Bottling on a silent airlock leaves residual malic acid in the wine, and that acid can later leak as gas or grow into haze. The measurement that separates a stall from completion is a malic-acid test, read with the method that produced it.
Stopped bubbles leave malolactic fermentation undecided. Lactic acid bacteria convert malic acid to lactic acid and a small amount of carbon dioxide. The Australian Wine Research Institute’s red-wine fact sheet, updated November 2024, says completion is judged by measuring L-malic acid: a “not detected” enzymatic result is usually under 0.05 g/L, and 0.1 g/L or less is low enough to treat that red-wine fermentation as complete. Paper chromatography, which the same institute’s methods page calls a qualitative presence-or-absence test, is a different report and carries no grams-per-litre detection limit on that page. A laboratory number covers the malic acid the named method measured on that date. It leaves residual sugar, sulfur dioxide, heat stability, and cold stability unmeasured.
This page compiles those AWRI documents, a University of Minnesota chromatography note from 7 October 2016, WineMaker Magazine, Denise Gardner’s winery note, and Purdue Extension FS-56-W (New 03/10), all read on 13 October 2026. It is a way to choose a test and file the result. It is not a recipe for starting or restarting the fermentation, and it sets no completion figure of its own.
Key points:
- Silence is undecided. No bubbles can be a finished fermentation or a stall.
- Match the test you can run. Chromatography, an enzymatic assay, and a laboratory report answer different questions.
- Copy the detection limit. A “not detected” line means “under this method’s limit,” which you write down from the report.
- Keep the published lines apart. AWRI’s enzymatic figures and Purdue’s chromatography and commercial-assumption figures are different statements.
- File six fields. Method, unit, detection limit, date, source, and the next check.
Photograph of a calm wine surface. The picture shows no rising bubbles. It does not show whether malic acid remains. This is not a batch from this site.
When the bubbles stop, has malolactic fermentation finished or stalled?
A stopped bubble stream is compatible with both. Malolactic fermentation is the bacterial conversion of malic acid. It releases some carbon dioxide, so a carboy can show tiny bubbles while it is working, and it can sit quiet while malic acid is still there.
Alison Crowe, answering a Finger Lakes reader in WineMaker Magazine, writes that this step is nothing like the yeast fermentation you can see and hear. If you inoculated, pinprick bubbles at the bung are sometimes audible and sometimes absent. She notes that cool, high-acid wines are especially quiet, and that a slow conversion can take months. During that quiet stretch the wine is short of the carbon dioxide blanket a vigorous ferment provides, so spoilage and volatile acidity are the practical risk she names. The month count is her description of a slow fermentation. It is not a deadline.
Denise Gardner makes the cellar version of the same point. Wineries that stop watching once the gas disappears have bottled wine whose malolactic fermentation was only paused. She ties that pause to cold storage: lactic acid bacteria slow as the cellar cools and can start again when the wine warms. Her article tells the reader to measure residual malic acid and to treat the wine as finished only when that measurement shows it is “dry.” The article, as saved, does not print a number for “dry.”
The AWRI red-wine sheet, updated November 2024, puts the same job on a schedule of L-malic measurements. Regular numbers show whether the fermentation has started, is delayed, or is stuck. The sheet also notes that with co-inoculation, malic acid can already be gone before the yeast has finished the sugar. In that case the quiet carboy still needs the alcoholic-fermentation checks. Those checks, gravity and residual sugar, are on the fermentation overview. This page does not repeat them, and it does not choose an inoculation time or a bacterial strain.
A stall, on the evidence above, is malic acid that is still present after the gas has stopped. Completion is a malic result that meets one named source’s line for the method that produced the result. Until that result exists, the batch stays in the undecided column.
Takeaway: Write “undecided” when the only observation is a quiet carboy, then run a malic test you can actually obtain.
Which malic test can you run, and what does each one report?
Use the test you can get this week. The AWRI methods page groups the common measurements as paper chromatography, thin-layer chromatography, an enzymatic assay, and high-performance liquid chromatography (HPLC). The AWRI laboratory-scope page says malic acid is measured to follow malolactic fermentation and to confirm it is complete, and it lists the two chromatography methods as qualitative. Heat stability and cold stability sit on that same scope page as separate tests. They are not part of a malic result.
