United States Food Safety Office of Laboratory QA/QC Division

Department of and Inspection Public Health 950 College Station Road

Agriculture Service Science Athens, GA 30605

_______________________________________________________________________________________________

Laboratory Guidebook
Notice of Change

Chapter new, revised, or archived: MLG Appendix 2.03

Title: Most Probable Number Procedure and Tables

Effective Date: 01/28/08

Description and purpose of change(s):

The MPN procedure used by the FSIS Field Service Laboratories to enumerate
Listeria monocytogenes, Salmonella spp. and E. coli O157:H7 product sample isolates
was added.

A transcription error in Table 4 was corrected.

The methods described in this guidebook are for use by the FSIS laboratories. FSIS does not
specifically endorse any of the mentioned test products and acknowledges that equivalent
products may be available for laboratory use.
_________________________________________________________________________________
QD-F-Micro-0004.03 Issuing Authority: Laboratory Quality Assurance Division (LQAD)
Page 1 of 1 Effective: 5/29/07
 United States Department of Agriculture
Food Safety And Inspection Service, Office of Public Health Science

MLG Appendix 2.03 Page 1 of 8

Title: Most Probable Number Procedure and Tables

Revision: 03 Replaces: 02 Effective: 01/28/08

APP 2 Procedure

APP 2.1 Introduction

APP 2.1.1 For the convenience of analysts using the procedures in this Guidebook, tables of
Most Probable Numbers (MPN) are provided in this appendix.

APP 2.1.2 MPN is a procedure to estimate the population density of viable microorganisms in a
test sample. It is based upon the application of the theory of probability to the
numbers of observed positive growth responses to a standard dilution series of
sample inoculum placed into a set number of culture media tubes. Positive growth
response after incubation may be indicated by such observations as gas production in
fermentation tubes or visible turbidity in broth tubes, depending upon the type of
media employed. The sample should be diluted in such a manner that higher
dilutions of the sample will result in fewer positive culture tubes in the series. The
number of sample dilutions to be prepared is generally based on the expected
population contained within the sample. If particularly high microbial populations
are expected, the sample must be diluted to a range where the MPN can be obtained.
Most reliable results occur when all tubes at the lower dilution are positive and all
tubes at the higher dilution are negative. Generally tenfold serial dilutions are used
in either a 3, 5 or 10 tube MPN series. When a higher number of tubes are
inoculated in the series, the confidence limits of the MPN are narrowed. For
particularly high microbial populations, the values obtained by MPN are generally
not considered to be as precise as population numbers derived from direct plating
methods; however, it should be emphasized that MPN values are only estimates
while plate counts are direct counts of living organisms expressed in cfu/ml. MPN
values are, however, particularly useful when low concentrations of organisms
(<100/g) are encountered in such materials as milk, food, water and soil where
particulate matter of the matrix may interfere with obtaining accurate colony counts.

APP 2.1.3 In application of probability theory to the determination of MPN values, it should be
kept in mind that the following assumptions are generally considered to be accepted:
(a) the organisms are randomly and evenly distributed throughout the sample, (b) the
organisms exist as single entities, not as chains, pairs or clusters and they do not
repel one another, (c) the proper growth medium, temperature and incubation
conditions have been selected to allow even a single viable cell in an inoculum to
produce detectable growth and (d) the population does not contain viable, sub-
lethally injured organisms that are incapable of growth in the culture medium used.

Approved by Laboratory Quality Assurance Division (LQAD)

 United States Department of Agriculture
Food Safety And Inspection Service, Office of Public Health Science

MLG Appendix 2.03 Page 2 of 8

Title: Most Probable Number Procedure and Tables

Revision: 03 Replaces: 02 Effective: 01/28/08

APP 2.1.4 The following 4 tables present MPN values and corresponding 95% confidence
limits for a 3 tube test series using 4 different, commonly used sets of inoculum
quantities that should be particularly useful relative to performing the
microbiological analyses previously described in various chapters of this Guidebook.
These MPN tables may be considered to be abbreviated since not all possible
combinations of positive and negative tubes within a series are presented. Those
combinations that occur often enough to have statistical significance are included,
while those that are improbable have been omitted. If laboratory analyses produce
combinations that are not included in the tables, then one should repeat the test on
another portion of the original sample (assuming the microbiological integrity of the
sample has not been compromised) as a possible performance error or contamination
is indicated. If this is not possible and an MPN is imperative, then more complete
tables should be consulted from other reference sources or the MPN can be
calculated by equation (other reference sources) on the basis of the observed results.

