# How to Perform the Most Probable Number (MPN) Method The most probable number (MPN) method is a statistical culture technique that estimates the concentration of viable microorganisms in a sample from the pattern of positive and negative replicate cultures. It is used in the MPN method in microbiology when direct plate counts are unreliable, when the target organism is present at low concentration, when the sample matrix is particulate or inhibitory, or when a liquid sample must be tested in a format that can handle a large volume. A trained analyst can set up a small three-tube MPN series in about 45 to 90 minutes, but elapsed time is usually 24 hours to 7 days because incubation and confirmation drive the schedule. This guide gives the bench-level procedure, the reasons behind each critical step, the controls that make the result defensible, and the mistakes that most often invalidate an MPN run. The MPN method does not count colonies on an agar surface. It scores growth or a biochemical signal in liquid cultures, then converts the score pattern into an estimated concentration with a confidence interval. That distinction matters for anyone learning microbiology. An MPN result is an estimate, not an exact count. The estimate is only as good as the dilution scheme, the selectivity of the medium, the incubation conditions, and the confirmation steps. ## The MPN Method in Microbiology: Principle and Scope The MPN method assumes that viable cells are randomly distributed in a well-mixed sample. When a measured aliquot is transferred to a culture tube, the number of viable cells in that aliquot follows a [Poisson distribution](/blog/research-skills/poisson-distribution-formula-and-examples). A tube is scored positive when at least one viable cell initiates growth and produces the expected signal. The method uses several dilutions, with multiple replicate tubes at each dilution, to bracket the transition from mostly positive tubes to mostly negative tubes. The dilution series produces a pattern such as 3 positive tubes, 2 positive tubes, and 1 positive tube. That pattern is compared with an MPN table or entered into an MPN calculator that uses the exact inoculum volumes and the number of replicates. Three assumptions support the calculation. First, the sample must be mixed well enough that each aliquot has the same expected cell burden. Second, the culture medium and incubation conditions must allow one viable target cell to grow and be detected. Third, the tubes must remain independent, with no cross-contamination and no inhibition that differs between tubes. When clumps, biofilms, or particulate food create uneven distribution, the Poisson model fits poorly and the MPN estimate becomes less reliable. The method is therefore strongest for liquids and for homogenates that can be mixed reproducibly. The positive signal can be turbidity, gas in a Durham tube, acid production with a pH indicator, a color change, fluorescence, or a PCR product. A Durham tube is a small inverted tube placed inside a broth tube to trap gas produced during fermentation. Turbidity is visible growth. A presumptive positive is a tube that shows the expected signal before confirmation. A confirmed positive is a tube whose growth or identity has been verified with a selective medium, a biochemical test, an immunologic test, or a nucleic acid test. The MPN method is widely used for water, food, beverages, soil, and environmental samples. Standard water methods use MPN formats for coliforms and Escherichia coli, including defined substrate tests and ISO 9308-2, which detects beta-D-glucuronidase activity with the fluorogenic substrate 4-methylumbelliferyl-beta-D-glucuronide [1]. Food methods use MPN for Salmonella, Listeria, E. coli, Vibrio, and other targets. Environmental and plant pathology methods use MPN-PCR for Ralstonia solanacearum in soil [2]. Industrial and sterilization studies use MPN for quantification when bioburden is low [3]. The common thread is a low or uneven target burden that makes a direct plate count insensitive or impractical. ## When to Use MPN and When Not To Use the MPN method when the expected concentration is low, when the sample is a liquid that can be inoculated directly, when the matrix contains particles that clog filters or obscure colonies, when a large volume must be tested, or when a selective broth is more reliable than a selective agar. Use it when the result can be reported as an estimate with a confidence interval, and when the validated method for the matrix specifies MPN. Do not use MPN when a precise count is required, when the sample contains a very high target concentration that would require many dilutions, or when clumping cannot be controlled. Do not treat an MPN result as interchangeable with a colony count from a pour plate or a membrane filter without