A bare tank, a heater, a sponge filter that has been running in your display, a lid and somewhere dim to put it — under an hour to set up. Then run it for at least 2 complete parasite life cycles, which is what a quarantine period actually is. At a typical 72 °F that means about 22 days for a comfortable three cycles — the hobby's four weeks, arrived at honestly. Warm the tank to 82 °F and the same confidence takes 13 days.
Quarantine is the least glamorous thing in the hobby and the one most likely to save a tank you have spent years on. It is also the subject with the widest spread of confident advice: search it and you will be told two weeks, four weeks, six weeks and eight weeks, by sources that all sound equally sure and none of which say where the number came from.
It can be worked out. What you are waiting for is a life cycle to complete more than once without anything appearing — and that cycle runs on a clock set by temperature.
Is it worth it?
The honest case is not that new fish are usually sick. Most are fine. The case is that the costs are lopsided:
- What it costs. A second-hand 10 or 20 gallon tank, a heater, a sponge filter and an air pump — roughly the price of two or three mid-range fish, once, and it lives in a cupboard between uses.
- What it prevents. Not one dead fish. A parasite loose in a stocked display, where you cannot medicate freely because of the invertebrates, cannot catch anything without tearing the scape apart, and cannot go fallow without somewhere to put twenty fish.
- What else it buys. A hospital tank on the day you need one, a grow-out for fry, somewhere to isolate a bully, and a safe place to let a stressed shipment eat and recover before it competes with established fish.
That last point is underrated. Even with a perfectly healthy fish, a couple of weeks eating properly in a quiet tank puts it into the display in far better shape than going straight from a shipping bag into a territory dispute.
What you actually need
- A tank, 10–20 gallons, sized to the fish rather than to the display. Bare glass. Second-hand is fine.
- A sponge filter and an air pump. The single most important item, and the reason is timing — see below.
- A heater and a thermometer. Not optional, and not only for tropical species: the temperature is what sets your quarantine length.
- A lid. Stressed new fish jump, and a bare tank gives them nothing to feel safe under.
- Somewhere to hide. A length of PVC pipe or a plain ceramic cave. No substrate, no wood, no live plants.
- A liquid test kit — ammonia and nitrite especially, for reasons the parameters chart covers.
- Its own net, siphon and bucket, kept with the tank and never used in the display. Cross-contamination through shared equipment is how a quarantine quietly fails.
Three reasons, all practical. Medication binds to substrate, wood and some decor, so a treated bare tank actually delivers the dose on the label. Ich cysts drop off the fish and settle — on bare glass you can siphon them out, in gravel you cannot. And you can see everything: waste, uneaten food, a spot on a flank, a fish breathing oddly in a corner. Observation is the entire product here.
Setting it up
- Put the sponge filter in your display tank now — weeks or months before you need it. This is the whole trick. A filter is only useful once it carries a bacterial population, and a quarantine tank set up on the afternoon you bring fish home is an uncycled tank containing a stressed animal. Leave a spare sponge running in the display permanently and it is ready whenever you are.
- Fill from the display, not the tap, on setup day. Using water the fish will eventually live in removes one parameter change from an animal that has already had several.
- Move the seeded sponge across and get the air pump running. Keep it wet in transit; a few minutes is nothing, an hour dry is the colony.
- Set the heater and let it stabilize before the fish arrives. See the next section for what to set it to — this is a decision, not a default.
- Dim it and cover it. A towel over three sides and low light: new fish settle faster and jump less in a tank that does not feel exposed.
- Acclimate and add the fish, then leave it alone for a day. How much acclimation depends on where it came from — a shipped bag wants less of it, not more — and no food on day one, because a fish that has been in a bag is not hungry and uneaten food in a bare tank is pure ammonia.
How long, and why
Here is the part every other guide asserts. A quarantine is not a number of weeks, it is a number of complete life cycles you have watched go by without seeing anything. One cycle proves nothing — it only means nothing has hatched yet. Two is the minimum that carries information. Three is comfortable.
