in the computed
bacterial density, determined as above
(I, A, 2).
2 The maximum acceptable colony density,
for quantitative determinations, is vari-
able with the bacterial group tested and
the medium used. Factors influencing
maximum acceptable colony density
include:
a Size of colonies. In principle, each
colony should represent one bacterial
cell deposited on the filter, or, con-
versely, each bacterial cell deposit-
ed on the filter should result in pro-
duction of a recognizable colony.
Madia producing relatively large
colonies (as in the fecal coliform
test) will support smaller numbers
of colonies on the filter than media
producing smaller colonies (such as
fecal streptococci). If the colony
size is large and the number of bac-
teria deposited on the filter is great,
some colonies will represent two or
more cells initially deposited on the
filter, and the quantitative reliability
of the test is impaired.
b Selectivity of medium. Highly
selective media permit growth of
relatively few colonies of extraneous,
unwanted, bacteria. The available
area of the filter is occupied pri-
marily by colonies of the group
tested. Thus, with a highly selective
medium such as that used for fecal
streptococci, it is reasonable to
expect good quantitative results with
relatively high colony counts. Con-
versely, media having limited
selectivity (such as Endo-type media
for coliforms) supports growth of
considerable numbers of extraneous
bacterial colonies, and it is necessary
to place arbitrary limitations on the
number of colonies per membrane in
quantitative studies.
c Biochemical interference between
neighboring colonies. Associated
with the physical crowding effects
noted in (b) above, sheen production
of coliform colonies may be inhibited
by overcrowding of colonies. This
reinforces the need for restriction
of colony density on the filter.
B Suspended Matter
1 Particulate matter in the sample can
ba a limitation in application of mem-
brane filters, especially when the
amount of suspended matter is relatively
great and the bacterial density is low.
2 Difficulties from suspended matter in
the sample may be apparent in several
ways.
a The pores of the filter may be
occluded, limiting the volume of
sample that can be filtered. This
problem has been noted in waters
rich in clays and in waters containing
large populations of certain diatoms
or other algae.
b Fibrous matter can be troublesome,
due to the tendency for a capillary
film of liquid culture medium to form
around the fibers. Colonies in contact
with such fibers tend to grow along
10-2
-------
Selection of Sample Filtration Volumes
III
the path of the fibers, assuming
highly irregular forms. Sometimes
these colonies cover abnormally
large areas of the filter surface.
c A more or less continuous mat of
particles may be collected from
some samples, with each particle
soon surrounded by a film of liquid
culture medium. On such filters,
distinct colonies usually fail to
develop as discrete entities, but
grow in a more or less continuous
film over the entire surface of the
filter.
3 Problems due to particulate matter
often can be reduced by filtration of
the selected volume of water in two or
more smaller increments, through
separate filters. In effect, this is a
means of enlarging the available sur-
face area of the filter.
Prefiltration of the sample through a
coarse filter for preliminary removal
of extraneous particulate matter is not
recommended in quantitative work.
Prefiltration invariably results in re-
moval of unpredictably large numbers of
bacterial cells.
In some cases the problem of particu-
lates cannot be solved, and it must
then be conceded that the membrane
filter method is not acceptable for such
samples. It then becomes necessary to
resort to other procedures, such as the
dilution tube method or agar plating
methods.
LIMITS ON NUMBER OF COLONIES ON
FILTERS WITH VARIOUS MEDIA
Referring to Figure 1, a specific number of
colonies is not shown for acceptable propor-
tionality between colony number and filtration
volume. Fixed limits cannot be stated for all
test situations, for these limits are somewhat
variable from one culture medium to another
and from one sample source to another.
The recommended limits shown in Table 1
are empirical values based on re-
search experience. It is believed that quan-
titative determinations of acceptable statistical
reliability can be obtained if the determinations
are based on colony counts within the limita-
tions shown.
IV RANGE OF BACTERIAL DENSITIES
COVERED BY SINGLE-VOLUME
FILTRATIONS
A The equation used in Section I of this out-
line can be used with any sample filtration
volume to determine the bacterial density
range over which acceptable counts can be
made.
For example, assume that a sample of
10 ml is used for a quantitative determi-
nation of total coliforms. Based on Table
1, quantitative determinations should be
based on a filtration volume yielding 20 -
80 coliform colonies. Compute
the coliforms per 100 ml based on 20
colonies and on 80 colonies per filter. This
will be the bacterial density range covered
by a 10 ml filtration volume, thus:
for 20 colonies:
No. coliforms per 100 ml
and for 80 colonies,
No. coliforms per 100 ml
100 X -~~
200
100 X
800
80
10
B
Thus, a 10 ml sample portion is appro-
priate for determination of total coliforms
in the range 200 - 800 per 100 ml.
Table 2 illustrates the ranges covered for
several filtration volumes, with colony
counts in the ranges 20 - 60, 20 - 80, and
20 - 100 per filtration volume.
10-3
-------
Selection CK Sample Filtration Volumes
Table 1. RECOMMENDED COLONY COUNT RANGES FOR.
QUANTITATIVE DETERMINATIONS WITH
MEMBRANE FILTER TESTS
Test
Total Coliform
Fecal Coliform
Fecal Streptococci
Total Counts
No. colonies
Minimum
20
20
20
20
i
Maximum
80
60
100
200
Medium
M Endo Broth MF,
LES Endo Medium
M FC Broth
M Enterococcus
Agar, KF Agar
M Enrichment
Broth
Remarks
Not more than 200
colonies of all types
Spreaders may
require adjustment
Table 2. RANGES COVERED BY REPRESENTATIVE
FILTRATION VOLUMES
Ml sample
filtered
100
10
1
0. 1
0.01
, Bacterial count per 100 ml based on
20 colonies
20
200
2000
20,000
200,000
60 colonies
60
600
6000
60, 000
600, 000
80 colonies
80
800
8000
80, 000
800, 000
100 colonies
100
1000
10, 000
100,000
1, 000, 000
C Application of a Series of Filtration
Volumes
1 Examination of Table 2 shows that for
quantitative work on membrane filters,
to extend the range of any test, it is
necessary to filter two or more different
sample volumes. The worker uses the
one sample volume yielding a quantita-
tively acceptable number of colonies to
compute the bacterial count per 100 ml.
2 Further, it can be seen that varying the
filtration volumes by decimal increments
will be inappropriate; there are values
within the total range covered in which
the colony number would fall outside the
critical counting range for the test being
made.
3 In order to give maximum assurance
that a series of varying filtration
will yield at least one membrane
with an acceptable number of colonies,
the range of filtration volumes should
be along these lines:
a Total coliform counts should be
based on filtration volumes varying
by a factor of 4, or less.
b Fecal coliform counts should be
based on filtration volumes varying
by a factor of 3, or less.
c Fecal streptococcus counts should
be based on filtration volumes vary-
ing by a factor of 5, or less.
V SELECTING FILTRATION VOLUMES
FOR MEMBRANE FILTER TESTS
A Total Coliform Counts
1 Determination of compliance with exist-
ing bacterial quality standards.
10-4
-------
Selection of Sample Filtration Volumes
For all tests to determine whether
water meets PHS Drinking Water
quality standards, minimum sample
sizes are prescribed as 50 ml, with
100 ml sample volumes suggested.
With tests in which it is assumed
that coliforms are present in some
numbers, and the test is to determine
whether some limiting standard (as
1000 per 100 ml in natural bathing
waters, prescribed by some
agencies), another approach is
suggested. Here, select the sample
filtration volume which would be
quantitatively most acceptable to
count coliforms at the limiting value.
For example, with a limiting value
of 1000 per 100 ml:
per 100 ml
=
Ko. colonies counte
No. ml of sample fih
2) Polluted raw surface water, 0.0
0.08, 0,15, and 0. 5 ml samples
will cover a count range of 4000
to 400, 000 per 100 ml.
3) Sewage and dilute sewage, with
filtration volumes of 0. 0003,
0.001, 0.003, and 0.01, will
cover a count range of 200, 000
to 27, 000, 000 oer 100 ml.
b If prior coliform data are available
Use the equation:
3asic filtration _ ^ 50
•/olume in ml Average coliform c<
Example: Assume that prior data
indicate average coliform count of
35, 000 per 100 ml. Using the
equation:
This previously given equation can be rearrange
-Sasic filt rat ion volume in ml = 100 X
50
35,000
^ ! mple filtration
• lume in nil
md from this:
No. colonies countf(.
No. organisms/100 in
mple filtration volume _ _
in ml
X
50
1000
(The value 50 is the midrange number
of colonies for an acceptable colony
count of 20 - 80 for computing coli-
forms per 100 ml)
In quantitative work, to determine
number of coliforms per 100 ml the
worker may or may not have prior
information or standards to use as
guidance in selecting filtrationvolum.es.
a In absence of prior bacteriological
data
1) Unpolluted raw surface water, 1,
4, 15, and 60 ml samples will
cover a count range of 33 - 8000
per 100 ml.
= 0. 143 ml
Round off the filtration volume to
0. 15 ml.
To assure a reasonable count-range,
filter increments of 0. 04 and 0. 60 ml
in addition. This will provide for
acceptable coliform counts in the
range of 3300 to 200, 000 per 100 ml
B Fecal Coliform Counts
1 Currently, no drinking water standards
are based on fecal coliform organisms.
Many states have environmental water
quality standards which are based on
fecal coliform organisms.
2 Determination of fecal coliforms in the
absence of prior data.
a Unpolluted raw surface water:
Filter 1, 3, 10, and 30 ml sample
portions. These volumes will
cover a fecal coliform range of
67 - 6000 per 100 ml.
10-5
-------
Selection of Sample Filtration Volumes
b Polluted raw surface water: Filter
portions of 0. 1, 0.3, 1.0, and 3.0
ml. This will cover a fecal coliform
count range of 670 to 60,000 per 100
ml.
c Sewage and dilute sewage: Filter
sample portions of 0. 0003, 0. 001
and 0. 003 ml. This will provide for
counts of 670, 000 to 20, 000, 000 per
100 ml.
Determination of fecal coliforms in
presence of prior data
a When previous fecal coliform counts
are available:
Filtration volume =
in ml
100 X -
40
Av. fecal coliform
count per 100 ml
Example: Prior data show 8000 fecal
coliforms per 100 ml.
40
Basic filtration volume in ml = 100 X -QQQQ-
= 0.5
Filter volumes of 0. 15, 0.5 and 1. 5
ml. This will be suitable for fecal
coliform counts over the range 1300
to 40,000.
b When previous total coliform data
are available but no fecal coliform
data are available, use the total
coliform value as above, but filter
3x and 9x the computed basic volume.
Example (from above): Computed
basic value = 0.5 ml
Filter volumes of 0. 5, 1. 5 and 5. 0
ml.
C Fecal Streptococcus Determinations
1 In absence of prior data
a Unpolluted raw surface water: Filter
sample portions of 1, 5, 25 and 100
ml. This will provide for fecal
streptococcus counts in the range 20
to 10,000 per 100 ml.
10-6
b Polluted surface water: Filter
sample portions of 0. 1, 0. 5, and
2. 0 ml. This will provide for fecal
streptococcus counts in the range
1000 to 100,000 per 100 ml. Pro-
vision for rather high counts of
fecal streptococci is made because
of possible situations in which pol-
lution of the water originates from
domestic or wild animals. In the
event that such pollution is highly
improbable, a filtration series of
0.2, 1.0, and 5. 0 ml (covering a
count range of 400 to 50, 000 per
100 ml) would be more appropriate.
2 When prior data are available
a If coliform, but not fecal strepto-
coccus data are available, compute
a basic filtration volume as in A, 2
above, but use the average coliform
count as a point of reference. If
significant pollution from domestic
or wild animals is believed present,
filter 0. 2X, IX and 5X the basic
filtration volumes. If the pollution
levels are believed due primarily to
human sources, use IX, 5X and 25X
the basic filtration volumes.
b If prior streptococcus data are avail-
able, use the equation
Basic filtration = 100 X
volume in ml
60
Av. Streptococcus
count per 100 ml
and filter 0.2X, IX, and 5X the basic
filtration volume for streptococci.
•
REFERENCES
1 The Federal Register. October 23, 1956,
pp 8110-11; and March 1, 1957. p 1271.
2 Clark, H. F., Kabler, P.W., and
Geldreich, E.E. Advantages and
Limitations of the Membrane Filter
Procedure. Water and Sewage Works.
Saotember 1957.
This outline Was prepared Dy H. L. Jeier,
Chief, program Support Training Branch,
USEPA, Cincinnati, Ohio 45268
Descriptors: Bacteria, Sample Testing
procedures, Filter, Membrane
-------
DETAILED MEMBRANE FILTER METHODS
I BASIC PROCEDURES
A Introduction
Successful application of membrane filter
methods requires development of good
routine operational practices. The
detailed basic procedures described in
this Section are applicable to all mem-
brane filter methods in water bacteriology
for filtration, incubation, colony counting,
and reporting of results. In addition,
equipment and supplies used in membrane
filter procedures described here are not
repeated elsewhere in this text in such detail.
Workers using membrane filter methods
for the first time are urged to become
thoroughly familiar with these basic
procedures and precautions.
B General Supplies and Equipment List
Table 1 is a check list of materials.
C "Sterilizing" Media
Set tubes of freshly prepared medium in a
boiling waterbath for 10 minutes. This
method suffices for medium in tubes up to
25X150 mm. Frequent agitation is needed
with media containing agar.
Alternately, coliform media can be
directly heated on a hotplate to the first
bubble of boiling. Stir the medium
frequently if direct heat is used, to avoid
charring the medium.
Do not sterilize in the autoclave.
D General Laboratory Procedures with
Membrane Filters
1 Prepare data sheet
Minimum data required are: sample
identification, test performed including
media and methods, sample filtration
volumes, and the bench numbers
assigned to individual membrane filters.
2 Disinfect the laboratory bench surface.
Use a suitable disinfectant solution and
allow the surface to dry before
proceeding.
3 Set out sterile culture containers in an
orderly arrangement.