validation. Comparative studies have examined MPN against direct plating methods for E. coli in raw meats and foods of animal origin, and method agreement depends on the matrix and the target [14,16]. A laboratory that switches methods needs a matrix-specific comparison, not an assumption that the numbers mean the same thing. The MPN method is also a poor choice when the target is present as spores or stressed cells that need a specific recovery step. The medium and incubation must be validated for the physiological state of the target. A presumptive positive that depends on a single biochemical reaction can overestimate the target if background organisms share that reaction. Confirmation is not optional when the result will guide a release decision, a regulatory report, or a clinical or veterinary interpretation. ## Materials and Reagents | Item | Working concentration or specification | Purpose | |--|--|--| | Sample diluent | 0.1% peptone water, phosphate-buffered saline at pH 7.2, or 0.85% sodium chloride | Suspends cells and preserves viability during dilution | | Selective or enrichment broth | Lauryl tryptose broth, brilliant green lactose bile broth at 2%, EC broth, tryptic soy broth, or LB-0 without sodium chloride for Vibrio work [4] | Supports target growth and suppresses background | | Durham tube | Small inverted gas trap inside the broth tube | Detects gas production | | Chromogenic or fluorogenic medium | 4-methylumbelliferyl-beta-D-glucuronide for beta-D-glucuronidase, read under 365 nm ultraviolet light [1] | Detects E. coli or coliform activity | | Quanti-Tray or defined substrate tray | Sealed multi-well tray with a self-contained medium [5] | MPN format for water and liquid samples | | PCR reagents | Validated primers, probes, master mix, and extraction reagents for the target [3,5,8] | Confirms positive tubes or enumerates by MPN-PCR | | Pipettes and sterile tips | Calibrated 1 mL, 0.1 mL, and 10 mL pipettes | Delivers accurate inoculum volumes | | Incubator | Set to the validated temperature for the target, for example 35 plus or minus 2 °C for coliforms | Controls growth and biochemical expression | | Vortex mixer | Bench vortex with a foam-free mixing setting | Resuspends cells before each dilution | | MPN table or calculator | Matched to tube count, replicate number, and inoculum volumes | Converts the positive pattern to an MPN estimate | | Ultraviolet lamp | 365 nm wavelength [1] | Reads fluorescence in MUG-based tests | | Autoclave and biohazard waste containers | Validated steam sterilization cycle | Inactivates waste and contaminated media | | pH meter and calibration standards | Calibrated to the medium specification | Confirms broth pH before use | Prepare media exactly according to the validated formulation. Do not adjust a selective broth with acid or base unless the formulation allows it. A pH shift can change selectivity and create false positives or false negatives. Bring frozen or refrigerated media to the specified temperature before inoculation, and inspect each tube for turbidity, precipitate, or a displaced Durham tube. Discard tubes with cracks, leaks, or gas bubbles trapped at the bottom before inoculation. ## Safety Notes Treat every sample and every inoculated tube as a biohazard. Wear gloves, a laboratory coat, and eye protection. Perform dilution and inoculation work in a biosafety cabinet when the sample may contain pathogens or when aerosols are likely. Autoclave contaminated waste at 121 °C for 15 minutes or use the institutional waste policy. Use ultraviolet goggles when reading MUG fluorescence at 365 nm [1]. Keep the ultraviolet lamp shielded and do not look directly at the source. Work with PCR reagents in a separate clean area, use dedicated pipettes, and include no-template controls to detect contamination [5,8]. Disinfect bench surfaces before and after work with a validated disinfectant. ## Step-by-Step MPN Procedure 1. Define the target, the matrix, the reporting unit, and the acceptance criteria before opening the sample. The reporting unit may be MPN per gram, MPN per milliliter, or MPN per 100 milliliters. A clear unit prevents a later error in conversion. Record the method reference, the incubation temperature, the incubation time, and the confirmation scheme. 2. Label every tube with the sample code, the dilution, the replicate number, the date, and the analyst initials. A three-tube series at four dilutions requires 12 tubes before controls. Label the tube caps as well as the sides because condensation can erase side labels. 3. Prepare the primary sample suspension. For a solid food, weigh 25 g into 225 mL of diluent to make a 1:10 suspension. For a liquid, use the validated volume, often 100 mL for water or 10 mL for a beverage. Homogenize for the validated time, then allow coarse particles to settle for a short period if the method permits. Do not centrifuge unless the validated method calls for it because centrifugation can change the viable count. 