Serial Dilution and Plate Count: A Standard Protocol for Bacterial Enumeration

Serial dilution followed by plate counting is the foundational method for estimating viable bacterial concentrations in liquid samples, solid suspensions, and surface swabs. This protocol converts an unknown bacterial density into a countable number of discrete colonies on solid growth medium, allowing calculation of colony-forming units per milliliter or per gram of original sample. The method assumes each visible colony arises from a single viable cell or clump of cells, and the accuracy of the final estimate depends on careful dilution technique, appropriate plating volumes, and disciplined counting rules. This article provides a complete workflow for laboratory students, technicians, researchers, and diagnostic professionals who need a defensible, repeatable enumeration procedure.

Scope and Purpose of Viable Plate Counting

Viable plate counting measures only living microorganisms capable of forming colonies under the specific incubation conditions provided. Dead cells, injured cells that cannot divide, and organisms requiring different nutrients or atmospheres will not be detected. This limitation distinguishes plate counts from total cell counts obtained by microscopy or from molecular methods that detect nucleic acids regardless of viability. The plate count method remains the primary approach for bacterial quantification in many food, clinical, environmental, and research applications, as noted in recent work on Salmonella quantification in meat samples [14].

The serial dilution step serves two purposes. First, it reduces a dense bacterial suspension to a manageable number of colonies per plate. Second, it provides multiple dilution levels so that at least one plate will fall within the statistically reliable counting range even when the original concentration is unknown. A simple method for estimating microbial counts from serial dilution experiments selects the best agar plate for estimation and accounts for colony size and plate area, both of which contribute to the likelihood of miscounting colonies on a plate [6]. This approach shows relative accuracy within plus or minus 0.1 log10 across a wide range of microbial concentrations and dilution ratios [6].

The protocol described here applies to pure cultures, mixed environmental samples, food homogenates, clinical specimens, and quality control samples. The same principles govern each application, though sample preparation differs. For plant pathogen studies, serial dilution plating is one of two recommended methods for measuring bacterial multiplication in planta, alongside quantitative PCR [13]. For commercial inoculants, traditional culture-based quantification remains the reference method against which faster molecular alternatives are validated [17].

Core Principles of Dilution Mathematics

A serial dilution reduces concentration by a fixed factor at each step. The most common scheme is a tenfold dilution, where one volume of sample is mixed with nine volumes of diluent. The dilution factor at each step is 10, and the cumulative dilution after n steps is 10 to the power of n. A 100-fold dilution can be achieved either by a single 1:100 dilution or by two successive 1:10 dilutions.

The dilution factor must be recorded precisely for every tube. A common error is confusing the dilution of the tube with the dilution of the plate. If 0.1 mL is removed from a 10 to the power of minus 5 tube and spread on a plate, the plate represents a further tenfold dilution of that tube. The total dilution factor for the plate is 10 to the power of minus 6. The countable colonies on that plate must be multiplied by 10 to the power of 6 to estimate the original concentration.

The choice of diluent affects cell survival. Phosphate-buffered saline, sterile saline, or buffered peptone water are standard choices. The diluent should be pre-chilled for samples that require cold chain maintenance. For environmental samples containing particulates, the diluent may include additives to disperse clumps, but any additive must be validated to not inhibit the target organisms.

Dilution errors propagate through the entire calculation. The precision of serial dilutions and viable bacterial counts has been examined in food microbiology literature, and the uncertainty introduced by pipetting and mixing steps contributes meaningfully to the final variance [23]. Using calibrated pipettes, consistent technique, and vortex mixing between transfers reduces this uncertainty. The impact of serial dilution and cell enumeration uncertainty on microbial inactivation studies has also been illustrated, showing that method noise can create apparent patterns in data that are actually artifacts of the measurement process [25].

At a Glance: Key Decisions in the Plate Count Workflow

Required Materials and Equipment

The materials list assumes a standard microbiology laboratory with basic aseptic technique capability. All items must be sterile before use.

Liquid diluent in tubes or bottles, typically 9 mL aliquots for tenfold dilutions. The diluent volume must be accurate because the dilution factor depends on it. Pipettes with sterile tips capable of delivering 0.1 mL and 1.0 mL accurately. A micropipettor with calibrated accuracy or serological pipettes. Sterile spreaders for spread plate technique, or molten agar held at the appropriate temperature for pour plate technique. Prepared agar plates in sterile petri dishes. The medium must support growth of the target organism. A vortex mixer for thorough mixing of dilution tubes. An incubator set to the appropriate temperature and atmosphere for the target organism. A colony counter with a magnifying lens and tally register. Personal protective equipment including a lab coat, gloves, and eye protection.