4 Label the culture containers.
Numbers correspond with the filter
numbers shown on the data sheet.
5 Place one sterile absorbent pad* in
each culture container, unless an agar
medium is being used.
Use sterile forceps for all manipulations
of absorbent pads and membrane filters.
Forceps sterility is maintained by
storing the working tips in about 1 inch
of methanol or ethanol. Because the
alcohol deteriorates the filter, dissipate
it by burning before using the forceps.
Avoid heating the forceps in the burner
as hot metal chars the filter.
*When an agar medium is used, absorbent pads are not used. The amount of medium should be
sufficient to make a layer approximately 1/8" deep in the culture container. In the 50 mm
plastic culture containers this corresponds to approximately 6-8 ml of culture medium.
NOTE: Mention of commercial products and manufacturers does not imply endorsement by the
Office Of Water Programs, Environmental Protection Agency.
W. BA. mem. 86. 8. 77
11-1
-------
Detailed Membrane Filter Methods
Table 1. EQUIPMENT, SUPPLIES AND MEDIA
Item
Funnel unit assemblies
Ring stand, with about a 3" split ring, to
support the filtration funnel
Forceps, smooth tips, type for
MF work
Methanol, in small wide- mouthed bottles.
about 20 ml for sterilizing forceps
Suction flasks, glass, 1 liter, mouth to
fit No. 8 stopper
Rubber tubing, 2-3 feet, to connect
suction flask to vacuum services, latex
rubber 3/16" I. D. by 3/32" wall
Pinch clamps strong enough for tight
compression of rubber tubing above
Pipettes, 10 ml, graduated, Mohr type.
sterile, dispense 10 per can per working
space per day. (Resterilize daily to
meet need).
Pipettes, 1 ml, graduated, Mohr type.
sterile, dispense 24 per can per working
space per day. (Resterilize daily to
meet need).
Pipette boxes, sterile, for 1 ml and
10 ml pipettes (sterilize above pipettes
in these boxes).
Cylinders, 100 ml graduated, sterile.
(resterilize daily to meet need).
Jars, to receive used pipettes
Gas burner, Bunsen or similar
laboratory type
Wax pencils, red, suitable for writing
on glass
Sponge in dilute iodine, to disinfect the
desk tops
Membrane filters (white, grid marked.
sterile, and suitable pore size for
microbiological analysis of water)
Absorbent pads for nutrient, (47 mm in
diameter), sterile, in units of 10 pads
per package. Not required if medium
•contains agar.
Petri dishes, disposable, plastic.
50 X 12 mm, sterile
Waterbath incubator 44.5 + 0.2°C
Vegetable crispers, or cake boxes,
plastic, with tight fitting covers, for
membrane filter incubations
Fluorescent lamp, with extension cord.
Ring stand, with clamps, utility type
Tot
M-Endo
Broth
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
1 Coliforms
L. E.S.
Coliform
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Delayed
Coliform
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Fecal
Coliform
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Fecal
Streptococcus
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Verified
Tests
X
X
11-2
-------
Detailed Membrane Filter Methods
Table 1. EQUIPMENT, SUPPLIES AND MEDIA (Cont'd)
Item
Half- round glass paper weights for
colony counting,with lower half of a
2- 02 metal ointment box
Hand tally, single unit acceptable,
hand or desk type
Stereoscopic (dissection) microscope,
magnification of 10X or 15X, prefer-
able binocular wide field type
Bacteriological inoculating needle
Wire racks for culture tubes,
10 openings by five openings pre-
ferred, dimensions overall approxi-
mately 6" X12"
Phenol Red Lactose Broth in 16 X
150 mm fermentation tubes with
metal caps, 10 ml per tube
Eosin Methylene Blue Agar
(Levine) in petri plates, prepared
ready for use
Nutrient agar slants, in screw
capped tubes, 16 X 126 mm
Gram stain solutions, 4 solutions
per complete set
Microscope, compound, binocular,
with oil immersion lens, micro-
scope lamp and immersion oil
Microscope slides, new, clean,
1" X3" size
Water proof plastic bags
for fecal coliform culture
dish incubation
M-Endo medium, MF dehydrated
medium in 25 X95 mm flat bottomed
screw-capped glass vials, 1.44 g
per tube, sufficient for 30 ml of
medium
Ethanol, 95% in small bottles or
screw-capped tubes, about 20 ml
per tube
Sodium benzoate solution, 12%
aqueous, in 25 X 150 mm screw-
capped tubes, about 10 ml per tube
L. E. S. Endo Agar MF, dehydrated
M-Endo medium, 0. 36 g per 25 X
95 mm flat bottomed screw-capped
glass vial, plus 0. 45 g agar, for 30 ml
Lactose Lauryl Sulfate Tryptose Broth
in 25 X 150 mm test tube without
included gas tube, about 25 ml, tor
enrichment in L. E. S. method
Total Conforms
M-Endo
Broth
X
X
X
X
X
L. E.S.
Coliform
X
X
X
X
X
X
Delayed
Coliform
X
X
X
X
X
X
Fecal
Coliform
X
X
X
X
Fecal
Streptococcus
X
X
X
Verified
Tests
X
X
X
X
X
X
X
X
X
11-3
-------
Detailed Membrane Filter Methods
Table 1. EQUIPMENT, SUPPLIES AND MEDIA (Cont'd)
Item
M-FC Broth for fecal coliform.
dehydrated medium in 25 X 95 mm
flat bottomed screw- capped glass
vials, 1. 11 g per tube, sufficient
for 30 ml of culture medium
Rosolic acid, 1% solution, in
0. 2N NaOH, in 25 X 150 mm flat
bottomed screw- capped tubes.
about 5 ml per tube, freshly
prepared
KF Agar, dehydrated medium in 25 X 150 mm
screw-capped tubes, sufficient for .10 nil. 2. :ig
per tube
Dilution bottles, 6-oz, preferable
boro- silicate glass, with screw-
cap (or rubber stopper protected
by paper) , each containing 99 ml
of sterile phosphate buffered
distilled water
Electric hot plate surface
Beakers, 400 - 600 ml (for water-
bath in preparation of membrane
filter culture media)
Crucible tongs, to be used at
electric hot plates, for removal
of hot tubes of culture media for
boiling waterbath
Total Coliforms
M - Endo
Broth
X
X
X
X
L.E.S.
Coliform
X
X
X
X
Delayed
Coliform
X
X
X
X
Fecal
Coliform
X
X
X
X
X
X
Fecal
Streptococcus
X
X
X
X
X
Verified
Test
11-4
-------
Detailed Membrane Filter Methods
6 Deliver enough culture medium to
saturate each absorbent pad, using
a sterile pipette.
Exact quantities cannot be stated
because pads and culture containers vary.
Sufficient medium should be applied so
that when the culture container is tipped,
a good-sized drop of culture medium
freely drains cut of the absorbent pad.
7 Organize supplies and equipment for
convenient sample filtration. In
training courses, laboratory instructors
will suggest useful arrangements;
eventually the individual will select a
system, of bench-top organization most
suited to his own needs. The important
point in any arrangement is to have all
needed equipment and supplies con-
veniently at hand, in such a pattern as
to minimize lost time in useless motions.
8 Lay a sterile membrane filter on the
filter holder, grid-side up, centered
over the porous part of the filter
support plate.
Membrane filters are extremely
delicate and easily damaged. For
manipulation, the sterile forceps
should always grasp the outer part
of the filter disk, outside the part
of the filter through which the sample
passes.
9 Attach the funnel element to the base
of the filtration unit.
To avoid damage to the membrane
filter, locking forces should only be
applied at the locking arrangement.
The funnel element never should be
turned or twisted while being seated
and locked to the lower element of the
filter holding unit. Filter holding units
featuring a bayonet joint and locking
ring to join the upper element to the
lower element require special care on
the part of the operator. The locking
ring should be turned sufficiently to
give a snug fit, but should not be
tightened excessively.
10 Shake the sample thoroughly.
11 Measure sample into the funnel with
vacuum turned off.
The primary objectives here are:
1) accurate measurement of sample;
and 2) optimum distribution of colonies
on the filter after incubation. To
meet these objectives, methods of
measurement and dispensation to the
filtration assembly are varied with
different sample filtration volumes.
a With samples greater than 20 ml,
measure the sample with a sterile
graduated cylinder and pour it into
the funnel. It is important to rinse
this graduate with sterile buffered
distilled water to preclude the loss
of excessive sample volume. This
should be poured into the funnel.
b With samples of 10 ml to 20 ml,
measure the sample with a sterile
10 ml or 20 ml pipette, and pipette
on a dry membrane in the filtration
assembly.
c With samples of 2 ml to 10 ml, pour
about 20 ml of sterile dilution water
into the filtration assembly, then
measure the sample into the sterile
buffered dilution water with a 10 ml
sterile pipette.
d With samples of 0. 5 to 2 ml, pour
about 20 ml of sterile dilution water
into the funnel assembly, then
measure the sample into the sterile
dilution water in the funnel with a
1 ml or a 2 ml pipette.
e If a sample of less than 0. 5 ml is to
be filtered, prepare appropriate
dilutions in sterile dilution water,
and proceed as applicable in item c
or d above.
When dilutions of samples are needed,
always make the flltrations as soon
as possible after dilution of the
sample; this never should exceed
NOTE: Mention of commercial products and manufacturers does not imply endorsement
by the Office of Water Programs, Environmental Protection Agency.
11-5
-------
Detailed Membrane Filter Methods
30 minutes. Always shake sample
dilutions thoroughly before delivering
measured volumes.
12 Turn on the vacuum.
Open the appropriate spring clamp or
valve, and filter the sample.
After sample filtration a few droplets
of sample usually remain adhered to
the funnel walls. Unless these droplets
are removed, the bacteria contained in
them will be a source of contamination
of later samples. (In laboratory
practice the funnel unit is not routinely
sterilized between successive filtrations
of a series). The purpose of the funnel
rinse is to flush all droplets of a sample
from the funnel walls to the membrane
filter. Extensive tests have shown that
with proper rinsing technique, bacterial
retention on the funnel walls is negligible.
13 Rinse the sample through the filter.
After all the sample has passed through
the membrane filter, rinse down the
sides of the funnel walls with at least
20 ml of sterile dilution water. Repeat
the rinse twice after all the first rinse
has passed through the filter. Cut off
suction on the filtration assembly.
14 Remove the funnel element of the filter
holding unit.
If a ring stand with split ring is used,
hang the funnel element on the ring;
otherwise, place the inverted funnel
element on the inner surface of the
wrapping material. This requires
care in opening the sterilized package,
but it is effective as a protection of the
funnel ring from contamination.
15 Take the membrane filter from the
filter holder and carefully place it.
grid-side up on the medium.
Check that no air bubbles have been
trapped between the membrane filter
and the underlying absorbent pad or
agar. Relay the membrane if necessary.
16 Place in incubator after finishing
filtration series.
Invert the containers. The immediate
atmosphere of the incubating membrane
filter must be at or very near 100%
relative humidity.
17 Count colonies which have appeared
after incubating for the prescribed
time.
A stereoscopic microscope magnifying
10-15 times and careful illumination
give best counts.
For reporting results, the computation
is:
bacteria/100 ml =
No. colonies counted XI00
Sample volume filtered in nil
Example:
A total of 36 colonies grew after
filtering a 10 ml sample. The
number reported is:
36 colonies
10ml
X 100 = 360 per 100 ml
Report results to two significant figures.
Example:
A total of 40 colonies grew after
filtering a 3 ml sample.
This calculation gives:
40 colonies
3 ml
X 100 T 1333. 33 per 100ml
But the number reported should be
1300 per 100 ml.
11-6
-------
Detailed Membrane Filter Methods
II MF LABORATORY TESTS FOR
COLIFORM GROUP
A Standard Coliform Test (Based on M-Endo
Broth MF)
1 Culture medium
a M-Endo Broth MF Difco 0749-02
or the equivalent BBL M-Coliform
Broth 01-494
Preparation of Culture Medium
(M-Endo Broth) for Standard MF
Coliform Test
Yeast extract 1.5
Casitone or equivalent 5.0
Thiopeptone or equivalent 5.0
Tryptose 10.0
Lactose 12.5
Sodium desoxycholate 0.1
Dipotassium phosphate 4.375 g
Monopotassium phosphate 1.375g
Sodium chloride 5.0 e
o
Sodium lauryl sulfate 0. 05 g
Basic fuchsin (bacteriological) 1. 05 g
Sodium sulfite 2.1 g
Distilled water (containing 1000 ml
20.0 ml ethanol)
This medium is available in
dehydrated form and it is rec-
ommended that the commercially
available medium be used in
preference to compounding the
medium of its individual constituents.
To prepare the medium for use.
suspend the dehydrated medium at
the rate of 48 grams per liter of
water containing ethyl alcohol at
the rate of 20 ml per liter.
As a time-saving convenience, it is
recommended that the laboratory
worker preweigh the dehydrated
medium in closed tubes for several
days, or even weeks, at one operation.
With this system, a large number
of increments of dehydrated medium
(e.g., 1.44 grains), sufficient for
some convenient (e.g., 30 ml)
volume of finished culture medium
are weighed and dispensed into
screw-capped culture tubes, and
stored until needed. Storage should
preferably be in a darkened disiccator.
A supply of distilled water containing
20 ml stock ethanol per liter can be
maintained.
When the medium is to be used, it
is reconstituted by adding 30 ml of
the distilled water-ethanol mixture
per tube of pre-weighed dehydrated
culture medium.
b Medium is "sterilized" as directed
in I, C.
c Finished medium can be retained
up to 96 hours if kept in a cool,
dark place. Many workers prefer
to reconstitute fresh medium daily.
Filtration and incubation procedures
are as given in I, D.
Special instructions:
a For counting, use the wide field
binocular dissecting microscope, or
simple lens. For illumination, use
a light source perpendicular to the
plane of the membrane filter. A
small fluorescent lamp is ideal for
the purpose.
b Coliform colonies have a "metallic"
surface sheen under reflected light
which may cover the entire colony, or
it may appear only in the center. Non-
coliform colonies range from
colorless to pink, but do not have
the characteristic sheen.
c Record the colony counts on the
data sheet, and compute the coliform
count per 100 ml of sample.