4. Prepare a decimal dilution series. Transfer 1 mL of the primary suspension into 9 mL of diluent to make the next dilution. Vortex for 5 to 10 seconds, change the pipette tip, and repeat. A separate tip for each dilution and each tube is mandatory. Carryover on a pipette tip can create a positive at a dilution that should be negative and distort the entire MPN pattern. 5. Select the dilution range so that the highest dilution has no positive tubes and the lowest dilution has at least one positive tube. A series that is all positive or all negative does not bracket the endpoint. It gives a censored result rather than a useful MPN estimate. For a three-tube method, prepare at least three dilutions. For a five-tube method, the same principle applies with five replicates at each dilution. 6. Inoculate the replicate tubes. For a three-tube MPN with 1 mL inocula, add 1 mL of each dilution to each of three tubes containing 9 mL of broth. For a 10 mL inoculum, use double-strength broth to keep the final medium concentration correct. For a 0.1 mL inoculum, use a calibrated pipette and deliver the volume directly into the broth. Touch the pipette tip to the inside wall just above the liquid and allow the volume to drain. Do not blow out the contents because that creates aerosols. 7. Place a Durham tube in each gas-detection tube before sterilization or add a sterile Durham tube as directed by the validated method. The open end must point down and the tube must be free of trapped air. A bubble in the Durham tube before incubation makes a gas-positive reading impossible to interpret. 8. Include controls in every run. A negative medium control contains sterile diluent instead of sample. A positive control contains a known target strain at a low, validated inoculum. A matrix control contains the sample matrix without the target when the matrix is known. A process control tests the diluent and the handling steps. For PCR confirmation, include a no-template control, a positive DNA control, and an inhibition control for each sample type [5,8]. 9. Incubate the tubes at the validated temperature and time. For coliform MPN, a common incubation is 35 plus or minus 2 °C for 24 to 48 hours. For E. coli confirmation, a parallel set may be incubated at 44.5 plus or minus 0.2 °C. For Salmonella, the pre-enrichment and selective steps follow the validated method. Do not stack tubes so tightly that the incubator temperature becomes uneven. Record the incubator temperature at the start and end of the run. 10. Read the presumptive result before shaking or moving the tubes. Record gas in the Durham tube, turbidity, color change, precipitate, and fluorescence under 365 nm [1]. A positive tube must show the expected signal for the target. A tube with turbidity but no gas may be positive for a non-gas-producing organism, so the reading scheme must match the target and the medium. 11. Confirm every presumptive positive. Transfer a loopful from each positive tube to the confirmation medium or test specified by the method. For E. coli, this may include a brilliant green lactose bile broth tube, an EC broth tube with MUG, or a chromogenic medium. For Salmonella, confirmation may include xylose lysine deoxycholate agar, triple sugar iron agar, and lysine decarboxylase testing. For Vibrio, TCBS agar can fail to distinguish V. cholerae, V. vulnificus, and V. alginolyticus, so molecular confirmation with a validated PCR is more specific [4]. 12. For MPN-PCR, extract DNA from each positive broth and run the validated PCR. Score the tube as positive only when the PCR target is detected and all controls pass. An MPN-PCR method can enumerate Campylobacter from environmental water [6], Salmonella Typhimurium with a shortened incubation [7], Ralstonia solanacearum in soil [2], histamine-producing bacteria in fish [8], and Listeria monocytogenes in fermented sausages [9]. The nucleic acid test does not remove the need for a culture step unless the method is validated for direct detection. 13. Record the number of confirmed positives at each dilution. Use the confirmed positives, not the presumptive positives, for the final MPN calculation. If a tube is presumptive positive but fails confirmation, score it as negative. A positive at a higher dilution with a negative at a lower dilution is a valid skip only when the method and the statistics allow it. In most bench settings, a skip indicates a pipetting error, a tube mix-up, or contamination. Repeat the run if the pattern will affect a decision. 