The Laboratory Biosafety Manual from the World Health Organization provides guidance on the containment practices appropriate for different risk groups of microorganisms [2]. Work with known pathogens requires biosafety cabinet use and additional containment measures. The risk assessment should be completed before beginning any enumeration protocol.

Sample Preparation and Initial Handling

The quality of the final count depends on the quality of the initial sample. A non-representative sample produces a precise but inaccurate result. For liquid cultures, vortex or mix thoroughly before sampling. For solid samples, a homogenization step is required. Weigh a representative portion, typically 10 grams, and add to 90 mL of sterile diluent to create a 10 to the power of minus 1 suspension. Homogenize using a stomacher or blender according to the sample type. For surface swabs, place the swab in a known volume of diluent and vortex vigorously to release adherent cells.

The time between sample collection and plating should be minimized. Holding samples on ice slows metabolic activity and reduces changes in viable count. Some organisms are sensitive to temperature shifts, and the holding conditions should be validated for the specific sample type.

For plant pathogen work, inoculum preparation may require adjustment to a target optical density before dilution plating. The relationship between optical density and cell count varies between instruments and strains, and calibration is required for accurate conversion [7]. An interlaboratory study comparing optical density calibration protocols found that calibrating optical density to estimated cell count using serial dilution of silica microspheres produced highly precise calibration across 244 laboratories [7]. This approach also assesses the instrument effective linear range [7].

Constructing the Dilution Series

Prepare a rack of sterile dilution tubes labeled with the sample identifier and the intended dilution factor. The first tube receives the initial sample. For a tenfold series starting with a liquid culture, add 1 mL of well-mixed culture to 9 mL of diluent. This is the 10 to the power of minus 1 dilution. Vortex for at least five seconds to ensure complete mixing. Transfer 1 mL from the 10 to the power of minus 1 tube to a fresh 9 mL diluent tube to create the 10 to the power of minus 2 dilution. Continue this process until the desired range of dilutions is prepared.

The number of dilutions needed depends on the expected concentration. A culture with an optical density around 1.0 may contain roughly 10 to the power of 8 to 10 to the power of 9 cells per milliliter. To reach the countable range of 25 to 250 colonies per plate with a 0.1 mL plating volume, dilutions through 10 to the power of minus 6 or 10 to the power of minus 7 are typically required. For samples with unknown concentrations, prepare a wider range of dilutions to ensure at least one falls in the countable range.

Each transfer must use a fresh sterile pipette tip. Reusing a tip contaminates the dilution series and invalidates the results. The pipette should be inserted into the liquid, the plunger depressed and released smoothly, and the contents dispensed into the next tube. Touch the tip to the wall of the receiving tube to ensure complete delivery. After each transfer, vortex the receiving tube before taking the next aliquot.

The single plate-serial dilution spotting method offers an alternative for samples with no prior idea of viable counts. This technique applies 20 microliter aliquots of six dilutions as micro-drops in six sectors on a single 9-centimeter plate [9]. Testing on diverse bacteria and yeast showed that at least one dilution level yielded 6 to 60 colonies per sector, comparable to the standard method using 100 microliter samples [9]. This approach reduces material consumption and is particularly suited to low-resource settings [9].

Plating Techniques: Spread Plate and Pour Plate

Two principal plating techniques are used for viable counts. The spread plate method delivers a known volume, typically 0.1 mL, onto the surface of a pre-poured agar plate. The inoculum is spread evenly across the surface using a sterile spreader until the liquid is absorbed. The pour plate method mixes a known volume, typically 1.0 mL, with molten agar in an empty petri dish. The agar is allowed to solidify, embedding the cells within the medium.

The spread plate method is preferred when surface colony morphology is important for identification or when the organism is inhibited by exposure to molten agar. The pour plate method allows a larger sample volume, increasing sensitivity for samples with low cell numbers. However, pour plates require the agar to be cooled to approximately 45 to 50 degrees Celsius before adding the sample, and some organisms are damaged by even brief exposure to elevated temperatures.