11-7
-------
Detailed Membrane Filter Methods
B Standard Coliform Tests (Based on L. E. S.
Endo Agar)
The distinction of the L. E.S. count is a
two hour enrichment incubation on LST
broth. M-Endo L. E.S. medium is used
as agar rather than the broth.
1 Preparation of culture medium
(L.E.S. Endo Agar) for L. E.S.
coliform test
a Formula from McCarthy, Delaney,
and Grasso (2)
manipulations of the pads.
(Agar occupies smaller half
bottom).
or
Bacto-Yeast Extract
Bacto-Casitone
Bacto-Thiopeptone
Bacto-Tryptose
Bacto-Lactose
Dipotassium phosphate
Monopotassium phosphate
Sodium chloride
Sodium desoxycholate
Sodium lauryl sulfate
Sodium sulfite
Bacto-Basic fuchsin
Agar
Distilled water (containing
2 0 ml ethyl alcohol)
1.2
3.7
3.7
7.5
9.4
3.3
g
1.0
3.7
0.1
0.05 g
1.6 g
0.8 g
15 g
1000 ml
b To rehydrate the medium, suspend
51 grams in the water-ethyl alcohol
solution.
c Medium is "sterilized" as directed
in I, C.
d Pour 4-6 ml of freshly prepared Agar
into the smaller half of the container.
Allow the medium to cool and solidify.
2 Procedures for filtration and incubation
a Lay out the culture dishes in a row
or series of rows as usual. Place
these with the upper (lid) or top
side down.
b Place one sterile absorbent pad in
the larger half of each container
(lid). Use sterile forceps for all
11-8
c Using a sterile pipette, deliver
enough single strength lauryl
sulfate tryptose broth to saturate
the pad only. Avoid excess medium.
d Follow general procedures for
filtering in I, D. Place filters on
pad with lauryl sulfate tryptose
broth.
e Upon completion of the filtrations,
invert the culture containers and
incubate at 35° C for 1 1/2 to 2
hours.
3 2-hour procedures
a Transfer the membrane filter from
the enrichment pad in the upper half
to the agar medium in the lower
half of the container. Carefully
roll the membrane onto the agar
surface to avoid trapping air
bubbles beneath the membrane.
b Removal of the used absorbent pad
is optional.
c The container is inverted and
incubated 22 hours + 2 hours + 0. 5 C.
4 Counting procedures are as in I, D.
5 L. E. S. Endo Agar may be used as a
single-stage medium (no enrichment
step) in the same manner as M-Endo
Broth, MF.
C Delayed Incubation Coliform Test
This technique is applicable in situations
where there is an excessive delay between
sample collection and plating. The procedure
is unnecessary when the interval be-
tween sample collection and plating is
within acceptable limits.
Preparation of culture media for
delayed incubation coliform test
a Preservative media M-Endo Broth
base
-------
Detailed Membrane Filter Methods
To 30 ml of M-Endo Broth MF
prepared in accordance with
directions in H, A, 1 of this
outline, add 1. 0 ml of a sterile
12% aqueous solution of sodium
benzoate.
L. E. S. MF Holding Medium-
Coliform: Dissolve 12.7 grams in
1 liter of distiUed water. No
heating is necessary. Final pH
7. 1 + 0. 1. This medium contains
sodium benzoate.
b Growth media
M-Endo Broth MF is used, prepared
as described in II. A, 1 earlier in
this outline. Alternately, L.E.S.
Endo Medium may be used.
2 General filtration followed is in I, D.
Special procedures are:
a Transfer the membrane filter from
the filtration apparatus to a pad
saturated with benzoated M-Endo
Broth.
b Close the culture dishes and hold
in a container at ambient temperature.
This may be mailed or transported
to a central laboratory. The mailing
or transporting tube should contain
accurate transmittal data sheets which
correspond to properly labeled dishes.
Transportation time, in the case of
mailed containers, should not exceed
three days to the time of reception
by the testing laboratory.
c On receipt in the central laboratory,
unpack mailing carton, and lay out
the culture containers on the labora-
tory bench.
d Remove the tops from the culture
containers. Using sterile forceps,
remove each membrane and its
absorbent pad to the other half of
the culture container.
e With a sterile pipette or sterile
absorbent pad, remove preservative
medium from the culture container.
f Place a sterile absorbent pad in
each culture container, and deliver
enough freshly prepared M-Endo
Broth to saturate each pad.
g Using sterile forceps, transfer the
membrane to the new absorbent pad
containing M-Endo Broth. Place
the membrane carefully to avoid
entrapment of air between the
membrane and the underlying
absorbent pad. Discard the
absorbent pad containing pre-
servative medium.
h After incubation of 20 + 2 hours
at 35°C, count colonies as in the
above section A, 2.
i If L.E.S. Endo Agar is used, the
steps beginning with (e) above are
omitted; and the membrane filter is
removed from the preservative
medium and transferred to a fresh
culture container with L. E. S. Endo
Agar, incubated, and colonies
counted in the usual way.
D Verified Membrane Filter Coliform Test
This procedure applies to identification
of colonies growing on Endo-type media
used for determination of total coliform
counts. Isolates from these colonies are
studied for gas production from lactose
and typical coliform morphology. In
effect, the procedure corresponds with
the Completed Test stage of the multiple
fermentation tube test for coliforms.
Procedure:
1 Select a membrane filter bearing
several well-isolated coliform-type
colonies.
2 Using sterile technique, pick all
colonies in a selected area with the
inoculation needle, making transfers
into tubes of phenol red lactose broth
(or lauryl sulfate tryptose lactose
11-9
-------
Detailed Membrane Filter Methods
broth). Using an appropriate data
sheet record the interpretation of
each colony, using, for instance,
"C" for colonies having the typical
color and sheen of coliforms; "NC"
for colonies not conforming to
coliform colony appearance on
Endotype media.
3 Incubate the broth tubes at 35° C+ 0. 5°C.
4 At 24 hours:
a Read and record the results from
the lactose broth fermentation tubes.
The following code is suggested:
No indication of acid or gas
production, either with or
without evidence of growth.
A Evidence of acid but not gas
(applies only when a pH indicator
is included in the broth medium)
G Growth with production of gas.
If pH indicator is used, use
symbol AG to show evidence of
acid. Gas in any quantity is a
positive test.
b Tubes not showing gas production are
returned to the 35° C incubator.
c Gas-positive tubes are transferred
as follows:
1) Prepare a streak inoculation on
EMB agar for colony isolation, and
using the same culture.
2) Inoculate a nutrient agar slant.
3) Incubate the EMB agar plates and
slants at 35° C + 0. 5°C.
5 At 48 hours:
a Read and record results of lactose
broth tubes which were negative at
24 hours and were returned for
further incubation.
b Gas-positive cultures are subjected
to further transfers as in 4c.
Gas-negative cultures are discarded
without further study; they are
coliform- negative.
c Examine the cultures transferred
to EMB agar plates and to nutrient
agar slants, as follows:
1) Examine the EMB agar plate for
evidence of purity of culture; if
the culture represents more than
one colony type, discard the
nutrient agar culture and reisolate
each of the representative colonial
types on the EMB plate and resume
as with 4c for each isolation.
If purity of culture appears evident,
continue with c (2) below.
2) Prepare a smear and Gram stain
from each nutrient agar slant
culture. The Gram stain should
be made on a culture not more
than 24 hours old. Examine under
oil immersion for typical coliform
morphology, and record results.
6 At 72 hours:
Perform procedures described in 5c
above, and record results.
7 Coliform colonies are considered
verified if the procedures demonstrate
a pure culture of bacteria which are
gram negative nonspore-forming rods
and produce gas from lactose at 35° C
within 48 hours.
E Fecal Coliform Count (Based on M-FC
Broth Base)
The count depends upon growth on a
special medium at 44. 5+0. 2°C.
1 Preparation of Culture Medium
(M-FC Broth Base) for Fecal
Coliform Count
11-10
-------
Detailed Membrane Filter Methods
a Composition
Tryptose 10. 0 g
Proteose Peptone No. 3 5. 0 g
Yeast extract 3 . 0 g
Sodium chloride 5. 0 g
Lactose 12. 5 g
Bile salts No. 3 1.5 g
Rosolic acid* (Allied 10. 0 ml
Chemical)
Aniline blue (Allied Chemical) 0. 1 g
Distilled water
1000 ml
b To prepare the medium dissolve
37. 1 grams in a liter of distilled
water which contains 10 ml of 1%
rosolic acid (prepared in 0. 2 N
NaOH).
Fresh solutions of rosolic acid give
best results. Discard solutions
which have changed from dark red
to orange.
c To sterilize, heat to boiling as
directed in I, C.
d Prepared medium may be retained
up to 4 days in the dark at 2-8°C.
2 Special supplies
Small water proof plastic sacks capable
of being sealed against water with
capacity of 3 to 6 culture containers.
3 Filtration procedures are as given in
I. D.
4 Elevated temperature incubation
a Place fecal coliform count mem-
branes at 44. 5 + 0. 2°C as rapidly
as possible.
Ill
Filter membranes for fecal coliform
counts consecutively and immediately
place them in their culture containers.
Insert as many as six culture containers
all oriented in the same way (i.e., all
grid sides facing the same direction)
into the sacks and seal. Tear off the
perforated top, grasp the side wires,
and twirl the sack to roll the open end
inside the folds of sack. Then submerge
the sacks with culture containers in-
verted beneath the surface of a 44. 5
+ 0. 2 C waterbath.
b Incubate for 22 + 2 hours.
5 Counting procedures
Examine and count colonies as follows:
a Use a wide field binocular dissecting
microscope with 5 - 10X magnification.
b Low angle lighting from the side is
advantageous.
c Fecal coliform colonies are blue,
generally 1-3 mm in diameter.
d Record the colony counts on the
data sheet, and report the fecal
coliform count per 100 ml of sample.
(I, D, 17 illustrates method)
TESTS FOR FECAL STREPTOCOCCAL
GROUP-MEMBRANE FILTER METHOD
A 48 hour incubation period on a choice of
two different media, giving high selectivity
for fecal streptococci, are the distinctive
features of the tests.
•'•'Prepare 1% solution of rosolic acid in 0. 2 N NaOH. This dye is practically insoluble in water.
11-11
-------
Dtailed Membrane Filter Methods
Test for Members of Fecal Streptococcal
Group based on KF-Agar
1 Preparation of the culture medium
a Formula: (The dehydrated formula
of Bacto 0496 is shown, but
equivalent constituents from other
sources are acceptable). Formula
is in grams per liter of reconstituted
medium.
Bacto proteose peptone #3 10. 0
Bacto yeast extract 10.0
Sodium chloride (reagent grade) 5. 0
g
g
g
g
g
g
g
Sodium glycerphosphate 10.0
Maltose (CP) 20.0
Lactose (CP) 1.0
Sodium azide (Eastman) 0.4
Sodium carbonate 0.636 g
(Na CO reagent grade)
Brom cresol purple 0. 015 g
(water soluble)
Bacto agar 20.0 g
b Reagent
2, 3, 5-Triphenyl tetrazolium
chloride reagent (TPTC)
This reagent is prepared by making
a 1% aqueous solution of the above
chemical passing it through a Seitz
filter or membrane filter. It can
be kept in the refrigerator in a
screw-capped tube until used.
c The dehydrated medium described
above is prepared for laboratory
use as follows:
Suspend 7. 64 grams of the dehydrated
medium in 100 ml of distilled water
in a flask with an aluminum foil
cover.
Place the flask in a boiling water-
bath, melt the dehydrated medium,
and leave in the boiling waterbath
an addional 5 minutes.
Cool the medium to 50°-60° C, add
1. 0 ml of the TPTC reagent, and
mix.
For membrane filter studies, pour
5-8 ml in each 50 mm glass or
plastic culture dish or enough to
make a layer approximately 1/8"
thick. Be sure to pour plates before
agar cools and solidifies.
For plate counts, pour as for standard
agar plate counts.
NOTE: Plastic dishes containing
media may be stored in a dark, cool
place up to 30 days without change
in productivity of the medium, pro-
vided that no dehydration occurs.
Plastic dishes may be incubated in
an ordinary air incubator. Glass
dishes must be incubated in an
atmosphere with saturated humidity.
2 Apparatus, and materials as given in
Table 1.
3 General procedure is as given in I.
Special instructions
a Incubate 48 hours, inverted with
100% relative humidity after
filtration.
b After incubation, remove the
cultures from the incubator, and
count colonies under wide field
binocular dissecting microscope,
with magnification set at 10X or
20X. Fecal streptococcus colonies
are pale pink to dark wine-color.
In size they range from barely
visible to approximately 2mm in
diameter. Colorless colonies are
not counted.
c Report fecal streptococcus count
per 100 ml of sample. This is
computed as follows:
No. fecal streptococci per 100 ml =
No. fecal streptococcus colonies
Sample filtration volume in ml
X100
11-12
-------
Detailed Membrane Filter Methods
B Verification of Streptococcus Colonies
1 Verification of colony identification
may be required in waters containing
large numbers of Micrococcus orga-
nisms. This has been noted
particularly with bathing waters, but
the problem is by no means limited to
such waters.
2 A verification procedure is described
in "Standard Methods for the Examination
of Water and Wastewater" 14th ed.
(1975). The W0rker should use
this reference for the step-by-
step procedure.
IV PROCEDURES FOR USE OF MEMBRANE
FILTER FIELD UNITS
A Culture Media
1 The standard coliform media used with
laboratory tests are used.
2 To simplify field operations, it is
suggested that the medium be sent to
the field, preweighed, in vials or
capped culture tubes. The medium
then requires only the addition of a
suitable volume of distilled water-
ethanol prior to sterilization.
3 Sterilization procedures in the field
are the same as for laboratory methods.
4 Laboratory preparation of the media,
ready for use, would be permissible
provided that the required limitations
on time and conditions of storage are
met.