14. Calculate the MPN from the confirmed pattern. Use the MPN table or calculator that matches the number of tubes, the number of dilutions, and the exact inoculum volumes. Do not use a table built for 1 mL, 0.1 mL, and 0.01 mL if your volumes are 10 mL, 1 mL, and 0.1 mL. The same positive pattern gives a different MPN when the volumes differ. Report the 95% confidence interval with the estimate. 15. Report the result with the method, the dilution series, the incubation conditions, the confirmation test, and the detection limit. If no tubes are positive, report the result as less than the detection limit for the scheme. If all tubes are positive, report the result as greater than the upper limit and repeat with higher dilutions. Do not report a single number without the supporting pattern and the confidence interval. ## Controls and Quality Assurance Quality assurance for an MPN run rests on three layers. The first layer is sterility. Uninoculated medium controls must remain clear. The diluent control must remain negative. Sterile tips, tubes, and diluents must be handled as sterile items. The second layer is growth promotion. A positive control with a known target strain must grow and produce the expected signal at the validated incubation time. A negative control with a non-target strain can show that the medium suppresses background, but it does not replace growth promotion with the target. The third layer is confirmation. Every positive tube that contributes to the MPN pattern should be confirmed by the method's reference test. Method validation is matrix specific. A study that compares an MPN method with an established method can show equivalence for one water type and not another. The ENUMERA Coli100 chromogenic method was evaluated against ISO 9308-2 for E. coli and total coliforms in 100 mL water samples across drinking water, surface water, wastewater, reclaimed water, and seawater, and the two methods gave statistically equivalent results across those matrices [1]. That is a validation statement for those matrices, not a universal claim for every sample. The Pseudalert and Quanti-Tray MPN test was compared with the ISO 16266 membrane filtration method for Pseudomonas aeruginosa in cooling tower water [5]. Such comparisons matter because an MPN test can be equivalent in one matrix and different in another. Count the confirmed positives twice and have a second analyst read a subset of tubes when the result is critical. Record the raw readings before confirmation. Store the records with the run so that a later review can reconstruct the pattern. If the laboratory uses an automated MPN system for heterotrophic bacteria in water, verify calibration and compare the automated reading with a manual reference at defined intervals [10]. ## Expected Results and How to Read Them A valid MPN run produces a dilution series with a transition from mostly positive tubes to mostly negative tubes. The positive pattern is written from the lowest dilution to the highest dilution. For example, 3 positive tubes at the first dilution, 2 at the second, and 1 at the third is written as 3-2-1. That code is looked up in the MPN table for the exact inoculum volumes and the exact number of tubes. The table returns an MPN estimate and a 95% confidence interval. | Observation | Interpretation | Action | |--|--|--| | Gas in the Durham tube and turbidity in the selective broth | Presumptive positive for a gas-producing target | Confirm with the method-specific test | | Turbidity without gas | Growth of a non-gas producer or a target that does not produce gas | Do not score as positive unless the method defines turbidity as the signal | | Fluorescence under 365 nm in a MUG tube | Presumptive positive for beta-D-glucuronidase activity [1] | Confirm the target and rule out background fluorescence | | All tubes positive at every dilution | The target concentration is above the upper limit of the series | Repeat with higher dilutions | | All tubes negative | The target is below the detection limit or the method failed | Check controls and repeat if the result is unexpected | | Positive at a higher dilution and negative at a lower dilution | Skip pattern, possible pipetting error or contamination | Review records and repeat the run | | Growth in the negative control | Medium, diluent, or handling contamination | Discard the run and investigate | The confidence interval is part of the result. A low MPN based on a few positive tubes has a wide interval, often spanning an order of magnitude. That width is not a laboratory error. It reflects the statistical nature of the method. For a result near a regulatory limit, the confidence interval can determine whether the sample passes or fails. Report the interval even when the point estimate is the only number that appears in a summary table. ## Troubleshooting | Symptom | Likely cause | Fix | |--|--|--| | All tubes positive at every dilution | Dilution series too low or sample load too high | Repeat with a higher dilution series and recalibrate