APP 2.1.5 On occasions when more than three dilutions of a sample are used in a decimal
series of dilutions of a 3-tube MPN determination; the following guidelines should
be followed. Results from only three consecutive dilutions are used to determine
the MPN. If one or more dilutions have all tubes positive, select the highest
dilution (smallest sample quantity) with positive results in all tubes and the next
two higher dilutions, as shown in examples a and b below. When none of the
dilutions yield all tubes positive, select the three lowest dilutions for which the
middle dilution contains the positive result, as shown in example c below. If a
positive result occurs in a higher unselected dilution, add the number of positive
tubes in this dilution to the results of the highest dilution of the three selected, as
shown in example d below. When all dilutions tested yield all tubes positive,
select the three highest dilutions (example e below). For additional information on
MPN estimations, consult the APHA’s Compendium of Methods for the
Microbiological Examination of Foods (3rd edition, 1992, chapter 6).

APP 2.2 SAMPLE PREPARATION FOR MOST PROBABLE NUMBER (MPN)

A series of 3 sequential MPN tubes may be inoculated directly from the 1:10
sample-culture medium homogenate if a low level of organisms is expected.
Prepare the sample in the initial culture medium as directed for the analysis. Set up
3 empty tubes and 6 tubes that contain 9 ml of the appropriate culture medium.
Add 10 mL of the homogenate to each of 3 empty tubes, 1 mL of the homogenate
to each of 3 tubes of culture medium, and 0.1 mL to each of the last 3 tubes of
culture medium. This series of tubes represents 1.0, 0.1 and 0.01 g of sample. The
remainder of the sample homogenate should also be cultured.

Approved by Laboratory Quality Assurance Division (LQAD)

 United States Department of Agriculture
Food Safety And Inspection Service, Office of Public Health Science

MLG Appendix 2.03 Page 3 of 8

Title: Most Probable Number Procedure and Tables

Revision: 03 Replaces: 02 Effective: 01/28/08

Alternatively, prepare serial 10-fold dilutions of the sample in the appropriate

culture medium and follow the outline below. This is the preferred method if it is
anticipated that the level of organisms may exceed 10 growth units per gram.

Volume of Original Source of Quantity of Quantity of Number of

Sample Inoculum Inoculum Culture Tubes
Medium
10 mL Sample 10.0 mL 90 mL 3
1.0 mL Sample 1.0 mL 9.0 mL 3
0.1 mL Dilution Tube 10-1 1.0 mL 9.0 mL 3
0.01 mL Dilution Tube 10-2 1.0 mL 9.0 mL 3
0.001 mL Dilution Tube 10-3 1.0 mL 9.0 mL 3
0.0001 mL Dilution Tube 10-4 1.0 mL 9.0 mL 3

To increase sensitivity in samples expected to have a low level of the organism of

interest, the FSIS laboratories perform the following procedure to prepare most
product samples for pathogen MPN analysis.

Start with a 65 ± 2 g test portion (representative of the entire sample) taken from
the sample reserve. Add 585 ml enrichment broth and stomach for 2 minutes.

Set up a 3-tube 5-dilution MPN (i.e., 15 individual dilutions for each sample)
representing 10, 1, 0.1, 0.01, and 0.001 g of sample. Three containers will hold
100 ml homogenate each to represent the 10 g sample dilution. Three containers
will hold 10 ml homogenate each to represent the 1 g sample dilution. Three
containers will hold 1 ml homogenate plus 9 ml enrichment broth each to
represent the 0.1 g sample dilution. Three tubes will contain 0.1 ml homogenate
plus 9.9 ml enrichment broth each to represent the 0.01 g sample dilution; and add
0.1 homogenate to 9.9 ml enrichment, vortex, and add 1.0 ml from that dilution to
each of three tubes containing 9.0 ml enrichment for the 0.001 g sample dilution.

Analyze each dilution (and the remainder of the sample homogenate) per
appropriate MLG chapter. MPN dilutions and the remainder of the sample
homogenate that screen positive will be confirmed.

Approved by Laboratory Quality Assurance Division (LQAD)

 United States Department of Agriculture
Food Safety And Inspection Service, Office of Public Health Science

MLG Appendix 2.03 Page 4 of 8

Title: Most Probable Number Procedure and Tables

Revision: 03 Replaces: 02 Effective: 01/28/08

APP 2.3 Completion of the MPN

After all MPN dilutions have been completed, refer to the MLG for each pathogen
being analyzed to determine incubation times and temperatures and procedures
for completion of the analyses.