The organism setting that clock in freshwater is ich, and its cycle is strongly temperature-dependent: 21 days at 49 °F, 6 days at 76 °F. Fit those two published figures and you get a rate response of about 2.3× per 10 °C — which turns the folk numbers into arithmetic:
| QT temperature | One cycle | 2 weeks covers | 4 weeks covers | For 3 clear cycles |
|---|---|---|---|---|
| 64 °F (18 °C) | 10.3 days | 1.4 cycles | 2.7 cycles | 31 days |
| 68 °F (20 °C) | 8.7 days | 1.6 cycles | 3.2 cycles | 26 days |
| 72 °F (22 °C) | 7.4 days | 1.9 cycles | 3.8 cycles | 22 days |
| 75 °F (24 °C) | 6.3 days | 2.2 cycles | 4.5 cycles | 19 days |
| 79 °F (26 °C) | 5.3 days | 2.6 cycles | 5.3 cycles | 16 days |
| 82 °F (28 °C) | 4.5 days | 3.1 cycles | 6.3 cycles | 13 days |
Computed by this site's engine from a two-point fit through the published cycle lengths (De-Hai Xu, USDA-ARS, in Tropical Fish Hobbyist: 21 days at 9.5 °C, 6 days at 24.5 °C). A fit through two figures, not a laboratory curve — read the ratios, and give yourself margin rather than counting to the last day.
Two things fall out of that table, and they point in opposite directions from the advice you usually get.
The hobby's four weeks is right. At any normal room temperature it covers between three and five complete cycles, which is exactly the margin you want. It is a good number that has been passed around without its reasoning for so long that it looks arbitrary.
"Two weeks" is not, unless you heat the tank. At 64 °F a fortnight is 1.4 cycles — less than the minimum that tells you anything, and a quarantine that ends before a single generation has completed is a waiting room, not a quarantine. In an unheated spare room in winter, even four weeks only reaches 2.7 cycles.
Which makes the heater setting a lever rather than a default. Running the quarantine at 79–82 °F, where the species tolerates it, compresses the cycle to 4.5–5.3 days and gets you three clean cycles in 13–16 days. Warmer water holds less oxygen, so raise the airflow when you raise the heat, and do not push a coldwater species up there to save yourself a week.
The parasite is only vulnerable for part of its life. It spends 5–7 days embedded under the fish's skin, where nothing you add to the water can reach it — that is the white spot you can see. It then drops off, encysts on a hard surface, and divides into hundreds of free-swimming theronts, which have only 2–3 days to find a fish before they die. That short window is why an empty tank clears itself, why bare glass you can siphon beats gravel you cannot, and why the cycle length is the number that matters. Above about 30 °C it cannot complete the cycle at all, which is where heat treatment comes from — a different subject from waiting one out.
The daily routine
Quarantine is mostly not doing things. The routine that matters is short:
- Look, every day, for a few minutes. Not a glance — actually watch. You are building a baseline of what this fish normally does so that a change is obvious.
- Feed lightly and remove anything uneaten. In a bare tank with a young filter, leftover food is the main ammonia source.
- Test ammonia and nitrite every day or two for the first couple of weeks. A small tank has almost no buffer against a filter that is not quite keeping up.
- Change water when a test says so, not on a schedule — matched for temperature, dechlorinated, and taken from the display if that is easy.
- Siphon the bottom. On bare glass this takes a minute and it physically removes cysts as well as waste.
Quarantine is where ammonia is most likely to appear — small volume, young filter, a stressed fish — and how much that matters depends on water you may not have thought about. A 0.5 ppm reading at 26 °C is 0.0039 mg/L of free NH₃-N in soft water at pH 7.2, which is under the long-term exposure guideline of 0.02 mg/L. The identical reading at pH 8.0 is 0.0236 mg/L — over the long-term exposure guideline. Running the tank warm adds to it too: the pH 8.0 case at 20 °C would be 0.0157 mg/L. The parameters chart has the full picture. The practical version is that a seeded filter and a light hand with food matter more in quarantine than anywhere else — and that hard-water keepers have less room than the number on the test card suggests.
What you're watching for
Observation, not diagnosis. What follows is what makes you look closer, not a key to what is wrong:
- Not eating, after the first day or two of settling in.
- Clamped fins, held tight to the body.
- Flashing — rubbing against surfaces.
- Rapid gilling, or hanging at the surface.
- Anything visible: white spots, fuzz, ragged fins, cloudy eyes, sores, unusual coloration.