B Operation of Millipore Water Testing Kit,
Bacte riological
1 Supporting supplies and equipment are
the same as for the laboratory
procedures.
2 Set the incubator voltage selector
switch to the voltage of the available
supply, turn on the unit and adjust as
necessary to establish operating
incubator temperature at 35 + 0.5°C.
3 Sterilize the funnel unit assembly by
exposure to formaldehyde or by
immersion in boiling water. If a
laboratory autoclave is available, this
is preferred.
Formaldehyde is produced by soaking
an asbestos ring (in the funnel base)
with methanol, igniting, and after a
few seconds of burning, closing the
unit by placing the stainless steel
flask over the funnel and base. This
results in incomplete combustion of
the methanol, whereby formaldehyde
is produced. Leave the unit closed
for 15 minutes to allow adequate
exposure to formaldehyde.
4 Filtration and incubation procedures
correspond with laboratory methods.
5 The unit is supplied with a booklet
containing detailed step-by-step
operational procedures. The worker
using the equipment should become
completely versed in its contents and
application.
C Other commercially available field kits
should be used according to manu-
facturer's instructions. It is emphasized
that the required standards of performance
are manditory for field devices as for
laboratory equipment.
D Counting of Colonies on Membrane Filters
1 Equipment and materials
Membrane filter cultures to be
examined
Illumination source
Simple lens, 2X to 6X magnification
Hand tally (optional)
2 Procedure
a Remove the cultures from the
incubator and arrange them in
numerical sequence.
11-13
-------
Detailed Membrane Filter Methods
Set up illumination source as that
light will originate from an area
perpendicular to the plane of
membrane filters being examined.
A small fluorescent lamp is ideal
for the purpose. It is highly
desirable that a simple lens be
attached to the light source.
Examine results. Count all coliform
and noncoliform colonies. Coliform
and noncoliform colonies. Coliform
colonies have a "metallic" surface
sheen under reflected light, which
may cover the entire colony or may
appear only on the center.
Noncoliform colonies range from
colorless to pink or red, but do not
have the characteristic "metallic"
sheen.
Enter the colony counts in the data
sheets.
Enter the coliform count per 100 ml
of sample for each membrane having
a countable number of coliform
colonies. Computation is as follows:
No. coliform per 100 ml =
McCarthy, J. A., Delaney, J.E. and
Grasso, R.J. Measuring Coliforms
in Water. Water and Sewage Works.
1961: R-426-31. 1961.
This outline was prepared by H. L.
Jeter, Chief, Program Support
Training Branch, USEPA, Cincinnati,
Ohio 45268
Descriptors: Biological Membranes,
Coliforms, Fecal Coliforms, Fecal
Streptococci, Filters, Indicator Bacteria,
Laboratory Equipment, Laboratory Tests,
Membranes, Microbiology, Water Analysis
No. coliform colonies on MF
No. milliliters sample filtered
REFERENCES
X100
1 Standard Methods for the Examination of
Water and Wastewater. APHA, AWWA.
WPCF 14th Edition. 1975.
11-14
-------
COLONY COUNTING ON MEMBRANE FILTERS
I INTRODUCTION
On removal of membrane filter cultures from
the incubator, the worker has several tasks
to perform, leading to the reporting of results
of the bacteriological examination. These
steps, together with the selection and use of
associated equipment, are considered in this
discussion. The following topics are included:
A Precautions on removal of membrane
filter cultures from the incubator.
B Selection of the best membrane filter for
colony counting (when more than one
membrane filter per sample was prepared,
representing a graded series of sample
increments.)
C Use of grid systems on filter surfaces as
counting aids.
D Recognition and counting of desired
colonies, including selection and use of
optical equipment.
E Calculations for reporting number of test
organisms per 100 ml of sample.
II REMOVAL OF CULTURE FROM
INCUBATOR
A Incubation time and temperature recom-
mendations should be closely adhered to.
This applies particularly to total coliform
counts. Some of our earlier training
manuals have suggested counting of colonies
after as few as 16 hours of incubation at
35°C. Currently, 22+2 hours is preferred.
B All membrane filter cultures should be
incubated in the inverted position, with
measures to avoid loss of culture medium
through leakage or evaporation. Some-
times an excessive amount of culture
medium is applied initially, or additional
moisture finds its way into the culture
container during incubation. In such
cases, when the culture is removed
from the incubator, it should be turned
"right side up" in such a way as to avoid
flooding the filter with excess liquid. If
excessive liquid is present, open the
culture container cautiously, and pour
off the excess.
C Drying Filters Before Colony Counts
1 Some workers advise opening all
cultures (especially total coliform tests
when Endo-type media are used) for
a short time (15 minutes to 1 hour)
for partial drying of coliform colonies
before counting. Advocates of this
step report that the typical surface
sheen characteristic of coliform
colonies is improved by this step.
2 Use of preliminary drying procedures
is a matter of personal preference.
In the opinion of the writer, the benefit
of preliminary drying is at best debat-
able, and at worst, may interfere with
subsequent study of the bacterial
colonies. Correct use of acceptable
lighting and optical equipment is a
far more important factor in ease and
accuracy of recognition of differentiated
colonies.
Ill SELECTION OF ACCEPTABLE MEMBRANE
FILTER CULTURE FOR EXAMINATION
A Non-Quantitative Tests
In bacteriologic examination of treated
waters, where waters meeting require-
ments result in development of very few
or no coliform colonies, the typical filtra-
tion volume is 100 ml, and but one filtra-
tion is made per sample. In this case,
there is no problem: the one membrane
NOTE: Mention of commercial products and manufacturers does not imoly endorsement bv the
Office of Water Programs, Environmental ProtectionAgency.
W. BA. mem. 85a. 8. 77
12-1
-------
Colony Counting on Membrane Filters
filter preparation is the basis of bacteri-
ologic evaluation of the sample.
B Quantitative Tests
1 When the bacteriological water quality
standard is for some fixed limiting
value, such as 70 per 100 ml for shell-
fish waters, again only a single sample
filtration volume may be used. In such
a case, the filtration of a single portion
of 50 ml will show directly whether the
water meets bacteriologic standards,
or if the limiting standard is being
exceeded.
9, On the other hand, if the objective of the
test was to show how many coliforms
were present per 100 ml of sample,
then it is necessary to filter a series of
sample increments from each sample,
each increment being placed on a separate
membrane filter. At the end of the
incubation period, the series of mem-
brane filters representing each sample
must be inspected, with selection of the
membrane filter bearing the number of
colonies most suitable for reporting
quantitative results. This is summarized
in Table 1. below:
The lower limit of 20 is set arbitrarily,
as a number below which statistically
valid results become increasingly
questionable with smaller numbers of
colonies. The upper limits represent
numbers above which interference from
colony crowding, deposition of extrane-
ous material, and other factors appear
to result in increasingly questionable
results. It is emphasized that these
limiting values are empirical, based
on laboratory observations alone, and
do not represent results of theoretical
calculations. It follows that it is quite
possible, with some sample sources,
to obtain acceptable quantitative results
with colony counts higher than the re-
commendations, but the minimum limit
of 20 colonies appears to apply to the
majority of sample sources. "
If no membrane filter bears a number
of colonies within the recommended
limits for the test, the worker has a
choice between - a) collecting a new
sample and repeating the test; and
b) using whatever results actually were
obtained, reporting an "educated guess"
as to the number of organisms per
100 ml. In the latter case, it is most
Table 1. NUMBERS OF COLONIES ACCEPTABLE
FOR QUANTITATIVE DETERMINATIONS
Test
Total coliform
Fecal coliform
Fecal streptococcus
Colony Counting Range
Minimum Maximum
20
20
20
80
60
100
Remarks
200 limit overall
12-2
-------
Colony Counting on Membrane Filters
IV
strongly urged th;it each result of this
type be specifically identified with a
qualifying statement, such as "Estimated
count, based on non-ideal colony density
on filter. "
Sometimes two or more filters, of a
series of filtration volumes from a
sample, produce colony counts within
the recommended counting range.
Colony counts should be made on all
such filters. See Section VI of this
outline for calculations based on such
results. These problems may arise
from the selection of a too-close range
of sample filtration volumes, from
colony differentiation failures related
to overcrowding on the filters, or from
physico-chemical interference with
colony development related to material
in the sample deposited in or on the
filters.
USE OF GRID SYSTEMS IN COLONY
COUNTS
Most manufacturers provide grid-imprinted
membrane filters for bacteriologic use.
The ink used in such filters must be bio-
chemically inert to the test organisms,
and, of course, must be applied in such a
manner as not to degrade the quality of the
filter. Examples of sucn gridding have
appeared from various manufacturers as
follows:
1 ... effective filtering area subdivided
into squares equal to 1/100 the effective
filtering area (when a filtering unit
with funnel-diameter of 35 mm is used).
2 ... grid markings which subdivide
the effective filtering area into squares
equal to 1/100 the effective filtering
area (9. 6
filters).
for 47 mm diameter
3 ... filters subdivided so that each
square of the grid represents 1/60 of
the effective filtering area.
B Some special studies may require use of
membrane filters without grid markings.
For example, the ink in some filters pre-
vents growth of Brucella melitensis. In
such cases it may be necessary to impro-
vise a viewing grid which can be placed
over the culture after incubation and
colony development.
C Applications of Grids
1 The grid dimensions are of no particular
significance in colony counting, provided
that their size permits easy and con-
tinuous orientation in counting of colonie's.
To be sure, a rough estimate of the total
number of differentiated colonies on a
filter is possible by counting a repre-
sentative number of squares and multi-
plying colony count by the appropriate
factor. For example, with many
filters, colonies in ten squares can be
counted, multiplied by 10, and the pro-
duct is a rough estimate of the total
number on the entire filter. It is em-
phasized that such procedure is for
rough estimates only, and should not be
condoned in quantitative work with
membrane filters.
2 The primary usefulness of the grid
system is for orientation during the
counting procedure. Some colonies
will touch lines on a grid system, and
a uniform practice must be established
to avoid missing some colonies or
counting others twice. The procedure
used by the writer is as follows:
a Counts are made in an orderly back-
and-forth sweep, from top to bottom
of the filter. See Figure 1.
b Inevitably, some colonies will be in
contact with grid lines. A suggested
routine procedure for counting colonies
in contact with lines is indicated in
Figure 2. Colonies are counted in
the squares indicated by the arrows,
and no effort is made to decide
whether "most of the colony" is in
one or the other square.
12-3
-------
Colony Counting or Membrane Filters
A
A r1
' /
Figure 1. The dashed circle indicates the
effective filtering area. The dashed back-
and-forth line indicates the colony counting
pathway.
V COUNTING OF COLONIES
A Equipment
1 A hand-tally is a useful device while
counts are being made.
2 Optical assistance in colony counts is
strongly recommended. Dependence
on naked-eye counts often results in
too-low results.
a Preferably, use a wide-field binocu-
lar dissecting microscope with
magnification of 10X or 15X.
b Optionally, but less desirably, a
simple lens with magnification at
least 5X can be used, provided that
acceptable illumination also is
present.
3 Lighting equipment
a For coliform counting., a large light
source is mandatory. Fluorescent
lamps in housings permitting place-
ment close to and as directly as
possible over the membrane filter is
the best lighting arrangement known
to the writer. Incandescent lamps,
whether simple light bulbs in a table
Figure 2. Enlarged portion of grid-marked
square of filter, with various ways colonies
can be in contact with grid-lines. Colonies
are counted in squares indicated by the arrows.
lamp or in elaborate microscope
lamp housings, are not satisfactory
for coliform colony counting on
membrane filters with Endo-type
media.
b For fecal coliforms or fecal strep-
tococci, the lighting requirements
are not so severe; in this case almost
any sufficiently bright light source,
which can be placed above the filter
(either at a high or at a low angle)
will suffice.
4 Lighting arrangement and counting
a As above, for coliform counting,
the fluorescent lamp should be at a
high angle (as nearly as possible
directly over the membrane filter)
so placed that an image of the light
source is reflected off the colony
surfaces into the microscope lens
system. Properly placed, the light
will demonstrate the "golden
metallic" surface luster of coliform
colonies, which may cover the entire
colony, or may appear only in an
area in the center of the colony.
The worker must learn to recognize
the difference between the typical
golden sheen of coliform colonies
and the merely shiny surface of non-
coliform colonies.
12-4
-------
Colony Counting or Membrane Filters
b Other types of colonies (fecal coli-
forms, fecal streptococci, etc.) do
not require such rigid control of the
light source. Low-angle lighting
can be helpful, to give a relief of
the colony profile from the colony
surface. This is valuable with small
colonies, such as frequently en-
countered in streptococcal studies.
In such cases, almost any light
source is acceptable, provided that
it is bright enough and that it is
applied from somewhere above the
membrane filter.
c The typical appearance of various
types of colonies is related to the
culture media applied; therefore,
this is not discussed in detail at this
point. See the outlines on culture
media and on laboratory procedures
for specified indicator organisms
for such information.
fecal streptococci), bacterial counts
always are reported in numbers per 100
ml. In standard practice, results are
expressed to two significant figures. For
example, if the calculation indicates
75, 400,or even 75, 444 organisms per 100
ml, the results would be reported as
75, 000 per 100 ml in each case. (The
digits 7 and 5 are the significant figures;
the three zeros only locate the decimal
point. )
When "total" bacterial counts are reported,
common practice is to report in number
per ml, not the number per 100 ml.
Quantitative work on enteric pathogens is,
at this time, limited to reporting of
occurrence of designated enteric pathogens,
correlated with measured density of pollu-
tion indicating bacterial groups. At such
time as the numerical determination of
enteric pathogens becomes feasible, it is
anticipated that reports will be in terms
of count per 100 ml, or even larger volume
units.
d In colony counting, count all colonies
individually, even if they are in
contact with each other (this is con-
trary to usual practice in colony
counting in agar cultures in Petri
dishes). Such colonies are recog-
nized quite easily when a microscope
is used for colony counting as re-
commended. Colonies which have
grown into contact almost invariably
show a very fine line of contact. The
worker must learn to recognize the
difference between two or more
colonies which have grown into con-
tact with each other, and single,
irregularly shaped, colonies which
sometimes develop on membrane
filters. Such colonies almost in-
variably are associated with a fiber
or particulate material deposited on
the filter, and tend to develop along
a path conforming to the shape and
size of the fiber or particulates.