The plating volume must be recorded for every plate. The calculation of the original concentration uses both the dilution factor and the plating volume. A plate receiving 0.1 mL from a 10 to the power of minus 5 tube represents 0.1 times 10 to the power of minus 5, or 10 to the power of minus 6 mL of the original sample. Each colony on that plate therefore represents at least 10 to the power of 6 colony-forming units per milliliter in the original sample.

For each dilution to be plated, prepare duplicate or triplicate plates. Replication allows calculation of the mean and provides a check on technique. If replicate plates differ substantially, the results are suspect and the dilution series may need to be repeated.

Incubation Conditions and Timing

Incubation conditions must match the metabolic requirements of the target organism. Temperature, atmosphere, and duration all affect colony formation. Standard bacterial incubation is 35 to 37 degrees Celsius for 18 to 24 hours, but many environmental organisms grow better at lower temperatures or require longer incubation. Anaerobic organisms require an anaerobic chamber or gas-generating sachets. Capnophilic organisms require elevated carbon dioxide.

The incubation duration should be sufficient for colonies to become visible but not so long that colonies merge or satellite colonies develop. Plates should be inspected at 24 hours and again at 48 hours if growth is slow. Some organisms form visible colonies only after several days. The incubation protocol should be validated for each organism and sample type.

Plates should be incubated inverted to prevent condensation from dripping onto the agar surface. Condensation can spread colonies across the plate and make counting impossible. The incubator should be monitored with a calibrated thermometer, and temperature excursions should be recorded.

Colony Counting Rules and Calculations

Counting rules exist to ensure that the selected plate provides a statistically reliable estimate. The conventional countable range is 25 to 250 colonies per plate for the standard 9-centimeter plate. Plates with fewer than 25 colonies have poor statistical precision because each colony represents a large multiple of the original concentration. Plates with more than 250 colonies are difficult to count accurately because colonies may merge or overlap.

The estimation method for serial dilution experiments takes into account colony size and plate area, both of which contribute to the likelihood of miscounting colonies on a plate [6]. The optimal count depends on the ratio of plate size to colony size, and the method can narrow the search for the best dilution plate [6]. For plates where colonies are large or numerous, the reliable counting range may be lower than 250.

When counting, mark each colony with a fine-tipped marker as it is counted to avoid double counting. Use a magnifying lens or colony counter for small colonies. Count all colonies on the plate, including pinpoint colonies. If the plate contains spreading colonies that cover a large area, the count may be unreliable and the next dilution should be used.

The calculation of the original concentration uses the following formula. Concentration in colony-forming units per milliliter equals the number of colonies divided by the plating volume in milliliters, multiplied by the reciprocal of the dilution factor. For example, if a plate receiving 0.1 mL from the 10 to the power of minus 6 dilution shows 150 colonies, the calculation is 150 divided by 0.1 times 10 to the power of 6, which equals 1.5 times 10 to the power of 9 colony-forming units per milliliter.

When multiple plates from the same dilution are counted, use the mean count in the calculation. When plates from adjacent dilutions both fall in the countable range, the results should be combined using a weighted average that accounts for the different dilution factors. Maximum likelihood estimators for colony-forming units provide a more rigorous approach to combining measurements taken at different dilutions [8]. These estimators, including Poisson and truncated Poisson methods, have been tested computationally and their error bounds and assumptions are documented [8].

Recording Results and Units

The standard unit for reporting viable counts is colony-forming units per milliliter for liquids or colony-forming units per gram for solids. The unit must always be stated. A count of 150 colonies from a 0.1 mL plating volume at the 10 to the power of minus 6 dilution is reported as 1.5 times 10 to the power of 9 colony-forming units per milliliter.

Results should be reported to two significant figures. The third significant figure is not meaningful given the inherent variability of the method. A count of 173 colonies should be reported as 1.7 times 10 to the power of 9, not 173 times 10 to the power of 7.

The laboratory record should include the sample identifier, the date and time of processing, the diluent used, the dilution series constructed, the plating method, the volume plated, the medium used, the incubation conditions, the colony counts for each plate, and the calculated concentration. This documentation supports traceability and allows the result to be audited. The Laboratory Quality Management System Handbook from the World Health Organization provides guidance on the documentation and quality assurance practices expected in diagnostic laboratories [1].

Quality Controls and Method Validation

Quality controls verify that the method is performing correctly and that the results are trustworthy. A negative control consists of plating sterile diluent to confirm that the diluent and plates are not contaminated. A positive control consists of plating a reference culture with a known concentration to confirm that the medium supports growth and the technique recovers the expected number of colonies.