pipettes | | All tubes negative despite an expected positive | Medium too selective, incubation too short, or target stressed | Check growth promotion, incubation temperature, and recovery medium | | Gas in the negative control | Air trapped in the Durham tube or contaminated medium | Degas the Durham tube, autoclave properly, and prepare fresh medium | | Inconsistent pattern with a skip | Pipette carryover, tube mix-up, or contamination | Use a new tip for every tube, relabel, and repeat | | Turbid tubes with no confirmable target | Background growth on a non-selective or over-incubated medium | Use a more selective medium and shorten incubation to the validated time | | PCR positive in the no-template control | PCR reagent or workspace contamination | Discard reagents, clean the workspace, and repeat with fresh aliquots | | PCR negative in a visibly positive tube | Inhibition, DNA loss, or a target outside the primer range | Include an inhibition control and use the validated extraction method | | Fluorescence difficult to read | Ultraviolet lamp aging or tube background | Check the lamp at 365 nm, use a dark reading box, and include a positive control [1] | | Clumped sample or uneven replicate counts | Poor homogenization or particulate matrix | Increase homogenization time, use a validated dispersant, and test replicate homogenates | | Incubator temperature drift | Overloaded incubator or failed thermostat | Record temperature, space tubes evenly, and service the incubator | ## Variations The three-tube MPN is the most common format for food and [clinical microbiology](/knowledge/diagnostics/microbiology/clinical-microbiology-from-specimen-collection-to-pathogen-identification) teaching. The five-tube MPN gives a narrower confidence interval and is used when the decision has a tighter tolerance. The ten-tube format is used in some water methods. The trade-off is cost, bench space, and hands-on time. More tubes improve precision but do not correct a non-random sample. The miniaturized MPN uses small volumes in microplates or microwells. A milk method used one dilution and three sets of five aliquots at 100 microliters, 10 microliters, and 1 microliter in 15 microwells and showed good correlation with pour-plate methods, with an R squared above 0.9 [11]. Miniaturization reduces medium use and bench space but increases the risk of evaporation, cross-well contamination, and pipetting error. Use a calibrated multichannel pipette and a plate seal or humidity chamber. The defined substrate MPN uses a tray with multiple wells and a chromogenic or fluorogenic medium. Quanti-Tray and Pseudalert are examples for water testing [5]. The sample is added to the medium, the tray is sealed, and the number of positive wells is converted to an MPN. The sealed format reduces handling and contamination. It also limits the analyst's ability to confirm individual wells, so the method must be validated for the target and the matrix. The MPN-PCR variation combines culture in a selective broth with a molecular confirmation. It improves specificity and speed when the target is slow-growing or when biochemical confirmation is ambiguous. Vibrio cholerae and Vibrio vulnificus were enumerated with an advanced MPN-PCR method that used LB-0 medium and a one-time PCR based on differences in salt requirement [4]. Campylobacter MPN-PCR from estuarine and wastewater samples showed superior intra-laboratory performance compared with a culture-based standard and an overall diagnostic accuracy of about 94 percent [6]. Salmonella Typhimurium was quantified with MPN combined with qPCR and a shortened incubation [7]. A miniaturized MPN based on ISO 6579:2002 was validated for Salmonella from poultry matrices [12]. Ralstonia solanacearum was quantified in soil by MPN-PCR [2]. Histamine-producing bacteria in fish were enumerated by MPN real-time PCR [8]. Listeria monocytogenes in fermented sausages was quantified by MPN-PCR [9]. Automated MPN systems measure growth or signal in a multi-well format and calculate the MPN from the instrument reading. An automated method for heterotrophic bacteria in water was based on the MPN method [10]. Automation reduces reading time but does not remove the need for calibration, sterility checks, and a manual reference method. The sterilization assessment of decellularized cardiovascular scaffolds used MPN quantification as part of a protocol that included controlled contamination and sterility testing [3]. That use shows how MPN can quantify a low bioburden when a direct count is not sensitive enough. ## Storage and Stability Notes Store dehydrated media in a sealed container at room temperature and protect it from moisture. Prepare broth on the day of use when possible. Autoclaved broth can be stored at 4 °C for a limited period defined by the laboratory's media policy. Do