APP 2.4 Example calculations

Sample quantities (g or ml)1 Reported MPN

positive estimate/
Example 10 1 0.1 0.01 0.001 values g or ml
a 3/32 3/3 2/3 0/3 0/3 3-2-0 9.3
b 3/3 3/3 3/3 2/3 0/3 3-2-0 93.
c 0/3 0/3 1/3 0/3 0/3 0-1-0 0.31
d 3/3 3/3 2/3 1/3 1/3 3-2-2 21.
e 3/3 3/3 3/3 3/3 3/3 3-3-3 >1100

1
The analyst should make sure that ALL sample dilution factors (including
the preparation of any sample homogenate) are correctly applied in
calculating the actual sample quantities subjected to MPN analysis.
b2
Numerator No. positive tubes
___________ = ____________________

Denominator No. tubes inoculated

APP 2.5 Reference(s):

Blodgett, R. FDA Bacteriological Analytical Manual Online, Appendix 2, Most

Probable Number Determination from Serial Dilutions, 2006.

Swanson, K.M.J, R. L. Petran and J. H. Hanlin. Chapter 6, Culture Methods for

Enumeration of Microorganisms in “Compendium of Methods for the
Microbiological Examination of Foods” Fourth Edition, 2001. Edited by F. P.
Downes and K. Ito. Published by American Public Health Association.

Approved by Laboratory Quality Assurance Division (LQAD)

MLG Appendix 2.03 MPN Table 1. Page 5 of 8

Table 1. MPN Index and 95% Confidence Limits for Various

Combinations of Positive Tubes in a 3 Tube Dilution Series Using
Inoculum Quantities of 10, 1 and 0.1 g (ml).

Combination of MPN Index 95% Confidence Limits

Positives per g (ml) Lower Upper
0-0-0 <0.03 --- 0.095
0-0-1 0.030 0.0015 0.096
0-1-0 0.030 0.0015 0.11
0-1-1 0.061 0.012 0.18
0-2-0 0.062 0.012 0.18
0-3-0 0.094 0.036 0.38

1-0-0 0.036 0.0017 0.18

1-0-1 0.072 0.013 0.18
1-0-2 0.11 0.036 0.38
1-1-0 0.074 0.013 0.20
1-1-1 0.11 0.036 0.38
1-2-0 0.11 0.036 0.42
1-2-1 0.15 0.045 0.42
1-3-0 0.16 0.045 0.42

2-0-0 0.092 0.014 0.38

2-0-1 0.14 0.036 0.42
2-0-2 0.02 0.045 0.42
2-1-0 0.15 0.037 0.42
2-1-1 0.20 0.045 0.42
2-1-2 0.27 0.087 0.94
2-2-0 0.21 0.045 0.42
2-2-1 0.28 0.087 0.94
2-2-2 0.35 0.087 0.94
2-3-0 0.29 0.087 0.94
2-3-1 0.36 0.087 0.94

3-0-0 0.23 0.046 0.94

3-0-1 0.38 0.087 1.1
3-0-2 0.64 0.17 1.8
3-1-0 0.43 0.09 1.8
3-1-1 0.75 0.17 2.0
3-1-2 1.2 0.37 4.2
3-1-3 1.6 0.40 4.2
3-2-0 0.93 0.18 4.2
3-2-1 1.5 0.37 4.2
3-2-2 2.1 0.40 4.3
3-2-3 2.9 0.90 10.
3-3-0 2.4 0.42 10.
3-3-1 4.6 0.90 20.
3-3-2 11. 1.8 41.
3-3-3 >11. 4.2 ---

Approved by Laboratory Quality Assurance Division (LQAD)

MLG Appendix 2.03 MPN Table 2. Page 6 of 8

Table 2. MPN Index and 95% Confidence Limits for Various

Combinations of Positive Tubes in a 3 Tube Dilution Series Using
Inoculum Quantities of 1, 0.1 and 0.01 g (ml).

Combination of MPN Index 95% Confidence Limits

Positives per g (ml) Lower Upper
0-0-0 <0.3 ----- 0.95
0-0-1 0.30 0.015 0.96
0-1-0 0.30 0.015 1.1
0-1-1 0.61 0.12 1.8
0-2-0 0.62 0.12 1.8
0-3-0 0.94 0.36 3.8

1-0-0 0.36 0.017 1.8

1-0-1 0.72 0.13 1.8
1-0-2 1.1 0.36 3.8
1-1-0 0.74 0.13 2.0
1-1-1 1.1 0.36 3.8
1-2-0 1.1 0.36 4.2
1-2-1 1.5 0.45 4.2
1-3-0 1.6 0.45 4.2

2-0-0 0.92 0.14 3.8

2-0-1 1.4 0.36 4.2
2-0-2 2.0 0.45 4.2
2-1-0 1.5 0.37 4.2
2-1-1 2.0 0.45 4.2
2-1-2 2.7 0.87 9.4
2-2-0 2.1 0.45 4.2
2-2-1 2.8 0.87 9.4
2-2-2 3.5 0.87 9.4
2-3-0 2.9 0.87 9.4
2-3-1 3.6 0.87 9.4