Method validation establishes the performance characteristics of the enumeration procedure for a specific sample type. Key parameters include accuracy, precision, limit of detection, and limit of quantification. The Bioanalytical Method Validation Guidance from the U.S. Food and Drug Administration describes the expectations for validation of analytical methods used in regulated studies [4]. While this guidance is written for bioanalytical methods, the principles of demonstrating accuracy, precision, and reproducibility apply to microbiological enumeration.

The Assay Guidance Manual from the National Center for Advancing Translational Sciences provides additional context on assay development and validation [3]. This resource covers experimental design considerations and quality control approaches applicable to quantitative assays.

For commercial products such as bacterial inoculants, the enumeration method must demonstrate acceptable repeatability and reproducibility. A study validating a propidium monoazide quantitative PCR assay for Bradyrhizobium diazoefficiens inoculants reported robust intra-assay repeatability with standard deviation below 0.3 and inter-assay reproducibility with coefficient of variation below 10 percent [17]. The same study showed strong linear correlation with standard plate counting with an R squared of 0.82 [17]. These performance metrics provide a benchmark for what constitutes acceptable method performance.

Common Failure Patterns and Troubleshooting

Several recurring problems compromise plate count results. Recognizing these patterns allows corrective action before results are reported.

No colonies on any plate suggests the sample contained very few viable cells, the diluent or medium was inhibitory, or the incubation conditions were incorrect. Check the positive control to distinguish between a true low count and a method failure. If the positive control grew, the sample itself had few viable cells. If the positive control also failed, the medium, diluent, or incubation conditions are suspect.

Too many colonies to count on all plates indicates the dilution series did not extend far enough. The sample was more concentrated than expected. Repeat the procedure with additional dilutions. For cultures with known growth characteristics, consult previous records to select an appropriate dilution range.

Spreading colonies that cover the plate surface make counting impossible. This often results from condensation on the agar, an overly wet plate surface, or motile organisms. Incubate plates inverted, allow plates to dry before use, and consider using a medium with reduced moisture content.

Colonies that are too small to see without magnification may indicate short incubation or suboptimal growth conditions. Extend the incubation period or adjust the temperature. Some organisms form visible colonies only after 48 to 72 hours.

Replicate plates that differ substantially suggest uneven mixing of the dilution tube or inconsistent plating technique. Vortex each tube thoroughly before sampling and use consistent spreading technique. If replicates continue to disagree, the pipette calibration should be verified.

The single plate-serial dilution spotting method addresses some of these failure modes by providing multiple dilution levels on a single plate, reducing the chance that all dilutions fall outside the countable range [9]. Automatic segmentation methods for CFU counting in single plate-serial dilution have been developed to reduce the labor of manual counting [24].

Limitations of the Plate Count Method

The plate count method has inherent limitations that affect interpretation of results. Only viable cells that can grow under the provided conditions are counted. Injured cells, viable but non-culturable cells, and organisms requiring unprovided nutrients or conditions are missed. The count therefore represents a lower bound on the true number of viable cells.

Clumped cells produce a single colony even if the clump contains multiple cells. The count is therefore expressed as colony-forming units instead of individual cells. For organisms that naturally grow in chains or clusters, such as staphylococci or streptococci, the count may substantially underestimate the true cell number. Isothermal microcalorimetry has been shown to provide accurate viable counts without requiring single cells, in contrast to plating techniques for which clustered cells skew the results [11].

The method is time-consuming and labor-intensive. The full workflow from sample receipt to final count typically requires 24 to 48 hours. This delay is acceptable for many applications but problematic when rapid results are needed. Faster alternatives include quantitative PCR, droplet digital PCR, and isothermal microcalorimetry. A droplet digital PCR assay for Salmonella quantification in meat samples showed a strong linear relationship with plate counting with an R squared above 0.99 [14]. A propidium monoazide quantitative PCR assay reduced processing time for Bradyrhizobium inoculant testing from 120 hours to 5 hours while maintaining strong agreement with the reference method [17].

The precision of the plate count method is limited by the counting range and the number of replicate plates. A single plate provides only a rough estimate. The precision of serial dilutions and viable bacterial counts has been quantified in food microbiology research, and the variance contributions of each step are documented [23]. For applications requiring high precision, additional replicate plates and multiple dilutions are needed.