not use broth with a pH shift, precipitate, or turbidity. Protect MUG-containing media from light because the fluorogenic substrate can degrade [1]. Bring the medium to the incubation temperature before use. Do not repeatedly warm and cool a broth tube. Store PCR reagents at the temperature specified by the manufacturer, usually -20 °C in a dedicated freezer without frost-free cycling. Keep primers and probes in aliquots to avoid freeze-thaw cycles. Store extracted DNA at 4 °C for same-day testing or at -20 °C for longer periods. Do not store a presumptive positive broth for days before confirmation unless the validated method allows it. Overgrowth can change the confirmation result. Sample storage is part of the method. Process water and liquid samples within the holding time specified by the method. Refrigerate food and environmental samples at 4 °C until analysis, but do not freeze them unless the validated method states that freezing is acceptable. Freezing can injure cells and lower the MPN. Record the collection time, the receipt time, and the analysis time. ## Common Mistakes and Limitations The most common mistake is failing to bracket the endpoint. An all-positive or all-negative series gives a censored result, not a useful MPN. A second common mistake is using the wrong MPN table. The same code of 3-2-1 means different concentrations when the inoculum volumes are different. Always match the table to the tube count, the number of dilutions, and the exact volumes. Pipette carryover is another frequent cause of a false positive. A visible bubble in the pipette tip, a touch to the broth, or a reused tip can transfer cells from a low dilution to a high dilution. Use a new sterile tip for every dilution and every tube. Deliver the volume slowly and do not blow out the tip. If the same pipette is used for multiple tubes, change the tip between tubes even when the dilution is the same. Overreading turbidity is common. Turbidity can come from the sample itself, from dead cells, or from a non-target organism. A turbid tube is not automatically a positive MPN tube. The reading scheme must define the signal. Gas in a Durham tube, a color change, fluorescence, or a confirmed PCR product is stronger evidence than turbidity alone. Confirm presumptive positives with the method-specific test. Ignoring the confidence interval is a reporting error. An MPN is an estimate with a wide interval at low counts. The method also carries a known bias, and the direction and size of that bias depend on the dilution scheme and the true concentration [13]. Do not present an MPN as if it were a colony count with no uncertainty. Do not average MPN values from replicate runs without using the proper statistical method. The MPN method assumes a random distribution. Clumped cells, biofilm fragments, and particulate food violate that assumption. A sample that contains clumps can produce a lower or higher MPN than the true concentration depending on how the clumps break apart. Homogenize according to the validated method and do not assume that a short shake is enough for a sticky matrix. Selectivity is another limitation. A medium that suppresses background in one matrix may fail in another. A presumptive positive that depends on a single enzyme can be produced by a non-target organism. The MPN-PCR methods cited here were developed because conventional selective media can be ambiguous. TCBS agar can fail to distinguish Vibrio cholerae, Vibrio vulnificus, and Vibrio alginolyticus, which is why a one-time PCR step improved specificity [4]. The method measures culturable cells under the chosen conditions. Cells that are viable but not culturable, stressed, or injured may not grow. A PCR step can detect DNA from cells that did not grow, so an MPN-PCR result must be interpreted with the culture step in mind. A positive PCR from a broth does not prove that the cell was viable at the time of sampling unless the method includes a viability step. Direct plating comparisons are useful but not interchangeable. The MPN procedure was compared with the Anderson-Baird-Parker direct plating method for E. coli biotype I in raw meats in an international study [14]. Another study evaluated MPN, Anderson-Baird-Parker, Petrifilm E. coli, and Fluorocult ECD for E. coli in foods of animal origin [15]. Those comparisons show that method choice affects the result. A laboratory that changes from MPN to a direct plating method needs a matrix-specific validation. Finally, do not use MPN for a high-count sample without a plan for the upper limit. If all tubes are positive, the result is greater than the upper limit. Repeat with a higher dilution series. If the sample is expensive or difficult to resample, prepare extra dilutions at the start. The cost of a repeat run is usually higher than the cost of a few extra tubes. ## Frequently