VI CALCULATIONS
A Counting Units
1 In reporting densities of indicator
organisms (coliforms, fecal coliforms,
B Typical Calculations
1 Select the membrane filter bearing the
acceptable number of colonies for re-
porting, and calculate indicators per
100 ml according to the general formula:
No. indicator organisms per 100 ml =
No. colonies of indicator organism
No. ml of sample filtered
2 Example:
a Assume that for a total coliform
count, volumes of 50, 15, 5, 1. 5,
and 0. 5 ml produced coliform colony
counts of 200, 110, 40, 10, and 5,
respectively.
b First, the worker actually would not
have counted coliform colonies on
all these filters. He would have
selected, by inspection, the mem-
brane filter(s) most likely to have
20-80 coliform colonies, limiting
actual counting to such colonies
(this does take some practice and
skill in making quick estimates, but
comes with experience).
12-5
-------
Colony Counting or Membrane Filters
c Having selected the membrane filter
probably most useful for reporting
purposes, coliform colonies are
counted according to accepted pro-
cedures, and the general formula
is applied:
40
Coliforms per 100 ml = -r- X 100
0
Coliforms per 100 ml = 800
C Special Situations in Calculating Densities
of Indicator Organisms
1 Assume a coliform count in which the
volumes of 1, 0.3, 0.1, 0.03, and
0.01 ml, respectively, produced coli-
form colony counts of TNTC, TNTC,
75, 30, and 8, respectively.
a Here, two sample volumes resulted
in production of coliform colonies
in the acceptable counting range.
k Suggestion: Compile the filtration
volumes and colonies from both
acceptable filters, as follows:
Volume, ml
0. 1
0.03
0. 13
Count
75
30
105
Calculate coliforms per 100 ml from
the composite result:
Coliforms per 100 ml =
105
X 100
0. 13
Coliforms per 100 ml = 81,000
2 Assume a coliform count in which
sample volumes of 1, 0.3, and . 01 ml
produced colony counts of 14, 3, and
0, respectively.
a Here, no colony count falls within
recommended limits.
b Suggestion: Calculate on the basis
of the most nearly acceptable value,
and report with qualifying remark,
thus:
Use 14 colonies from 1 ml of sample:
14
1.0
X 100 = 1400
Report: "Estimated Count, 1400
per 100 ml, based on non-ideal
colony count".
3 Assume a coliform count in which the
volumes I, 0.3, and 0.01 ml produced
coliform colony counts of 0, 0, and 0,
respectively.
a Here, no actual calculation is possible,
even for "estimate" reports.
b Suggestion: Calculate the number
of estimated coliforms per 100 ml
that would have been reported if
there had been 1 coliform colony
on the filter representing the largest
filtration volume, thus:
Use 1 colony, and 1 ml: - X 100 = 100
Report: "Less than 100 coliforms
per 100 ml".
4 Assume a coliform count in which the
volumes of 1, 0. 3, and 0. 01 ml pro-
duced coliform colony counts of TNC,
150, and 110 colonies.
a Here, all colony counts are above
the recommended limits.
b Suggestion: Use Example 2, above,
and report an estimated count based
on non-ideal colony counts:
110
0.01
X 100 = 1,100,000
Report; "Coliform count estimated
at 1, 100,000 per 100 ml, based on
non-ideal colony count".
12-6
-------
Colony Counting or Membrane Filters
Assume that, in Example 4, the volumes
of 1. 0, 0.3, and 0. 01 ml, all produced
too many coliform colonies to show
separated colonies, and that the labora-
tory bench record showed TNTC (Too
Numerous to Count).
Suggestion: Use 80 colonies as the
basis of calculation with the smallest
filtration volume, thus:
80
0.01
X 100 = 800,000
Report: "> 800, 000 coliforms per 100
ml sample. Filters too crowded. "
VII CONCLUSION
The foregoing discussion has presented a
number of factors which determine the quan-
titative reliability of membrane filter results.
It cannot be too strongly emphasized that the
correct use of acceptable colony counting
equipment is one of the most important single
factors in successful application of membrane
filter methods. Here, there is perhaps a
greater exercise of personal skill and judgment
than in any other aspect of membrane filter
methodology. There is no substitute for prac-
tice and experience, supported by liberal use
of supporting colony verification studies, to
produce a skilled worker in colony counts
on membrane filters.
This outline was prepared by H. L. Jeter,
Chief, Program Support Training Branch,
USEPA, Cincinnati, Ohio 45268
Descriptors: Filters, Membrane, Bacteria,
Microorganisms, Measurement Testing
Procedures
12-7
-------
VERIFIED MEMBRANE FILTER TESTS
I INTRODUCTION
A The purpose of a verified membrane filter
test procedure is to establish the validity
of colony differentiation and interpretation
in the test being applied. Specifically, a
verified membrane filter test may prove
useful 1) as a self-training device for new
workers, 2) as a research tool in evalu-
ation of new membrane filter media and
procedures, or 3) to provide supporting
evidence of colony interpretation in cases
where the analytical results may be subject
to professional or official challenge.
B Reduced to essentials, a verified mem-
brane filter test consists of 1) interpre-
tation of the colonies appearing on a
selective, differential medium, 2) re-
covery of purified bacterial cultures from
differentiated colonies, and 3) application
II
of supplemental test procedures to
determine the validity of the original
interpretation of the membrane filter
colonies.
In this discussion, primary attention is given
to a verified membrane filter coliform test.
In addition, verification procedures are
presented for members of the fecal coli-
form group and for fecal streptococci.
VERIFIED TEST FOR MEMBERS OF THE
COLIFORM GROUP
Anabbreviatedprocedure corresponds to the
Confirmed Test of Standard Methods through
use of lactose broth (or lactose lauryl
tryptose broth) followed by confirmation
in brilliant green lactose bile broth. The
procedure is shown diagrammatically as
follows:
10 - 20 sheen colonies from membrane
filter (each tested separately)
•*
Lactose (or lauryl tryptose) broth
Incubate 24 hours at 35°C
I
No gas
J/
Reincubate 24 hours at 35°C
±
Gas
4
No
Negative coliform test
Colony not coliform
~l
Gas-
Brilliant Green lactose bile broth
Incubate up to 48 hours at 35°C
No gas
Negative coliform test
Colony not coliform
Gas
Positive coliform test
Colony was coliform
Diagram 1. ABBREVIATED COLIFORM VERIFICATION PROCEDURE
W. BA. mem. 83a. 8. 77
13-1
-------
Verified Membrane Filter Tests
B A more elaborate verification of membrane
filter test for coliforms resembles the
Completed Test of Standard Methods. The
test is started in exactly the same way as
the abbreviated test, and may be repre-
sented diagrammatically as a continuation
from the lactose broth stage of Diagram 1.
See Diagram 2.
C While the diagrams (1 and 2) are pre-
sented in terms of sheen colonies (inter-
preted as coliforms), the careful worker
also should subject a similarly represen-
tative number of non-sheen colonies
(judged to be noncoliforms) to the same
test procedure. This will reveal whether
the medium being studied fails to differ-
entiate appreciable numbers of colonies
which in reality are coliforms, even
though they did not demonstrate the
desired differential characteristic.
Ill VERIFICATION OF FECAL COLIFORM
TESTS ON MEMBRANE FILTERS
A The procedure described here is based
on the principle that, with use of m-FC
Broth and incubation in a water bath at
44. 5°C for 24 hours, fecal coliform
colonies on membrane filters develop a
blue color, (sometimes a greenish-blue).
Extraneous bacteria are believed to fail
to develop colonies, or else consist of
such colonies develop some color other
than the blue color of fecal coliforms
(colorless, buff- or brownish-color, or
even red colonies may develop on the
medium).
Gas-positive lactose broth tubes from
Diagram 1, above
v
Streak on eosin methylene blue agar
plates for colony isolation
Incubate 24 hours at 35°C
Transfer an isolated nucleated colony
(or at least two well-isolated representative
colonies in the absence of nucleated colonies)
to
I
lactose broth (or lactose
lauryl tryptose broth)
and to
.
nutrient agar slant
I
Incubate up to 48 hours
at 35°C
Incubate not more than 24 hours
at 35°C
Prepare Gram-stained smear and
examine under oil immersion
No gas
Negative test
(Colony was not
^oJ iform)
Gas
and
4.
Gram-negative, non-spore
forming rods, pure culture
based on morphology
V
Positive coliform test (Colony
originally selected was coliform)
Lack of any morpho-
logical feature
described on left
Negative coliform test
(Colony originally chosen
was not coliform)
Diagram 2. EXTENDED COLIFORM VERIFICATION PROCEDURE
13-2
-------
Verified Membrane Filter Tests
B The verified test for fecal coliforms
is indicated in Diagram 3, below:
10 - 20 blue colonies from membrane
filter (each tested separately)
Phenol red lactose broth (or lactose
broth or lauryl tryptose broth)
v
Incubate 24 hours at 35°C
X
No gas
4
Reincubate 24 hours
at 35°C
•v
Gas
No gas
Negative test for
fecal coliforms.
Colony was not a
fecal coliform colony
•f
Gas
EC Broth
4
Incubate 24 hours at 44. 5°C t 0. 5° C in a
water bath
V
No gas
Negative test for fecal
coliforms. Colony was
not a fecal coliform
Cfas
Positive test for
fecal coliforms.
Colony was fecal
coliform
Diagram 3. A VERIFICATION PROCEDURE FOR FECAL
COLIFORMS ON MEMBRANE FILTERS
IV VERIFICATION OF FECAL STREPTO-
COCCUS COLONIES ON MEMBRANE
FILTERS
A The procedure is used in the evaluation of
results from a medium similar to the
m-Enterococcus Agar (Slanetz) described
in the current edition of Standard Methods.
The membrane filter procedure utilizes
48 hour incubation at 35°C, and colonies
which are pink to red, either in their
entirety or only in their centers, are
regarded as fecal streptococci. Most
such colonies are 1-2 mm in diameter,
and some may be larger. Occasionally,
some samples may be encountered in
V
which numerous extremely small colonies,
approximately 0. 1 mm in diameter, are
present in great numbers. Almost in-
variably, these are not fecal streptococci.
See diagram 4 for a representation of
a verification test.
CALCULATIONS BASED ON VERIFICATION
STUDIES
A A percent verification can be determined
for any colony-validation test:
Percent verification =
No. of colonies meeting verification test
No. of colonies subjected to verification
X 100
13-3
-------
Verified Membrane Filter Tests
10 - 20 Pink to red colonies from
membrane filter (each tested separately)
[lO . 20 PINK TO RED COLONIES FROM
MEMBRANE FILTER (EACH TESTED SEPARATELY)]
PINK-RED COLONY
GROWTH IN BRAIN-HEART INFUSION BROTH WITHIN 1 DAYS AT 45 C AND
GROWTH AT 45 C AND 10 C
CONFIRM WITH GROWTH IN 65% NoCI AND pH 9.5 IN
BRAIN- HEART INFUSION BROTH AND REDUCTION OF 0 1 % METHYLENE BLUE
• REDUCTION OF K2TeO_ TTC AND ^____
FERMENTATION OF 0- SORBlTOL AND GLYCEROL
STARCH HYDROLYSIS
GROWTH AT 45 C ONLY
STARCH HYDROLYSIS
HYDROLYSIS OF GELATIN
S FAECAUS S FAECALI5 AND
VAfi. LIOUEFACIENS S. FAECALIS VAR. ZYMOGENES
HEMOLYSIS
S \.
POSITIVE NEGATIVE
I i
ATYPICAL S FAECALIS PEPTONIZ ATION '
(VEGETATION SOURCE) LITMUS MILK
FERMENTATION OF l-ARABINOSE
S. FAECIUM
POSITIVE
1
S fAECAllS VAR
LIOUEFACIENS
(INS
LACTOSE FERMENTATION
ACIO ONLY NO CHANGE
\ \
S BOVIS S. EOUINUS
(LIVESTOCK AND POULTRY
SOURCES)
Diagram 4. FLOW SHEET AND SEQUENCE OF TESTS TO PERFORM VERIFICATION
STUDIES ON COLONIES BELIEVED TO BE FECAL STREPTOCOCCI
Example: Twenty-five sheen colonies
on Endo-type membrane filter medium
were subjected to verification studies
shown in Diagram 1. Twenty-two of
these colonies proved to be coliforms
according to provisions of the test:
22
Percent verification = — X 100
^0
= 88
B A percent verification figure can be
applied to a direct membrane filter
count per 100 ml to determine the veri-
fied membrane filter count per 100 ml
of the test organism.
Verified count per 100 ml
of the test organism
Percent verification ., count per 100 ml
100 of test organism
Example: For a given sample, by a
direct membrane filter test, the fecal
coliform count was found to be 42, 000
per 100 ml. Supplemental studies on
selected colonies showed 92% verification.
Verified fecal
coliform count
Rounding off;
92
X 42,000
100
= 0.92 X 42,000
= 38,640
= 39,000 per 100 ml
13-4
-------
Verified Membrane Filter Tests
C A percentage of false-negative tests also
can be determined (See II, C)
Percent false negative =
No. "negative" colonies found positive y .„„
Total No. "negative" colonies tested
Example: On a total coliform test, 25
nonsheen (coliform negative) colony
types were subjected to the coliform
verification procedure shown in Diagram
1. Two of these colonies proved to be
coliform colonies.
VII
Percent false negatives = —
^b
100
VI
A
B
SOME APPLICATIONS OF PERCENT
VERIFICATION CALCULATIONS
In comparisons between two or more differ-
ent membrane filter media, the medium
which has the highest percentage of veri-
fication, and the lowest percentage of
false negatives (based on a broad range
of sample types and sources) is the better
medium.