Biosafety and Waste Handling

Serial dilution and plate counting involve handling live microorganisms, some of which may be pathogenic. The Laboratory Biosafety Manual from the World Health Organization provides the framework for safe handling of biological materials [2]. The risk assessment should consider the organism, the concentration, the procedure, and the potential for aerosol generation.

Vortex mixing and pipetting can generate aerosols. Work with known pathogens should be performed in a biosafety cabinet. Even with non-pathogenic organisms, aseptic technique and careful handling reduce the risk of laboratory contamination. Gloves should be worn and changed when contaminated. Hands should be washed after removing gloves.

All contaminated materials, including pipette tips, spreaders, dilution tubes, and plates, must be decontaminated before disposal. Autoclaving at 121 degrees Celsius for at least 15 minutes is the standard method. Plates should be placed in biohazard bags and autoclaved without opening, as opening plates can release spores or aerosols. Liquid waste should be treated with appropriate disinfectant or autoclaved.

The protective value of personal protective equipment depends on its quality and correct use. A study of isolation gowns found that unrated, Level 1, and Level 2 gowns did not provide effective bacterial isolation barriers when bacteria contacted one side of the gown material [12]. Level 3 and Level 4 gowns provided substantially better protection [12]. This finding underscores the importance of selecting appropriate protective equipment for the level of risk.

Professional Escalation Criteria

Certain results warrant escalation to a supervisor or qualified professional. These criteria protect both the validity of the results and the safety of laboratory personnel.

If a sample from a clinical or food source yields counts exceeding regulatory limits or action thresholds, the result must be reported promptly through the appropriate channels. The specific thresholds depend on the sample type and applicable regulations. The laboratory should have a documented procedure for notifying the relevant authority.

If the positive control fails while the test samples show growth, the validity of the test results is questionable. The run should be repeated after correcting the control failure. Results from a failed control run should not be reported as valid.

If the sample is suspected to contain a pathogen requiring higher containment than is available, work should stop and the sample should be referred to a facility with appropriate capabilities. The Laboratory Biosafety Manual provides guidance on risk group classification and the containment levels required for each group [2].

If repeated attempts to obtain countable plates fail due to technical issues, the method should be reviewed and potentially revised. Persistent problems with spreading colonies, contamination, or poor recovery indicate a systematic issue that requires investigation.

Frequently Asked Questions

Why is the countable range set at 25 to 250 colonies per plate?

The range balances statistical precision with practical counting ability. Below 25 colonies, each colony represents a large multiple of the original concentration, so the relative error is high. Above 250 colonies, colonies may merge or overlap, making accurate counting difficult. The optimal count also depends on colony size and plate area, as described in the estimation method for serial dilution experiments [6].

What is the difference between colony-forming units and cell numbers?

Colony-forming units represent the number of colonies that form on the plate, which may be lower than the true number of cells if cells are clumped or grow in chains. Each colony may arise from a single cell or from a cluster of cells. The count is therefore reported as colony-forming units instead of cells. Isothermal microcalorimetry can provide viable counts without requiring single cells, unlike plating techniques where clustered cells skew results [11].

How do I choose the right dilution range for an unknown sample?

Prepare a wider range of dilutions than you expect to need. For a sample with an unknown concentration, prepare dilutions from 10 to the power of minus 1 through 10 to the power of minus 8. After incubation, select the plate with 25 to 250 colonies for counting. The single plate-serial dilution spotting method can reduce the number of plates needed by testing multiple dilutions on one plate [9].

Can I combine counts from two different dilutions?

Yes, when plates from adjacent dilutions both fall in the countable range, the results can be combined using a weighted average. Maximum likelihood estimators provide a more rigorous approach to combining measurements from different dilutions and have documented error bounds and assumptions [8]. An online calculator for these estimators is available from the authors of that work [8].

Why do my replicate plates show different counts?

Variation between replicate plates arises from pipetting error, incomplete mixing of the dilution tube, and random distribution of cells in the sample. Vortex each dilution tube thoroughly before sampling and use calibrated pipettes. If replicates consistently differ by more than expected, verify pipette calibration and technique.

How does the plate count method compare to molecular methods?

Plate counting measures only viable cells that can grow under the provided conditions, while molecular methods such as quantitative PCR detect nucleic acids from both viable and non-viable cells. Droplet digital PCR has shown strong linear correlation with plate counting for Salmonella quantification [14]. Propidium monoazide quantitative PCR can selectively detect viable cells and has been validated for bacterial inoculant testing [17]. Molecular methods are faster but require validation against the culture-based reference method.

