The short answer
Most plate count variation comes from technique, not from the medium. The four biggest factors: the diluent (type, temperature, holding time), the temperature of the medium when poured, the colony count range you accept, and whether the sample was homogenised. Lot-to-lot variation in a medium that meets specification is usually under 10%, while technique routinely introduces 30 to 50%. Troubleshoot technique first and test the medium with reference strains last.
Know what normal looks like
Plate counts carry inherent statistical variation. Two duplicate plates at the same dilution, both with 30 to 300 colonies, normally differ by 10 to 20% under the Poisson distribution; plates with only 10 colonies can differ by 50% without anything being wrong. So the first check is that you are comparing plates within 30 to 300 (or whatever range the method specifies); plates below 25 should not be counted at all.
Checking in order of impact
1. Diluent
- Type. Use what the method specifies. 0.1% peptone water, buffered peptone water, phosphate buffer and physiological saline differ in how well they let injured cells recover; plain water kills cells by osmotic shock.
- Temperature. Diluent straight from the refrigerator cold-shocks cells. Let it reach room temperature.
- Holding time. Complete pouring within 15 to 20 minutes of diluting; cells die or multiply in diluent left standing.
- Volume. A 9 mL dilution tube that has evaporated to 8.5 mL in the autoclave introduces a 5% error per step, 15% over three steps. Check volumes after sterilisation.
2. Pouring temperature
Hold the medium at 44 to 46 °C. Too hot (above 48 °C) kills a fraction of cells, especially injured ones; too cool and it sets early and mixes unevenly. Calibrate the water-bath thermometer, and allow 30 minutes or more after placing the bottle in the bath before it is truly at temperature.
3. Plate depth and drying
Pour 15 to 20 mL. Thin plates dry out over 48 hours and colonies stay small; thick plates hide colonies in the lower layer. For spread plates, dry the surface until no droplets remain or colonies will run together.
4. Medium storage
Use poured plates within 7 to 14 days of refrigeration, bagged to prevent drying; plates held longer have dehydrated surfaces and give low counts. Reseal opened powder against moisture; see Storing dehydrated media and shelf life.
5. Counting range and reading
25 to 250 or 30 to 300, as the method states. If only one dilution is in range, use only that one; if two are, apply the method's weighted formula rather than a simple average. The same analyst recounting the same plate should agree within 5%, different analysts within 10%; beyond that, harmonise the reading criteria (what counts as one colony, how spreaders are handled).
6. Incubation conditions
Incubator temperature is not uniform; top and bottom shelves can differ by 2 °C, and stacks of plates warm slowly in the middle. A 2 °C or 4-hour difference can shift a 48-hour aerobic count by 20%.
7. Sample homogenisation
Incompletely homogenised solid samples or unshaken liquids make the 1 mL you take unrepresentative. This is the largest source of between-sample variation and has nothing to do with the medium.
8. The medium itself
Only after the above are excluded, test the medium with reference strains. Inoculate quantitatively with ATCC strains (for example E. coli ATCC 25922, S. aureus ATCC 25923), five plates each of old and new lot, and compare recovery. A difference above 30%, or recovery on the new lot below what the method requires, indicates a medium problem. STBIO ships a COA with every lot and cultural response data is available for comparison.
A checklist
| Item | Standard |
|---|---|
| Diluent | Type specified by the method, room temperature, volume checked after sterilisation |
| Dilution to pouring | Within 20 minutes |
| Medium temperature | 44 to 46 °C, measured |
| Pour volume | 15 to 20 mL |
| Plate storage | Refrigerated, bagged, within 14 days |
| Counting range | Per method; never below 25 |
| Incubator | Calibrated; stacks no higher than 6 plates |
| Controls | Positive control strain and blank with every batch |
FAQ
Counts dropped after switching medium brand. Is the new brand worse? Not necessarily. Brands differ within specification, but more often the preparation changed with the brand (grams per litre, sterilisation time). Compare side by side with reference strains, five plates per brand, and look at recovery.
Duplicate plates differ by a factor of two. Repeat? If both counts are in range and still differ twofold, that exceeds statistical variation; repeat, and check diluent mixing and pipetting technique.
Which peptone for 0.1% peptone diluent? An ordinary bacteriological peptone, for example Peptone, Bacteriological; this is what the Taiwan FDA food methods and most international methods specify.