Spotting a sick fish

The signs, from subtle to serious

Healthy fish hold their fins open, feed eagerly, sit at their normal level in the water column, and breathe steadily. Illness shows as departures from that baseline — which is why knowing your fish when they are well is what lets you catch problems when they are not.

Behaviour shifts before the body does. Watch for fish hanging at the surface gulping, or sitting on the bottom; "flashing" — darting to rub against surfaces; fins clamped tight to the body; loss of appetite; hiding or listlessness in a normally active fish; separation from a shoal. Physical signs come later: rapid or laboured gilling, faded or darkened colour, white spots or patches, frayed or reddened fins, a bloated body with raised scales, ulcers, stringy white waste.

• Behaviour: gasping at the surface, sitting on the bottom, flashing, hiding, off food
• Breathing: fast or laboured gill movement
• Fins: clamped, frayed, or reddened at the base
• Body: white spots, cottony tufts, ulcers, bloating with raised scales, colour loss
• Waste: long, pale, stringy faeces

Check the water before the medicine cabinet

When a fish looks sick, test the water first — not the fish. Ammonia and nitrite should read zero in an established tank; any trace is toxic and a common cause of gasping, lethargy, and reddened gills that get mistaken for infection. Check nitrate and temperature too, and confirm the filter is running — a missed water change, an overstocked tank, a dead fish decaying out of sight, or a stalled biological filter will sicken fish that no medication can fix.

Correcting the water — a careful water change, restored filtration, a stable temperature — is often the whole treatment, and it is the right first move even when a true disease is also present, because clean water gives a fish the reserves to recover. A liquid test kit reading ammonia, nitrite, nitrate, and pH is the most useful diagnostic tool most aquarists own.

A quick triage: signs to likely causes

Once the water checks out, the pattern of signs narrows the field. Treat this as a starting point for which guide to read next, not a final diagnosis — several conditions overlap, and confirming often takes a closer look. Matched against clean water, though, these associations hold up well and stop you treating the wrong thing.

Orient first, then confirm. A flashing fish with no visible marks could be flukes, early ich, or simply an irritant in the water; only a closer look will say which.

• Discrete white salt-grain spots → ich (white spot)
• Fine gold/rust dust, best seen in side-light → velvet
• Flashing/rubbing with no visible marks → flukes, early ich, or a water irritant
• Cottony white tufts → water mould (Saprolegnia, an oomycete — not a true fungus) or, if fast-spreading, columnaris
• Ragged, receding fins → fin rot (check the water first)
• Bloated body with raised scales → dropsy (systemic, guarded prognosis — see below)
• Bulging eye(s) → pop-eye (one eye trauma; both eyes systemic)
• Floating/sinking/listing → swim-bladder disorder
• Laboured gilling, gasping → gill problem, low oxygen, or water quality
• White, stringy faeces + wasting → internal infection (e.g. Spironucleus)

Differential diagnosis: the look-alikes that fool people

Several of the most common presentations share enough features to trigger the wrong treatment — and the wrong treatment at the wrong time can kill faster than the original disease.

Ich vs. velvet: ich cysts sit on the surface as discrete white grains, each large enough to count; they are not attached to individual scales. Velvet (Piscinoodinium or Oodinium) is a finer, velvety or gold dust distributed across the body and often the gills first — in early cases it is visible only in raking sidelight. A skin scrape under a microscope resolves the question immediately: ich shows large (~450 µm), spherical, ciliates with a horseshoe-shaped macronucleus; velvet shows smaller (~100 µm) dinoflagellates with a visible chloroplast in the early trophont stage. Both flash and scratch, but velvet tends to affect gills before the body surface is obviously dusted, so rapid gilling in what looks like an otherwise normal fish is a velvet signal.

Flukes vs. ich vs. irritant: a fish flashing with no visible marks on the body at all is almost never ich — the cysts would be visible by the time flashing becomes frequent. Monogenean gill flukes (Dactylogyrus, Gyrodactylus) produce intense flashing and respiratory distress with a clean body surface. A gill or skin scrape showing actively moving, hook-bearing worms confirms flukes; their absence, combined with clean water chemistry, points to a chemical irritant (chloramine, new silicone outgassing, new decor leaching compounds).