What should I do if my sample contains a pathogen?

Consult the Laboratory Biosafety Manual for the containment level appropriate to the organism [2]. Work with known pathogens in a biosafety cabinet using appropriate personal protective equipment. If the required containment is not available, refer the sample to a facility with appropriate capabilities.

How should I report results with very low or very high counts?

For plates with fewer than 25 colonies, report the count with a note that the result is below the reliable counting range. For plates with more than 250 colonies, report the result as too numerous to count and indicate the lowest dilution that was plated. The estimation method for serial dilution experiments can help identify the optimal plate for counting and narrow the search for the best dilution [6].

Related Diagnostic Guides

• Drop Plate Method for Viable Cell Counting: Protocol and Best Practices
• Spread Plate Method: Protocol, Advantages, and Applications
• How to Use a Hemocytometer for Bacterial Cell Counting: Protocol and Calculations
• Pour Plate Method: Principle, Procedure, and How to Count Colonies
• Spread Plate Method: Advantages, Disadvantages, and Step-by-Step Protocol

References and Further Reading

• Laboratory Quality Management System Handbook. World Health Organization.
• Laboratory Biosafety Manual. World Health Organization.
• Assay Guidance Manual. National Center for Advancing Translational Sciences.
• Bioanalytical Method Validation Guidance. U.S. Food and Drug Administration.
• NCBI Literature Resources. National Center for Biotechnology Information.
• Estimation method for serial dilution experiments.. Journal of microbiological methods, 2014.
• Robust estimation of bacterial cell count from optical density.. Communications biology, 2020.
• Maximum likelihood estimators for colony-forming units.. Microbiology spectrum, 2024.
• Optimization of single plate-serial dilution spotting (SP-SDS) with sample anchoring as an assured method for bacterial and yeast cfu enumeration and single colony isolation from diverse samples.. Biotechnology reports (Amsterdam, Netherlands), 2015.
• A High-throughput Interbacterial Competition Platform.. Bio-protocol, 2020.
• Isothermal microcalorimetry for thermal viable count of microorganisms in pure cultures and stabilized formulations.. BMC microbiology, 2019.
• Isolation gowns as a potential work hazard.. Annals of work exposures and health, 2020.
• Plant Pathogenicity Phenotyping of Ralstonia solanacearum Strains.. Methods in molecular biology (Clifton, N.J.), 2018.
• Droplet Digital Polymerase Chain Reaction Assay for Quantifying Salmonella in Meat Samples.. 2026.
• Protocol for multiplexed detection of antibody binding to cell-surface-expressed hemagglutinin by flow cytometry.. 2026.
• Protocol for determining minimum inhibitory concentrations of essential oils against bacterial pathogens using broth microdilution.. 2026.
• PMA-qPCA: Accelerating the market release of high-quality Bradyrhizobium diazoefficiens inoculant.. 2025.
• Effect of alkaline pretreatment on cephalosporin degradation and microbial susceptibility in wastewater of cephalosporin drug manufacturing plant.. 2026.
• Effectiveness of Antibacterial Extract Bawang Suna (Allium schoenoprasum L.) against Methillicin-Resistant Staphylococcus aureus (MRSA) using Total Plate Count. Majalah Obat Tradisional, 2022.
• Effectiveness Ofantibacterialextract Bawang ( Allium schoenoprasum L.) Against Methicillin-Resistant Staphylococcus aureus (MRSA)Using Total Plate Count. 2022.
• A faster and more economical alternative to the standard plate count (SPC) method for microbiological analyses of raw milks. 2007.
• Comparative studies on microbial activity in Azotobacter chroococcum inoculated wheat (Triticum aestivum) rhizosphere by plate count and manometric methods. 1994.
• Estimating the precision of serial dilutions and viable bacterial counts. International Journal of Food Microbiology, 2002.
• Automatic segmentation method for CFU counting in single plate-serial dilution. Chemometrics and Intelligent Laboratory Systems, 2019.
• Tail or artefact? Illustration of the impact that uncertainty of the serial dilution and cell enumeration methods has on microbial inactivation. Food Research International, 2019.

This article is educational and does not replace validated laboratory procedures, institutional biosafety review, manufacturer instructions, or professional interpretation.