Asked Questions ### What does the MPN method actually measure? The MPN method estimates the concentration of viable microorganisms that can grow and produce the defined signal under the chosen culture conditions. It does not count every cell in the sample. ### How many tubes and dilutions should I use for a basic MPN? Use at least three dilutions with three replicate tubes each for a basic three-tube MPN. Use five replicates when you need a narrower confidence interval. The series must bracket the endpoint with at least one positive at the lowest dilution and no positives at the highest. ### Why is turbidity alone a weak positive signal? Turbidity can come from the sample, from dead cells, or from non-target organisms. A valid MPN reading uses the defined signal for the target, such as gas, a color change, fluorescence, or a confirmed PCR product. ### Can I use the MPN method for solid foods? Yes. Prepare a validated primary homogenate, usually 25 g in 225 mL of diluent, then make a decimal dilution series. Particulate foods need thorough homogenization because clumping violates the random-distribution assumption. ### How do I report an MPN result when no tubes are positive? Report the result as less than the detection limit for the dilution scheme. State the detection limit, the method, and the confirmation test. Do not report zero as if it were an exact count. ### What is MPN-PCR? MPN-PCR combines the MPN culture series with a PCR test on the positive broths. It improves specificity and can shorten confirmation time. The PCR result must be read with the culture result and with the appropriate controls. ### Why is the 95% confidence interval so wide? The interval reflects the statistical uncertainty of an estimate based on a small number of replicate tubes. Low counts and few positives produce wider intervals. More replicates and a well-placed dilution series narrow the interval. ### How do I prevent contamination in a three-tube MPN series? Use a new sterile tip for every dilution and every tube, keep the workspace clean, include negative controls, and do not touch the broth with the pipette. Label every tube and read the controls before reading the sample tubes. ## Sources 1. [A chromogenic medium for the enumeration of Escherichia coli and total coliforms in 100 mL water samples: validation of an MPN-based method according to ISO 9308-2.](https://pubmed.ncbi.nlm.nih.gov/42612281/) 2. [Sensitive quantitative detection of Ralstonia solanacearum in soil by the most probable number-polymerase chain reaction (MPN-PCR) method](https://pubmed.ncbi.nlm.nih.gov/24584461/) 3. [A sterilization method for decellularized xenogeneic cardiovascular scaffolds.](https://pubmed.ncbi.nlm.nih.gov/29183849/) 4. [Simultaneous isolation and enumeration of virulent Vibrio cholerae and Vibrio vulnificus using an advanced MPN-PCR method](https://pubmed.ncbi.nlm.nih.gov/34870749/) 5. [Comparison of the Pseudalert™/Quanti-Tray® MPN test for the enumeration of Pseudomonas aeruginosa in cooling tower water with the ISO 16266 membrane filtration culture-based method](https://pubmed.ncbi.nlm.nih.gov/31965680/) 6. [Environmental monitoring of waterborne Campylobacter: evaluation of the Australian standard and a hybrid extraction-free MPN-PCR method](https://pubmed.ncbi.nlm.nih.gov/25709604/) 7. [Development of a rapid method to quantify Salmonella Typhimurium using a combination of MPN with qPCR and a shortened time incubation.](https://pubmed.ncbi.nlm.nih.gov/28400022/) 8. [Quantification of total and specific gram-negative histamine-producing bacteria species in fish using an MPN real-time PCR method.](https://pubmed.ncbi.nlm.nih.gov/21839377/) 9. [Quantification of Listeria monocytogenes in fermented sausages by MPN-PCR method](https://pubmed.ncbi.nlm.nih.gov/15287877/) 10. [Automated measurement and quantification of heterotrophic bacteria in water samples based on the MPN method](https://pubmed.ncbi.nlm.nih.gov/20835882/) 11. [A simple low-effort method for estimating bacterial loads in milk using a most probable number (MPN) approach with small aliquots.](https://pubmed.ncbi.nlm.nih.gov/42097370/) 12. [A validated miniaturized MPN method, based on ISO 6579:2002, for the enumeration of Salmonella from poultry matrices](https://pubmed.ncbi.nlm.nih.gov/20059618/) 13. [Some remarks on the bias of the MPN method.](https://pubmed.ncbi.nlm.nih.gov/1888659/) 14. [ICMSF methods studies. XIII. An international comparative study of the MPN procedure and the Anderson-Baird-Parker direct plating method for the enumeration of Escherichia coli biotype I in raw meats.](https://pubmed.ncbi.nlm.nih.gov/396015/) 15. [Evaluation of the MPN, Anderson-Baird-Parker, Petrifilm E. coli and Fluorocult ECD method for enumeration of Escherichia coli in foods of animal origin.](https://pubmed.ncbi.nlm.nih.gov/1445766/)