In productivity comparisons between two
or more different membrane filter media,
the medium which produces the highest
verified membrane filter counts per 100
ml (based on a broad range of sample
types and sources) is the better medium.
The worker is cautioned NOT to apply
percentage of verification determined
from one sample, to other samples. For
example, do not determine a percentage
verification on m-Endo broth for a sample
taken from the Ohio River on September 6,
and then seek to apply that percentage
verification to another coliform determina-
tion from the Little Miami River, on the
same date. Even the application of the
verification percentage to another Ohio
River sample, either on the same date
from a different station, or on another
date from the same station, should be
undertaken with great caution. Such
application of verification percentages
from one sample to another should be
taken only after sufficient studies have
been made demonstrate the suitability
of such a procedure.
USE OF VERIFICATION STUDIES IN
MF-MPN COMPARISONS
A Comparisons of data obtained from MF
versus MPN methods have been the source
of great concern to microbiologists. For
the current basis of comparisons, see
Standard Methods (either llth or 12th
edition) "-- with a proviso that it should
be used for determining the potability
of drinking water only after parallel
testing had shown that it afforded infor-
mation equivalent to that given by the
standard multiple-tube test. "
B Some workers have sought to apply this
requirement on the basis of statistical
calculations, based on comparisons of
numerical values from membrane filter
tests with numerical values obtained from
multiple-tube tests. Further study of this
problem, and methods different workers
have applied to the problem, can be made
Dn the basis of the appendpd reference Hst
C Numerical comparisons between raw or
verified membrane filter results on split
samples, compared with multiple -tube
results, also should take into account the
question of the reliability of the multiple -
tube test. The numerical results of the
Completed Test for conforms, for example,
can be compared with the results of the
Confirmed Test, to determine a percentage
of verification for the multiple -tube test:
Percent verification =
Completed Test Coliforms per 100 ml
Confirmed Test Coliforms per 100 ml
Example: On a given sample, the test
was carried to the Completed Test
stage. Afterward, both a Confirmed
Test and a Completed Test coliform
result were obtained, consisting of
13-5
-------
Verified Membrane Filter Testg
Table 1. VALIDITY OF MF AND MPN "CONFIRMED TEST'
Number
MF Coliform Test
Source of Percent
supplies Minimum Maximum verified
Wells - Springs
Lakes - Lagoons
Creeks
Rivers
Sewage
Totals
16
23
19
22
11
91
1.0 7,600 96.6
1.0 420,000 79.6
32 260,000 75.8
320 890,000 69.7
1,400,000 28,000,000 68.6
78. 1
MPN Confirmed Test
Percent
Minimum Maximum verified
7.0 11,000 64.
79 490,000 70.
120 460,000 66.
700 350,000 75.
460,000 49,000,000 73.
70.
6
9
4
7
8
3
*A11 coliform values are per 100 ml of sample
49, 000 per 100 ml for the Confirmed
Test and 33, 000 per 100 ml for the
Completed Test.
Percent verification -
33,000
49,000
= 67
X 100
See Table 1 for some studies of MF
verification studies, and parallel
multiple-tube verification studies
(Confirmed Test carried to Completed
Test). These studies have been con-
ducted in research laboratories of this
Center, and demonstrate the difficulty
and problems associated comparative
evaluation of membrane filter versus
multiple-tube methods. The student is
invited to study this table at leisure.
REFERENCES
1 Delaney, I.E., McCarthy, J.A. and
Grasso, R.J. Measurement of E^ coll
Type I by the Membrane Filter. Water
and Sewage Works, 109, 289-294. 1962.
2 Geldreich, E. E. , Clark, H.F., Huff,
C. B. and Best, L.C. Fecal-Coliform-
Organism Medium for the Membrane
Filter Technique. JAWWA 57, 208-
214. 1965.
13-6
3 Geldreich, E.E., Jeter, H.L. and Winter,
J.A. Technical Considerations in
Applying the Membrane Filter Pro-
cedures. In Press. 1966.
4 Hoffman, D. A. , Kuhns, J.H. , Stewart,
R.C. and Crossley, E.I. A Comparison
of Membrane Filter Counts and Most
Probable Numbers of Coliform in San
Diego's Sewage and Receiving Waters.
JWPCF 36, 109-117. 1964.
b McCarthy, J.A., Delaney, J,E. and
Grasso, R.J. Measuring Coliforms
in Water. Water and Sewage Works,
108, 238-243. 1961.
6 Thomas, H.A., Jr. and Woodward, R.L.
Estimation of Coliform Density by the
Membrane Filter and the Fermentat' v
Tube Methods. American Journal
Public Health 45, 1431-1437. 1955.
This outline was prepared by H. L. Jeter,
Chief, Program Support Training Branch,
USEPA, Cincinnati, Ohio 45268
Descriptors: Quality Control, Filters,
Membrane, Enteric Bacteria,
Microorganisms, Laboratory Tests
-------
COLLECTION AMU HANDLING OF SAMPLES FOR
BACTERIOLOGICAL EXAMINATION
I INTRODUCTION
The first step in the examination of a water
supply for bacteriological examination is
careful collection and handling of samples.
Information from bacteriological tests is
useful in evaluating water purification,
bacteriological potability, waste disposal.
and industrial supply. Topics covered
include: representative site selection,
frequency, number, size of samples,
satisfactory sample bottles, techniques of
sampling, labeling, and transport.
II SELECTION OF SAMPLING LOCATIONS
The basis for locating sampling points is
collection of representative samples.
A Take samples for potability testing from
the distribution system through taps.
Choose representative points covering
the entire system. The tap itself should
be clean and connected directly into the
system. Avoid leaky faucets because of
the danger of washing in extraneous
bacteria. Wells with pumps may be
considered similar to distribution systems.
B Grab samples from streams are frequently
collected for control data or application of
regulatory requirements. A grab sample
can be taken in the stream near the surface.
C For intensive stream studies on source
and extent of pollution, representative
samples are taken by considering site,
method and time of sampling. The
sampling sites may be a compromise
between physical limitations of the
laboratory, detection of pollution peaks,
and frequency of sample collection in
certain types of surveys. First, decide
how many samples are needed to be
processed in a day. Second, decide
whether to measure cycles of immediate
pollution or more average pollution.
Sites for measuring cyclic pollution are
immediately below the pollution source.
Sampling is frequent, for example, every
three hours.
A site designed to measure more average
conditions is far enough downstream for
a complete mixing of pollution and water.
Keep in mind that averaging does not
remove all variation but only minimizes
sharp fluctuations. Downstream sites
g may not nrcd to be so frequent.
Samples maybe collected 1/4, 1/2 and
3/4 of the stream width at each site or
other distances, dt'pending on survey
objeriJves. Often only one sample in the
channel of the stream is collected.
Samples wre usually taken near the surface.
D Samples from lakes or reservoirs are
frequently collected a< the draw off and
usually abou^' tin"1 samo depth and maybe
collected over (''if- onlift- suiface.
E Collect samples of bathing beach water
at locations and times where the most
bathers swim.
Ill NUMBER, FREQUENCY AND SIZE
OF SAMPLES
A For determining sampling frequency for
drinking water, consult the USPHS
Standards.
1 The total number, frequency, and site
are established by agreement with
either atat^ of PHS authorities.
2 The minimum number depends upon the
number of users. Figure 1 indicates
that the smaller populations call for
relatively more samples than larger
ones. The rmmbers on the left of the
graph refer to actual users and not the
population sbo"'ti by census.
3 In the event that colifo'-m limits of the
standard are exceeded, daily samples
must b-.1 taken a! the some site.
Examinations :-!n uld. continue until two
consecutive samples show coliform
level is S9l;sfautory. Such samples
are to be considered as special samples
and '-hall not b«3 included in the total
Hunjijer <>f ,;.unples examined.
4 Sampling programs described above
represent a minimum number which
may be increased by reviewing
authority.
W. BA. sa. Id.1.78
14-1
-------
Collections and Handling of Samples for Bacteriological Examination
B For stream investigations the type of
study governs frequency of sampling.
C Collect swimming pool samples when use
is heavy. The high chlorine level rapidly
reduces the count when the pool is not in
D
Population
served:
25 to 1,000
1,001 to 2,500
2, 501 to 3, 300
3, 301 to 4, 100
4, 101 to 4, 900
4, 901 to 5,800
5,801 to 6,700
6, 701 to 7, 600
7, 601 to 8, 500
8,501 to 9,400
9,401 to 10, 300
10, 301 to 11, 100
11, 101 to 12,000
12,001 to 12,900
12, 901 to 13, 700
13,701 to 14,600
14,601 to 15, 500
15, 501 to 16, 300
16, 301 to 17, 200
17, 201 to 18, 100
18, 101 to 18, 900
18, 901 to 19, 800
19, 801 to 20, 700 •""•—
20,701 to 21,500
21,501 to 22, 300
22,301 to 23, 200
23, 201 to 24, 000
24, 001 to 24, 900
24, 901 to 25, 000
25,001 to 28, 000
28,001 to 33, 000
33,001 to 37, 000
37,001 to 41, 000 •* —
41,001 to 46, 000 »•—
46,001 to 50, 000
50,001 to 54, 000
54,001 to 59,000
59,001 to 64, 000
64,001 to 70, OOq
70,001 to 76,000,
76,001 to 83, 000
83,001 to 90,000
Minimum number of
samples per month
4
- -- 2
3
__ 4
5
6
7
8
9
10
11
12
- 13
14
- 15
16
17
18
jg
-------------- ~~ 20
~~~~ 21
" "~ 22
"""" 23
"~"~""~""""~~ 24
__--___---_- 25
26
-------------- 27
28
29
- 30
- 35
- 40
45
50
55
60
65
70
75
• - 80
• - 85
90
use. Residual chlorine tests are
necessary to check neutralization of
chlorine in the sample.
Lake beaches may be sampled as required
depending on the water uses.
Population
served;
Minimum number of
samples per month
90,001 to 96, 000
96,001 to 111,000
111,001 to 130,000
130, 001 to 160,000
160,001 to 190,000
190,001 to 220,000
220,001 to 250,000
250,001 to 290,000
290,001 to 320,000
320,001 to 360,000
360, 001 to 410, 000
410, 001:to 450,000
450,001 to 500,000
500,001 to 550,000 ---
550,001 to 600,000
600, 001 to 720, 000 —
720,001 to 780,000
780,001 to 840,000 ---
840,001 to 910,000 ---
910, 001 to 970,000
970,001 to 1,050,000 -:
1,050,001 to 1, 140,000
1, 140,001 to 1,230,000
1, 230,001 to 1, 320,000
1, 320,001 to 1,420,000
1,420,001 to 1,520,000
1, 520, 001 to 1, 630, 000
1,630,001 to 1,730,000
1, 730,001 to 1,850,000
1,850,001 to 1, 970,000
1, 970,001 to 2,060, 000
2,060,001 to 2, 270,000
2,270,001 to 2,510, 000
2,510,001 to 2,750, 000
2, 750,001 to 3,020,000
3,020,001 to 3, 320,000
3, 320,001 to 3, 620, 000
3, 620,001 to 3, 960,000
3, 960,001 to 4, 310, 000
4,310,001 to 4,690,000
4, 690, 001 or more
95
100
110
120
130
140
150
160
170
180
i90
ilOO
210
220
230
240
250
-.60
i;70
280
290
iOO
330
340
350
360
370
380
390
400
410
420
430
440
450
460
470
480
500
FIGURE I
14-2
-------
Collection and Handling of Samples for Bacteriological Examination
E Salt water or estuarine beaches are
sampled as needed with frequency
depending on use.
F Size of samples depends upon examination
anticipated. Generally 100 ml is the
minimum size.
IV BOTTLES FOR WATER SAMPLES
A The sample bottles should have capacity
for at least 100 ml of sample, plus an
air space. The bottle and cap must be of
bacteriological inert materials. Resistant
glass or heat resistant plastic are
acceptable. At the National Training
Center, wide mouth ground-glass
stoppered bottles (Figure 2) are used.
All bottles must be properly washed and
sterilized. Protect the top of the bottles
and cap from contamination by paper or
metal foil hoods. Both glass and heat
1
\\\\\\\\\\\1
FIGURE 2
resistant plastic bottles may be
sterilized in an autoclave. Hold plastic
at 121°C for at least 10 minutes. Hot
air sterilization. 2 hours at 170°C, may
be used for dry glass bottles.
Add sodium thiosulfate to bottles intended
for halogenated vater s.-iniples. A quantity
of 0. 1 ml of a 10% solution provides 100
mg per liter concentration in a 100 ml
sample. This level shows no effect upon
viability or growth.
Supply catalogs List wide mouth ground
glass stoppered bottles of borosilicate
resistance glass, specially for water
samples.
V TECHNIQUE OF SAMPLE COLLECTION
Follow aseptic technique as nearly as possible.
Nothing but sample water must touch the inside
of the bottle or cap. To avoid loss of sodium
thiosulfate, fill the bottle directly and do not
rinse. Always remember to leave an air space.
A In sampling from a distribution system,
first run the faucet wide open until the
service line is cleared. A time of 2-3
minutes generally is sufficient. Reduce
the flow and fill the sample bottle without
splashing. Some authorities stress flaming
the tap before collection, but the use of this
technique is now generally considered as value-
less. A chlorine determination is often made
. on the site.
B The bottle may be dipped into some
waters by hand. Avoid introduction
of bacteria from the human hand and
from surface debris. Some suggestions
follow- Hold the bottle near the base
with one hand and with the other remove
the hood and cap. Push the bottle
rapidly into the water mouth down and tilt up
up towards the current to fill. A depth
of about 6 inches is satisfactory. When
there is no current move the bottle
through the water horizontally and away
from the hand. Lift the bottle from the
water, spill a small amount of sample
to provide an air space, and return the
uncontaminated cap.
14-3
-------
Collection and Handling of Samples for Bacteriological Examination
C Samples may be dipped from swimming
pools. Determine residual chlorine on
the pool water at the site. Test the
sample at the laboratory to check chlorine
neutralization by the thiosulfate.