- Sitting on the bottom or hiding permanently when the species does not normally.
- Stringy white feces, which often points to something internal.
If you see any of it, the useful next step is to identify the problem before treating it, and the timer starts again from the day symptoms clear. We do not publish medication dosing or disease identification on this site — that is a job for a good species reference or an aquatic vet, and a page that guessed at it would be worth less than nothing. What this page can tell you is that treating blind removes the one thing quarantine gives you, which is information.
If the display is already infected
The same arithmetic runs backwards. With every fish removed, the parasite has nothing to complete its cycle on: the encysted stage hatches, the theronts get their 2–3 days to find a host, and they fail. Leaving the display fallow — fishless and warm, with the filter still running — clears it without medicating the tank at all. Give it the same three cycles you would give a quarantine, plus margin, and keep the temperature up so those cycles run quickly.
Everything on this page is freshwater ich, Ichthyophthirius multifiliis. Saltwater white spot is Cryptocaryon irritans: a different parasite with a much longer and more variable cycle, and a fallow period measured in months rather than weeks. The two get conflated constantly because they look the same on the fish. If you are running a reef or a FOWLR tank, take the approach here and none of the numbers.
Ending quarantine
- Count the cycles, not the days on the calendar — the table above turns one into the other for your temperature.
- Bring the temperature back down slowly if you raised it, over a day or two, to whatever the display runs at.
- Match the water. Check pH and hardness in both tanks before the move; if they have drifted apart, drip the fish across rather than netting it straight over.
- Move the fish, not the water. Net it across and leave the quarantine water behind. There is no reason to transfer several gallons of the water you have spent a month watching.
- Put the sponge filter back in the display to keep it alive for next time, unless the tank was treated — in which case rinse everything, dry it, and reseed a fresh sponge.
What you'll be told, and what's true
- "Two weeks of quarantine is enough." It depends entirely on the temperature, which is why the flat answer is wrong. At 82 °F a fortnight covers 3.1 cycles and is genuinely fine. At 64 °F it covers 1.4 — not even two — and tells you almost nothing.
- "A quarantine tank needs to be a proper setup with substrate and plants." The opposite. Bare glass lets you see and siphon what settles, and stops substrate and decor absorbing any treatment you end up using. A quarantine tank should look empty.
- "You can quarantine in a breeder box in the display." That is isolation, not quarantine. The whole tank shares water, and every parasite that matters travels in it. A breeder box is useful for keeping a fish from being bullied and useless for keeping a pathogen out.
- "It looked healthy in the shop, so it's fine." The gap between infection and visible symptoms is exactly what the waiting period exists to cover — a fish carrying something can look perfect on the day you buy it. That is also why the clock restarts if you add a second fish partway through.
Common questions
Do I really need a quarantine tank?
Plenty of people keep fish for years without one. The case for it is not that trouble is likely, it is that the cost is lopsided: a quarantine setup is roughly the price of two or three fish, and the thing it protects against is losing everything in the display at once. If you buy from one shop you trust and you are prepared to accept that risk occasionally, skipping it is a defensible choice — just make it as a choice rather than by accident.
Can I quarantine in a bucket or a plastic tote?
Yes, and it works fine. A food-grade bucket or storage tote with a heater, a sponge filter, a lid and somewhere dark to sit is a quarantine tank in every way that matters. You lose the ability to see the fish from the side, which is a real loss when the whole point is observation — a cheap glass tank is better for that alone. What you cannot do is skip the heater or the filter.
Should I medicate during quarantine even if the fish looks fine?
This page does not give medication doses, deliberately, and prophylactic treatment is a genuine disagreement among experienced keepers rather than a settled question. What is not in dispute: medicating blind removes your ability to see what is actually wrong, several common treatments damage the filter you are relying on, and blanket dosing contributes to resistance. If a fish shows symptoms, identify the problem first — a species reference or an aquatic vet, not a guide page.
Do plants and snails need quarantining too?
They can carry the same parasite — ich cysts do not care whether they arrived on a fish or a plant — along with pest snails, planaria and hitchhiking eggs. A plant quarantine is a different process from a fish one, though: plants can sit in a lit, fishless container for a few weeks, which starves out anything that needs a host, and that is the same reasoning as the fallow period below.