3-0-0 2.3 0.46 9.4

3-0-1 3.8 0.87 11.
3-0-2 6.4 1.7 18.
3-1-0 4.3 0.90 18.
3-1-1 7.5 1.7 20.
3-1-2 12. 3.7 42.
3-1-3 16. 4.0 42.
3-2-0 9.3 1.8 42.
3-2-1 15. 3.7 42.
3-2-2 21. 4.0 43.
3-2-3 29. 9.0 100.
3-3-0 24. 4.2 100.
3-3-1 46. 9.0 200.
3-3-2 110. 18. 410.
3-3-3 >110. 42. ---

Approved by Laboratory Quality Assurance Division (LQAD)

MLG Appendix 2.03 MPN Table 3. Page 7 of 8

Table 3. MPN Index and 95% Confidence Limits for Various

Combinations of Positive Tubes in a 3 Tube Dilution Series Using
Inoculum Quantities of 0.1, 0.01 and 0.001 g (ml).
Combination of MPN Index 95% Confidence Limits
Positives per g (ml) Lower Upper
0-0-0 <3.0 --- 9.5
0-0-1 3.0 0.15 9.6
0-1-0 3.0 0.15 11.
0-1-1 6.1 1.2 18.
0-2-0 6.2 1.2 18.
0-3-0 9.4 3.6 38.

1-0-0 3.6 0.17 18.

1-0-1 7.2 1.3 18.
1-0-2 11. 3.6 38.
1-1-0 7.4 1.3 20.
1-1-1 11. 3.6 38.
1-2-0 11. 3.6 42.
1-2-1 15. 4.5 42.
1-3-0 16. 4.5 42.

2-0-0 9.2 1.4 38.

2-0-1 14. 3.6 42.
2-0-2 20. 4.5 42.
2-1-0 15. 3.7 42.
2-1-1 20. 4.5 42.
2-1-2 27. 8.7 94.
2-2-0 21. 4.5 42.
2-2-1 28. 8.7 94.
2-2-2 35. 8.7 94.
2-3-0 29. 8.7 94.
2-3-1 36. 8.7 94.

3-0-0 23. 4.6 94.

3-0-1 38. 8.7 110.
3-0-2 64. 17. 180.
3-1-0 43. 9.0 180.
3-1-1 75. 17. 200.
3-1-2 120. 37. 420.
3-1-3 160. 40. 420.
3-2-0 93. 18. 420.
3-2-1 150. 37. 420.
3-2-2 210. 40. 430.
3-2-3 290. 90. 1000.
3-3-0 240. 42. 1000.
3-3-1 460. 90. 2000.
3-3-2 1100. 180. 4100.
3-3-3 >1100. 420. -

Approved by Laboratory Quality Assurance Division (LQAD)

MLG Appendix 2.03 MPN Table 4. Page 8 of 8

Table 4. MPN Index and 95% Confidence Limits for Various

Combinations of Positive Tubes in a 3 Tube Dilution Series Using
Inoculum Quantities of .01, 0.001 and 0.0001 g (ml).

Combination of MPN Index 95% Confidence Limits

Positives per g (ml) Lower Upper
0-0-0 <30. --- 95.
0-0-1 30. 1.5 96.
0-1-0 30. 1.5 110.
0-1-1 61. 12. 180.
0-2-0 62. 12. 180.
0-3-0 94. 36. 380.

1-0-0 36. 1.7 180.

1-0-1 72. 13. 180.
1-0-2 110. 36. 380.
1-1-0 74. 13. 200.
1-1-1 110. 36. 380.
1-2-0 110. 36. 420.
1-2-1 150. 45. 420.
1-3-0 160. 45. 420.

2-0-0 92. 14. 380.

2-0-1 140. 36. 420.
2-0-2 200. 45. 420.
2-1-0 150. 37. 420.
2-1-1 200. 45. 420.
2-1-2 270. 87. 940.
2-2-0 210. 45. 420.
2-2-1 280. 87. 940.
2-2-2 350. 87. 940.
2-3-0 290. 87. 940.
2-3-1 360. 87. 940.

3-0-0 230. 46. 940.

3-0-1 380. 87. 1100.
3-0-2 640. 170. 1800.
3-1-0 430. 90. 1800.
3-1-1 750. 170. 2000.
3-1-2 1200. 370. 4200.
3-1-3 1600. 400. 4200.
3-2-0 930. 180. 4200.
3-2-1 1500. 370. 4200.
3-2-2 2100. 400. 4300.
3-2-3 2900. 900. 10000.
3-3-0 2400. 420. 10000.
3-3-1 4600. 900. 20000.
3-3-2 11000. 1800. 41000.
3-3-3 >11000. 4200. -

Approved by Laboratory Quality Assurance Division (LQAD)