D Sample bathing beach water by wading out
to the two foot depth and dipping the
sample up from about 6 inches below the
surface. Use the procedure described in
V. B.
E Wells with pumps are similar to
distribution systems. With a hand pumped
well, waste water for about five minutes
before taking the sample. Sample a well
without a pump by lowering a sterile
bottle attached to a weight. A device which
opens the bottle underneath the water
will avoid contamination by surface debris.
F Various types of sampling devices are
available where the sample point is
inaccessible or depth samples are desired.
The general problem is to put a sample
bottle in place, open it, close it, and
return it to the surface. No bacteria but
those in the sample must enter the bottle.
The J - Z sampler described by Zobell
in 1941, was designed for deep sea
sampling but is useful elsewhere (Figure)
3). It has a metal frame, breaking
device for a glass tube, and sample
bottle. The heavy metal messenger
strikes the lever arm which breaks
the glass tubing at a file mark. A
bent rubber tube straightens and the
water is drawn in several inches from
the apparatus. Either glass or collapsible
rubber bottles are sample containers.
Commercial adaptations are available.
Note the vane and lever mechanism on
the New York State Conservation
Department's sampler in Figure 4.
When the apparatus is at proper depth
the suspending line is given a sharp
pull. Water inertia against the vane
raises the stopper and water pours
into the bottle. Sufficient sample is
collected prior to the detachment of
the stopper from the vane arm allowing
a closure of the sample bottle.
The New York State Conservation
Department's sampler is useful for
shallow depths and requires nothing
besides glass stoppered sample bottles.
Reproduced with permission of the Journal
of Marine Research 4:3, 173-188 (1941) by
the Department of Health, Education and
Welfare.
14-4
-------
Collection and Handling of Samples for Bacteriological Examination
FIGURE 4
A commercial sampler is available
which is an evacuated sealed tube with
a. capillary tip. When a lever on the
support rack breaks the tip, the tube
fills. Other samplers exist with a
lever for pulling the stopper, while
another uses an electromagnet.
VI DATA RECORDING
A Information generally includes: date, time
of collection, temperature of water, locatio
of sampling point, and name of the sample
collector. Codes are often used. The '
location description must be exact enough
to guide another person to the site.
Reference to bridges, roads, distance to
the nearest town may help. Use of the
surveyors' description and maps are
recommended. Mark identification on the
bottles or on securely fastened tags.
Gummed tags may soak off and are
inadvisable.
B While a sanitary survey is an indispensable
part of the evaluation of a water supply, its
discussion is not within the scope of this
lecture. The sample collector could supply
much information if desired.
VI SHIPPING CONDITIONS
The examination should commence as soon
as possible, preferably within one hour. A
maximum elapsed time between collection and
examination is 30 hours for potable water
samples and 6 hours for other water samples
(time from collection to laboratory delivery).
An additional 2 hours is allowed from delivery
to laboratory to the completion of first-day
laboratory procedures. Standard Methods
(14th Edition) recommends icing of samples
between collection and testing.
VII PHOTOGRAPHS
A photograph is a sample in that it is evidence
representing water quality. Sample collectors
and field engineers may carry cameras to
record what they see. Pictures help the general
public and legal courts to better understand
laboratory data.
REFERENCES
1 APHA, AWWA, WPCF, Standard Methods
for the Examination of Water and Wastewater.
(12 Ed.) 1965.
2- Prescott, S.C,, Winslow, C.E.A., and
McCrady, M. H. Water Bacteriology. 6th Ed.,
368 pp.
1946.
John Wiley and Sons, Inc., New York.
Haney, P. D., and Schmidt, J. Representative
Sampling and Analytical Methods in Stream
Studies, Oxygen Relationships in Streams,
Technical Report W58-2 pp. 133-42. u. S.
Department of Health, Education and Welfare,
Public Health Service, Robert A. Taft Sanitary
Engineering Center, Cincinnati, Ohio. 1958.
Velz, C. J. Sampling for Effective Evaluation
of Stream Pollution. Sewage and Industrial
Wastes, 22:666-84. 1950.
14-5
-------
Collection and Handling of Samples for Bacteriological Examination
Bathing Water Quality and Health III for Bacteriological Analysis. Journal
Coastal Water. 134pp. U. S ' of Marine Research, 4:3:173-88. 1941
Department of Health, Education, and
Welfare, Public Health Service, Robert
A. Taft Sanitary Engineering, This outline was originally prepared by
Cincinnati, Ohio. 1961. A. G. Jose, former Microbiologist FWPCA
Training Activities, SEC and updated by
the Training Staff, National Training Center,
Zobeli, C. E. Apparatus for Collecting DTTB, MDS, WP0 EPA, Cincinnati, OH
Water Samples from Different Depths 45268
Descriptors: Equipment Microbiology,
Sampling, Water Sampling
14-6
-------
TESTING THE SUITABILITY OF DISTILLED WATER
FOR THE BACTERIOLOGY LABORATORY
I INTRODUCTION
A Standard Methods for the Examination of
Water and Wastewater (12th Edition) states;
"Only distilled water or demineralized
water which has been tested and found free
from traces of dissolved metals and bac-
tericidal and inhibitory compounds may
be used for the preparation of culture media
and reagents. Bactericidal compounds may
be measured by a biologic test procedure
. . . . " This outline describes a suitable
procedure.
B A need for such a test has been shown in
the lack of reproducibility of plate counts
and a possible cause of inconsistent re-
sults in split sample examinations.
IV REAGENTS
A Use reagents of the highest purity. Some
brands of potassium dihydrogen phosphate
(KH2PO4) have large amounts of impurities.
The sensitivity of the test is controlled in
part by the purity of the reagents employed.
1 Carbon source - Sodium citrate, reagent,
crystals (Na3C6H5O7 • 2H2O) 0.29 g
dissolved in 500 ml of redistilled water.
2 Nitrog_en_ source - Dissolve 0. 60 g of
ammonium sulfate, reagent, crystals,
(NH4)2SO4) in 500 ml of redistilled
water.
3 Salt mixture solution - Dissolve the
following compounds in 500 ml of re-
distilled water.
II THEORY OF THE TEST PROCEDURE
A Growth of Aerobacter aerogenes in a
chemically defined minimal growth medium.
The addition of a toxic agent or a growth
promoting substance will alter the 24 hr
population by an increase or decrease of
20% or more, when compared to a control.
Ill APPARATUS AND MATERIALS
A Glassware - rinse all glassware in freshly
redistilled water from a glass still. The
sensitivity of the test depends upon the
cleanliness of the sample containers,
flasks, tubes, and pipettes. Use only
borosilicate glassware.
B Culture - any strain of coliform IMViC
type —H- (A. aerogenes). This can be
easily obtained from any polluted river or
sewage sample.
Magnesium sulfate, reagent, crystals
(MgS0
0.26 g.
Calcium chloride, reagent, crystals
(CaCl2 • 2H20) 0. 17 g.
Ferrous sulfate, reagent, crystals
(FeSO • HO) 0.23 g.
Tt £t
Sodium chloride, reagent, crystals
(NaCl) 2.50 g.
Phosphate buffer solution - Use a 1 to
25 dilution of a stock phosphate solution
prepared by dissolving 34. 0 gm of
potassium dihydrogen phosphate
(KH2PO4) in 500 ml of distilled water,
adjusting to pH 7. 2 with 1 N NaOH and
diluting to 1 liter with distilled water.
Toxic control - dissolve 0.40 grams
CuSO4 • 5H2O in 100 ml of redistilled
water. Dilute 1:1000 for 1 mg per liter
Cu before use.
W.BA. lab. 12e. 8. 77
15-1
-------
Testing the Suitability of Distilled Water
B Sterilization of Reagents
Unknown distilled water sample - either
boil for one minute or sterilize by mem-
brane filtration.
Prepare reagents with redistilled water
heated to boiling for 1 to 2 minutes.
Phosphate buffer solution may be sterilized
by MF filtration or boiling.
C Solutions are useful up to two weeks when
stored at 5°C in sterilized glass stoppered
bottles. The salts solution must be stored
in the dark because sunlight results in
copious ferric ion precipitation. A slight
turbidity arising in the first 3-5 days
does not detract from the usefulness of
the reagents.
V PROCEDURE
A Collect 150 - 200 ml of water sample in a
sterile borosilicate glass flask and sterilize.
Label 3 flasks or tubes: A, B, and F.
Add water Samples and redistilled
water to each flask as indicated at the
bottom of the page.
B Add a suspension of Aerobacter aerogenes
(IMViC type --++) of^sTiclTd^lis'itirthat'e^ch
flask will contain 25-75 cells per ml.
Make an initial bacterial count by plating
a 1 ml sample in plate count agar. Incu-
bate tests A-F at 32° or 35°C for 20 - 24
hr. Make plate counts using dilutions of
1, 0.1, 0.01, 0.001 and 0.0001 ml.
VI PREPARATION OF A BACTERIAL
SUSPENSION
A Bacterial Growth
On the day prior to performing the distilled
water suitability test, inoculate a strain of
Aerobacter aerogenes onto a nutrient agar
slant with a slope of approximately 2-1/2
inches in length contained in a 125 mm X
16 mm screw cap tube. Streak the entire
agar surface to develop a continuous
growth film and incubate 18-24 hrs at
35°C.
B Harvesting Viable Cells
Pipette 1 - 2 ml of sterile dilution water
from a 99 ml water blank onto the 18 - 24
hr culture. Emulsify the growth on the
STANDARD TEST
Media
Reagents
Phosphate buffer (73+ 1)
Water, 1 mg per liter Cu
TOTAL VOLUME
Control
A
— ° 5
^25
1 5
X .
,™._?i n
30.0
Unknown
Dist. Water
B
2. 5
o t;
1 5
X
21.0
X
30.0
Toxic
Control
F
— 2.5
? 5
2 "i
1. 5
21.0
30.0
OPTIONAL TEST
H
OT
H
J
y
5
H
d.
n
f
Food
Available
C
Jl
/ \
/ \
X
0 S
1.5
•V"
21.0
5.0
30.0
Nitrogen
Source
D
A
i \
i \
— 2.5
— 1.5
X
21.0
2.5
30.0
Carbon
Source
E
IT
/ \
/ \
X
. b
n 5
1.5
X
21.0
2. 5
30.0
0
H
i
r
H
rn
H
OPTIONAL TEST
15-2
-------
Testing the Suitability of Distilled Water
FLK
[ 1-2 ml
slant
wash]
C
[-)
E
OPTIONA L
TEST
slant by gently rubbing the bacterial film
with the pipette, being careful not to tear
the agar, and pour the contents back into
the original 99 ml water blank.
C Dilution of Bacterial Suspension
Make a 1 - 100 dilution of the original
bottle into a second water blank, and a
further 1 - 100 dilution of the second
bottle into a third water blank, shaking
vigorously after each transfer. Then
pipette 0. 1 ml of the third dilution
(1:1, 000, 000) into each of the flasks A,
B, and F (see Standard Methods for
Examination of Dairy Products, 12th ed.).
This procedure should result in a final
dilution of the organisms to a range of
25-75 viable cells for each ml of test
solution.
D Verification of Bacterial Density
Variations among strains of the same
organism, different organisms, media,
and surface area of agar slopes will
possibly necessitate adjustment of the
dilution procedure to arrive at a specific
density range between 25 - 75 viable cells.
To establish the growth range numerically
for a specific organism and medium, make
a series of plate counts from the third
dilution to determine the bacterial density.
Then choose the proper volume from this
third dilution which when diluted by the 30
ml in the flasks A, B, and F will
C
contain 25-75 viable cells per ml. If
the procedures are standardized as to
surface area of the slant and laboratory
technique, it is possible to reproduce re-
sults on repeated experiments with the
same strain of microorganisms.
E Procedural Difficulties:
1 Chlorine or chloramine distilling over
into receiver. Distilled water should be
checked by a suitable quantitative pro-
cedure like the starch-iodide titration.
If chlorine is found, sufficient sodium thio-
sulfate or sodium sulfite must be added.
2 Unknown water sample stored in soft
glass containers or glass containers
without liners for metal caps.
3 Contamination of reagents of distilled
water with a bacterial background.
4 Incorrect dilution of A. aerogenes to
get 25 - 75 cells per ml.
5 Gross contamination of the sample de-
termined by the initial colony count be-
fore incubation.
F Calculation:
1 For growth inhibiting substances:
c_olony count per ml Flask B
colony count per ml Flask A
a Ratio 0.8 to 1.2 (inclusive) shows
no toxic substances.
15-3
-------
Testing for Suitability of Distilled Water
b Ratio less than 0. 8 shows growth
inhibiting substances in water sample.
2 For toxic control
colony count per ml Flask F _, ..
—z—^ —*- =-———-;- = Ratio
colony count per ml Flask A
OPTIONAL TEST
*For nitrogen and carbon sources that
promote growth**
colony count per ml Flask C _ ,.
—=,— -^- z——:—:—T- = Ratio
colony count per ml Flask A
*For nitrogen sources that promote
growth**
colony count per ml Flask D _ ,.
- = Ratio
colony count per ml Flask A
*For carbon sources that promote
bacterial growth**
colony count per ml Flask E _ R
colony count per ml Flask A
G Interpretation of Results:
1 The colony count from Flask A after
20 - 24 hours, at 35° C will depend on
the number of organisms initially
planted in Flask A and on the strain of A.
aero genes used in the test procedures.
This is the reason the control Flask A
must be run for each individual series
of tests. However, for a given strain
of A. aerogenes under identical
environmental conditions, the terminal
count should be reasonably constant
when the initial plant is the same.
Thus, it is essential that the initial
colony count on Flask A and Flask B
should be approximately equal to secure
accurate data.
When the ratio exceeds 1.2, it may be
assumed that growth stimulating sub-
stances are present. However, this
procedure is an extremely sensitive
test and ratios up to 3.0 would have
little significance in actual practice.