| Test you can obtain | What the report is | Unit to copy | Detection limit in the sources read here | Choose it when |
|---|---|---|---|---|
| Paper chromatography, or thin-layer chromatography | A spot compared with acid standards. AWRI calls both qualitative. Purdue calls paper chromatography semi-quantitative | Present or absent. Purdue also describes how visible the malic spot is | AWRI methods page: no g/L limit. Purdue FS-56-W (New 03/10): no malic spot indicates under 30 mg/L; a barely visible spot is about 200 mg/L | You can run a chromatography kit and you need to know whether a malic spot is still there |
| Enzymatic assay | A number. AWRI describes a specific enzyme converting malic acid, read as absorbance at 344 nm from the by-product NADH. Kits exist and the sample is diluted | g/L or mg/L, as the kit or the report prints it | The methods page states none. AWRI’s November 2024 red-wine sheet says a “not detected” enzymatic result is usually under 0.05 g/L | You have a UV spectrophotometer and a malic kit, or a report that names this method |
| Laboratory enzymatic assay or HPLC | The number plus the method name on the report. AWRI describes HPLC as precise and relatively fast, with a high capital cost, and as something wineries use when they already run HPLC for organic acids | The unit printed on the report | Copy the detection limit from the report. If the report omits it, write “not stated”. Do not insert 0.05 g/L | You need a number with a stated limit, and you do not have the instrument |
Photograph of a bench spectrophotometer and cuvettes. The screen is dark. Nothing in the picture is a malic-acid reading.
University of Minnesota enologists Drew Horton and Matthew Clark, on 7 October 2016, show what the paper test looks like. Their figure uses tartaric, malic, and lactic standards beside four red wines from the Horticultural Research Center. They read “very little malic acid evidence” on that sheet as those four wines having gone through malolactic fermentation, and they note that lactic acid travels farthest and that tartaric acid remains. That is a reading of one chromatogram. It is not a grams-per-litre cutoff. They also say the solvent needs ventilation. The AWRI methods page says the same about both chromatography solvents and points to a fume cupboard or other ventilated bench. This page does not repeat a solvent recipe.
Crowe’s practical split matches the table. Most home winemakers, she writes, do not own the enzymatic equipment, so they either run paper chromatography or they have a wine laboratory do the enzymatic assay. She is explicit that chromatography is an indicator: experienced readers watch the malic spot shrink and the lactic spot grow, and the assay is still not a good quantitative measurement. She prefers the enzymatic number partly because the chromatography solvent is toxic in a house. Gardner is stricter about the spot test: do not treat paper chromatography, or the lack of gas, as proof the fermentation has finished. Measure residual malic acid.
Thin-layer chromatography is the faster qualitative cousin. AWRI says it works like the paper method and cites Iland and co-authors (2000) for the speed advantage. It still identifies acids by how far the spots move. It still does not print a detection limit in g/L on the AWRI methods page.
Takeaway: If the only test in reach is a chromatogram, use it to see whether a malic spot remains, and plan a quantitative measurement before you apply a g/L completion line.
What does a laboratory malic result cover?
A laboratory malic number covers the acid the method measured on that date. Read the report for four things before you compare it with anyone’s completion line: the method name, the unit, the detection limit, and whether it says L-malic acid or malic acid.
The November 2024 red-wine sheet asks for L-malic acid concentration. Its completion wording is tied to enzymatic analysis: aim for “not detected,” usually under 0.05 g/L, and treat 0.1 g/L or less as low enough for the malolactic fermentation to be considered complete. The September 2020 sheet, Achieving successful malolactic fermentation, uses the same two figures and a wider scope. That sheet covers most red wines and some white and sparkling wines, and it says 0.1 g/L or less is low enough to consider the fermentation virtually complete and to limit spoilage risk. Same numbers, different wording, different wine scope. Use the sheet that matches the wine in the vessel, and write that sheet’s name in the record. Do not blend the two sentences into a new threshold.
The detection limit is the boundary of “not detected.” If the report prints a limit of 0.2 g/L, a not-detected result means the acid was under 0.2 g/L by that method. It does not mean the acid was under the 0.05 g/L figure AWRI uses for a usual enzymatic not-detected result. If the report prints no limit, write “not stated” and do not borrow 0.05 g/L from the fact sheet to fill the gap. HPLC, on the methods page, is calibrated with malic-acid standards in water. That page does not state an HPLC detection limit either, so an HPLC report has to carry its own.
Photograph of unlabeled sample vials. The glass does not carry a method, a unit, or a detection limit. Those fields come from the report.
What the number leaves out is the rest of stability. The AWRI scope page lists alcohol, residual sugar, sulfur dioxide, heat stability, and cold stability as other determinations. A malic result does not measure them. It does not show that bacteria are gone. It does not show that the wine can be bottled. Gardner’s point about a paused fermentation finishing in the package is the reason: malic acid left above the line you chose can still be a substrate later. The bottling guide is where those other checks sit. Whether the fermentation was meant to run on the fine lees is a different decision, already set out in the first-rack guide.
Takeaway: Transcribe method, unit, detection limit, and date from the report, then say in one line which published completion statement you are comparing that number with.
How do you fill in the malic-result record?