Therefore, Test C, D, and E do not
appear necessary except in special
circumstances, when the ratio is
between 1.2 and 3.0.
Usually Flask C will be very low and
flasks D and E will have a ratio of less
than 1. 2 when the ratio of Flask B/
Flask A is between 0. 8 and 1.2. The
limiting factors of growth in Flask A
are the nitrogen and organic carbon
present. An extremely large amount
of ammonia nitrogen with no organic
carbon could increase the ratio in
Flask D above 1. 2 or the absence of
nitrogen with high carbon concentration
could give ratios above 1. 2 in Flask
E with an A/B ratio between 0. 8 and
1.2.
A ratio below 0. 8 indicates the water
contains toxic substances and this ratio
includes all allowable tolerances. As
indicated in item 2 (above), the 1. 2
ratio could go as high as 3. 0 without
any undesirable results.
We are unable to recommend corrective
measures in specific cases of defective
distillation apparatus. However, a
careful inspection of the distillation
equipment and a review of production
and handling of the distilled water
should enable the local laboratory
personnel to correct the cause of the
difficulty.
*Do not attempt to calculate ratios, 3, 4, or
5 when ratio 1 indicates a toxic reaction.
** Ratio in excess of 1.2 indicates available
source for bacterial growth.
15-4
-------
Testing for Suitability of Distilled Water_
CASE EXAMPLES
Test results for various distilled water samples
SOURCE
1
2
3
4
5
6
TEST
COUNT
< 100
74, 000
18, 000
21, 000
310, 000
850, 000
CONTROL
COUNT
120,000
170, 000
14, 000
14, 000
60, 000
37, 000
RATIO
0.4
1.3
1.5
5.2
22.9
INTERPRETATION
Toxic Substance
Toxic Substance
Excellent water
Excellent water
Growth Substance
Growth Substance
REFERENCES
1 Standard Methods for Examination of Water
and Wastewater. 12th Edition. 1965.
p 578.
2 Geldreich, E. E. and H. F. Clark. Dis-
tilled Water Suitability for Microbiologi-
cal Applications. Journal Milk and Food
Technical. In Press. 1965.
This outline was orenar-ori bv E. E
Chief Bacteriologist, Water supply Programs
Division, WPO, EPA, Cincinnati, OH 45268.
Descriptors: Bacteria, Microbiology,
Laboratory, Water Supply, Distillation,
Water Quality Control
15-5
-------
RESIDUAL CHLORINE AND TURBIDITY
I. INTRODUCTION
The Interim Primary Drinking Water Regulations (Federal Register, December 24,
1975) permits the options of substitution of up to 75 percent of the bac-
teriological samples with residual chlorine determinations. Any community
or non-community water system may avail themselves of this option with
approval from the State based upon results of sanitary surveys. Residual
chlorine determinations must be carried out at the frequency of at least
four for each substituted microbiological sample.
Since many potable water plants carry out their own microbiological deter-
minations, it will be necessary that these laboratories be certified for
the bacteriological parameters. Residual chlorine determinations may be
carried out by any person acceptable to the State and the analytical
method and techniques used must be evaluated in some manner to assure that
reliable information is obtained.
Since the presence of high turbidity can interfere with the disinfection
capability of chlorine, a maximum allowable limit has been set for turbidity
as follows:
A. One turbidity unit (TU) as determined by a monthly average except
that five or fewer turbidity units may be allowed if the supplier
of water can demonstrate to the State that the higher turbidity
does not
1. Interfere with disinfection,
2. Prevent maintenance of residual of disinfectant throughout
distribution system, or,
3. Interfere with microbiological determinations.
B. Five turbidity units based on an average of two consecutive days.
The Criteria and Procedures Document for Water Supply Laboratory Certifi-
cation suggests that some quality control guidelines be instituted for
the residual chlorine and turbidity measurements at the State level for
the purpose of ensuring data validity for these critical measurements.
In response to public comments regarding the proposed Primary Regulations
(Federal Register, December 24, 1975) it is stated that operators per-
forming residual chlorine and turbidity analyses "....be certified,
approved, or at least minimally trained to perform the analytical tasks
before a State could accept their analytical determinations "
CH.TURB. 3.9.77 16-1
-------
II. RESIDUAL CHLORINE
Since residual chlorine analysis would be carried out in "field" conditions
or in the small laboratories of treatment plants, perhaps by unskilled
operators, it is necessary to keep the analytical method as simple as
possible. For a number of years, operators had utilized the orthotolidine
technique in a kit form to determine the chlorine residual. Recent
studies and regulatory guidelines have dictated against this test procedure.
The acceptable test procedure is now the DPD Test (13th Ed., Standard Methods
for the Examination of Water and Wastewater, pgs. 129-132), for which kits
are available from at least two companies and which meet requirements for
accuracy and reliability. These kits are capable of measuring both free
and combined chlorine of which only the free chlorine is measured to meet
compliance requirements. Kit procedures call for a premeasured single
powder or tablet reagent added to the test cell with the sample and a
resultant color development measures by comparison the standardized colors
within one minute. Standard Methods includes cautions regarding temperature
and pH control regarding this test parameter and this test procedure, the
DPD Test, is least effected by temperature and the pH is adjusted by the
added reagents. The only interfering substance, oxidized manganese, can
be determined in a preliminary step and compensated for in the final test
value.
III. TURBIDITY
Turbidity has long been used in the water supply industry for indicating proper
operational techniques. Turbidity should be clearly understood to be an ex-
pression of the optical property of a sample which causes light to be scattered
and absorbed rather than transmitted in straight lines through the sample.
The standard method for the determination of turbidity has been based on the
Jackson candle turbidimeter. However, the lowest turbidity value which can
be measured directly on the Jackson turbidimeter is 25 units which is well
above the monitoring level. Because of these low level requirements, the
nephelometric method was chosen and procedures are given in Standard Methods
(13th Ed., 1971).
IV. NEPHELOMETRIC MEASUREMENTS FOR COMPLIANCE MONITORING
The subjectivity and apparatus deficiencies involved in visual methods of
measuring turbidity make each unsuitable as a standard method.
Since turbidity is an expression of the optical property of scattering or
absorbing light, it was natural that optical instruments with photometers
would be developed for this measurement.
(3 6}
The type of equipment specified for compliance monitoring^ ' ; utilizes
nephelometry.
A. Basic Principle^ '
The intensity of light scattered by the sample is compared (under defined
conditions) with the intensity of light scattered by a standard reference
solution (formazin). The greater the intensity of scattered light, the
greater the turbidity. Readings are made and reported in NTUs (Nephelometric
Turbidity Units).
16-2
-------
B. Schematic
Turbidity Particles
Scatter Light
Sample Cell
(Top View)
Figure 2 NEPHELOMETER
(90° Scatter)
Light passes through a polarizing lens and on to the sample in a cell.
Suspended particles (turbidity) in the sample scatter the light.
Photocell(s) detect light scattered by the particles at a 90° angle to the
path of the incident light. This light energy is converted to an electric
signal for the meter to measure.
1. Direction of Entry of Incident Light to Cell
a. The lamp might be positioned as shown in the schematic so the
beam enters a sample horizontally.
b. Another instrument design has the light beam entering the sample
(in a flat-bottom cell) in a vertical direction with the photocell
positioned accordingly at a 90° angle to the path of incident light.
2. Number of Photocells
The schematic shows the photocell(s) at one 90° angle to the path of
the incident light. An instrument might utilize more than one photo-
cell position, with each final position being at a 90° angle to the
sample liquid.
3. Meter Systems
a. The meter might measure the signal from the scattered light in-
tensity only.
b. The meter might measure the signal from a ratio of the scattered
light versus light transmitted directly through the sample to a
photocell.
16-3
-------
4. Meter Scales and Calibration
a. The meter may already be calibrated in NTUs. In this case,
at least one standard is run in each instrument range to be
used in order to check the accuracy of the calibration scales.
b. If a pre-calibrated scale is not supplied, a calibration curve
is prepared for each range of the instrument by using appropriate
dilutions of the standard turbidity suspension.
C. EPA Specifications for Instrument Design^ '
Even when the same suspension is used for calibration of different
nephelometers, differences in physical design of the turbidimeters will
cause differences in measured values for the turbidity of the same sample.
To minimize such differences, the following design variables have been
specified by the U. S. Environmental Protection Agency.
1. Defined Specifications
a. Light Source
Tungsten lamp operated at not less than 85% of rated voltage
and at not more than rated voltage.
b. Distance Traveled by Light
The total of the distance traversed by the incident light plus
scattered light within the sample tube should not exceed 10 cm.
c. Angle of Light Acceptance of the Detector
Detector centered at 90° to the incident light path and not to
exceed + 30° from 90°.
(Ninety degree scatter is specified because the amount of scatter
varies with size of particles at different scatter angles).
d. Applicable Range
The maximum turbidity to be measured is 40 units. Several ranges
will be necessary to obtain adequate coverage. Use dilution for
samples if their turbidity exceeds 40 units.
2. Other EPA Design Specifications
a. Stray Light
Minimal stray light should reach the photocell(s) in the absence
of turbidity.
16-4
-------
3.
Some causes of stray light reaching the photocell(s) are:
1) Scratches or imperfections in glass cell windows.
2) Dirt, film or condensation on the glass.
3) Light leakages in the instrument system.
A schematic of these causes is shown in Figure 3.
Light Leakage from
Transmitted Light
Light Scatter by glass tube
(Top View)
Figure 3 NEPHELOMETER
SOURCES OF STRAY LIGHT
Stray light error can be as much as 0.5 NTU. Remedies are
close inspection of sample cells for imperfections and dirt,
and good design which can minimize the effect of stray light
by controlling the angle at which it reaches the sample.
b. Drift
The turbidimeter should be free from significant drift after a
short warm-up period. This is imperative if the analyst is
relying on a manufacturer's solid scattering standard for setting
overall instrument sensitivity for all ranges.
c. Sensitivity
In waters having turbidities less than one unit, the instrument
should detect turbidity differences of 0.02 unit or less.
Several ranges will be necessary to obtain sufficient sensitivity
for low turbidities.
Examples of instruments meeting the specifications listed in 1 and 2
above include:
a. Hach Turbidimeter Model 2100 and 2100A.
b. Hydroflow Instruments DRT 100, 200, and 1000.
16-5
-------
4. Other turbidimeters meeting the listed specifications are also
acceptable.
D. Sources of Error
1. Sample Cells
a. Discard scratched or etched cells.
b. Do not touch cells where light strikes them in instrument.
I Q\
c. Keep cells scrupulously clean, inside and out.v ;
1) Use detergent solution.
2) Organic solvents may also be used.
3) Use deionized water rinses.
4) Rinse and dry with alcohol or acetone.
2. Standardizing Suspensions^ '
a. Use turbidity - free water for preparations. Filter distilled
water through a 0.45ym pore size membrane filter if such filtered
water shows a lower turbidity than the distilled water.
b. Prepare a new stock suspension of Formazin each month.
c. Prepare a new standard suspension and dilutions of Formazin
each week.
3. Sample Interferences
a. Positive
1) Finely divided air bubbles
b. Negative
1) Floating debris
2) Coarse sediments (settle)
3) Colored dissolved substances
(absorb light)
16-6
-------
E. Reporting Results^ '
NTU _ RECORD TO NEAREST
0.0-1.0 0.05
1-10 0.1
10-40 1
40-100 5
100-400 10
400-1000 50
>1000 100
F. Precision and Accuracy^ '
1. In a single laboratory (EMSL), using surface water samples at
levels of 26, 41, 75 and 180 NTU, the standard diviations were
+0.60, +0.94, +1.2 and +4.7 units, respectively.
2. Accuracy data is not available at this time.
V. STANDARD SUSPENSIONS AND RELATED
One of the critical problems in measuring turbidity has been to find a
material which can be made into a reproducible suspension with uniform sized
particles. Various materials have been used.
A. Natural Materials
1. Diatomaceous earth
2. Fuller's earth
3. Kaolin
4. Naturally turbid waters.
Such suspensions are not suitable as reproducible standards because
there is no way to control the size of the suspended particles.
B. Other materials
1 . Ground glass
2. Microorganisms
3. Barium Sulfate
4. Lates spheres
Suspensions of these also proved inadequate.
16-7
-------
C. Formazin
1. A polymer formed by react .g hydrazine sulfate and hexamethylenete-
tramine sulfate.
2. It is more reproducible than previously used standards. Accuracy
of + one percent for replicate solutions has been reported.
3. In 1958, the Association of Analytical Chemists initiated a standard-
ized system of turbidity measurements for the brewing industry by:
a. Defining a standard formula for making stock Formazin solutions
and
b, Designating a unit of measurement based on Formazin, i.e., the
Formazin Turbidity Unit (FTU).
4. During the 1960's Formazin was increasingly used for water quality
turbidity testing. It is the currently recognized standard for
compliance turbidity measurements.
D. Units
1. At first results were translated into Jackson Turbidity Units (JTU).
However, the JTU was derived from a visual measurement using con-
centrations (nig/liter) of silica suspensions prepared by Jackson.
They have no direct relationship to the intensity of light scattered
at 90 degrees in a nephelometer.
2. For a few years, results of nephelometric measurements using specified
Formazin standards were reported directly as Turbidity Units (TUs).
3. Currently, the unit used is named according to the instrument used for
measuring turbidity. Specified Formazin standards are used to calibrate
the instrument and results are reported as Nephelometric Turbidity
Units (NTUs).
VI. SUMMARY
The importance of residual chlorine determination can be seen in its possible
effect on the health of the consumers. The Criteria and Procedures for
Laboratory Certification suggests that some form of quality assurance should
be instituted on a state level to assure valid data for both the chlorine and
turbidity measurements. The comments on the public responses to the proposed
Interim Primary Regulations also suggests some form of quality assurance on
the state level to be instituted. Consequently, the Regional Certification team
should point out to the principal laboratories the importance of some kind of
effort being instituted. States might wish to offer some kind of formal
training effort as part of the approval mechanism for the operators doing
the chlorine and/or turbidity measurements.
16-8
------- |