One row per measurement. Leave every cell blank until you have the report or the chromatogram in front of you. The row below is the form, not an example result.
| Field | What to write | What to leave out |
|---|---|---|
| Method | Paper chromatography, thin-layer chromatography, enzymatic, or HPLC, and whether a laboratory ran it | “Bubbles stopped” or “tasted finished” |
| Unit | Present/absent, mg/L, or g/L, exactly as printed | A unit converted from a different method |
| Detection limit | The limit printed on the report or kit insert. For a chromatogram with no printed limit, write “qualitative; no g/L limit on the AWRI methods page” unless you are applying Purdue’s spot descriptions and you name Purdue | 0.05 g/L, unless the report itself says that |
| Date | The day the sample was measured | The day the airlock went quiet |
| Source | One completion statement, with wine scope, method, and year | An average of AWRI and Purdue |
| Next check | The matching row in the small table under this one | A bottling date based on this row alone |
| This row shows | Next check |
|---|---|
| Chromatography only, spot present or absent | A quantitative malic measurement, enzymatic or laboratory, before you apply a g/L completion line |
| A number above the single source you named | Another malic measurement. The November 2024 red-wine sheet calls temperatures under 16 °C or over 25 °C unfavourable for the fermentation, and Gardner notes that cold can pause it. This page does not give a restart recipe |
| A number at or under that named line, with a detection limit fine enough to support the line | The other pre-bottling checks on the bottling guide. This malic row does not cover them |
“Fine enough” is a comparison you can do on the form. A detection limit of 0.2 g/L cannot support a completion line of 0.1 g/L or 0.05 g/L, because “not detected” at 0.2 g/L still allows acid between 0.1 and 0.2 g/L. In that case the next check is a method with a tighter stated limit, or a reported number that sits on the correct side of the line you named.
Takeaway: The record is finished when all six fields are filled from documents, and the next check is a test, not a guess from the airlock.
Why do the published completion lines differ?
They differ because they describe different methods and, in Purdue’s case, a commercial assumption beside a spot reading. Putting them in one table keeps the figures attached to the sentence that stated them.
| Statement as read | Scope | Figure | Where it is written |
|---|---|---|---|
| “Not detected” is usually under 0.05 g/L by enzymatic analysis. 0.1 g/L or less is low enough to consider the fermentation complete | Red wine; L-malic acid; enzymatic | 0.05 g/L and 0.1 g/L | AWRI, Malolactic fermentation in red wine, updated November 2024 |
| “Not detected” is usually under 0.05 g/L by enzymatic analysis. 0.1 g/L or less is low enough to consider the fermentation virtually complete and to limit spoilage risk | The September 2020 sheet’s scope: most red wines and some white and sparkling wines; enzymatic | 0.05 g/L and 0.1 g/L | AWRI, Achieving successful malolactic fermentation, updated September 2020 |
| No malic spot indicates under 30 mg/L. A barely visible spot is about 200 mg/L. The method is semi-quantitative | Paper chromatography | 30 mg/L and about 200 mg/L | Purdue Extension FS-56-W, New 03/10 |
| It is commercially assumed that completed malolactic fermentation leaves under 300 mg/L (0.3 g/L). A visible haze in white wine from Oenococcus oeni can be expected above 300 mg/L | A commercial assumption, plus a white-wine haze statement | 0.3 g/L | Purdue Extension FS-56-W, New 03/10 |
| Measure residual malic acid and confirm the wine is “dry.” Do not treat chromatography or the absence of gas as completion | Winery monitoring | No figure in the saved article | Denise Gardner |
| “Very little malic acid evidence” on a chromatogram of four red wines, read against tartaric, malic, and lactic standards | Those four wines at the University of Minnesota Horticultural Research Center | No g/L cutoff | Horton and Clark, 7 October 2016 |
A printed enzymatic result of 0.2 g/L would sit under Purdue’s 0.3 g/L commercial assumption and above the AWRI red-wine line of 0.1 g/L. That sentence only compares the two published lines. It is not a reading from a batch, and it is not a third threshold. The Source cell of the record is where you pick one line and keep the other out.
The fermentation overview already records Purdue’s spot reading and the 0.3 g/L commercial assumption as one of four finish checks. This page does not replace that check. It adds the method choice, the detection limit, and the rule against averaging.
Takeaway: Cite one source in the record. A second source with a different figure stays in this table, not in the same cell.
How this page was put together
The claims above come from the public documents listed at the end, read on 13 October 2026 from saved copies. AWRI states the enzymatic and HPLC methods and both of its completion wordings. Purdue FS-56-W states the chromatography spot readings and the 0.3 g/L commercial assumption. Gardner states that gas and paper chromatography are not completion, and does not state a number. The Minnesota note states how one chromatogram was read. Crowe states why sight and sound fail and why an enzymatic number is a different kind of result. No figure on this page was measured here, averaged, or extended to a wine the source did not name. Kit prices and home solvent steps are left out. The photographs are scene pictures of still wine, a bench instrument with a dark screen, and unlabeled vials. They are not a cellar log and not a test result.