Columnaris vs. Saprolegnia: both produce white-to-grey cottony patches, but the framing differs. Saprolegnia growths are fluffy, three-dimensional tufts, often following an injury or stress event, and spread slowly. Columnaris (Flavobacterium columnare) typically starts as a pale saddle-patch behind the dorsal fin or around the mouth and spreads fast — within 24-48 hours in warm water. Raising temperature, which is standard advice for ich, dramatically accelerates columnaris; confirming the distinction before altering temperature is therefore important. A scrape showing a 'haystack' or rotating-column motility pattern under phase contrast, or the characteristic yellow-green pigment under magnification, is strongly suggestive of columnaris.

Dropsy vs. bloat from overfeeding/constipation: raised, pinecone-like scales (the Koi term 'pineconing') indicate fluid accumulation in the body cavity and are genuinely diagnostic of dropsy. Uniform swelling without scale protrusion — especially in goldfish, cichlids, or goldfish-adjacent species — is more likely to be constipation or fatty liver from diet. The distinction matters because dropsy carries a grim prognosis and warrants isolation; a constipated fish often recovers with a fast and high-fibre diet.

Treatment tradeoffs and contraindications

Before reaching for any agent, identify which fish and which system will be exposed — the same medication that is routine in one context is dangerous in another.

Scaleless and soft-scaled fish: loaches (Cobitidae, Botiidae), most corydoras and other callichthyid catfish, mormyrids, many tetras, and eels lack the protective scale layer that buffers lipid-soluble compounds. Salt, formalin, organophosphates, and many copper-based treatments penetrate their skin at a fraction of the dose tolerated by scaled fish. If these species are in the system, the effective treatment window is narrow. Hospital-tank treatment of the affected individual — without the sensitive species — is strongly preferable.

Copper and invertebrates: copper compounds (ionic copper, chelated copper) are lethal to shrimp, snails, and other invertebrates at concentrations well below therapeutic levels for fish. A single treatment cycle in a display tank with snails or shrimp will kill them. Copper also precipitates on calcareous substrate (coral sand, limestone rock) and leaches back for weeks, making an 'inverted tank' effectively contaminated long after dosing ends. Do not use copper in any system you plan to restock with invertebrates without a sustained carbon and water-change regime first.

Antibiotics and the biological filter: broad-spectrum antibiotics, particularly those targeting gram-negative bacteria (the class that covers most finrot and ulcer pathogens), can suppress nitrifying bacteria enough to cause an ammonia spike. The effect is variable — some products are more filter-friendly than others — but treating in a hospital tank and monitoring ammonia during and after any antibiotic course is the safest approach. This is especially important in a lightly stocked system where nitrifier biomass is modest.

Planted tanks: many standard medications are phytotoxic. Malachite green (common in ich treatments) will stunt or kill sensitive stem plants and mosses. Metronidazole is broadly tolerated by plants but should still be used in hospital tanks where possible. Potassium permanganate harms plants severely. Formalin sharply depletes dissolved oxygen — supplemental aeration is mandatory during any formalin treatment, and the oxygen drop is further compounded at higher temperatures.

Temperature interactions: for ich, raising temperature to the upper tolerable limit of the stock (not beyond what the most sensitive species can sustain) speeds the parasite's life cycle and shortens treatment time. For columnaris, heat has the opposite effect — the bacterium grows faster in warmer water, and a temperature increase intended for ich can convert a manageable columnaris infection into a rapid kill. Never raise temperature without first ruling out columnaris as a co-infection or primary cause.

Resistance: the routine use of antibiotics for conditions where they offer no benefit — viral disease, pure parasitism, non-bacterial finrot secondary to water quality — selects for resistance without providing any therapeutic advantage. Once resistance develops in the biofilm of a chronic aquarium system, it is very difficult to clear. Use antibiotics only when a bacterial cause is strongly indicated, and use a full course rather than stopping early.

When to reach for a microscope or a vet

Much can be read with the naked eye and a test kit, but some of the most serious fish problems are invisible without magnification. A skin or gill scrape examined under a compound microscope at 40–100× is the decisive tool for parasite identification — it takes the question from 'probably flukes' to 'confirmed Dactylogyrus, treat accordingly.' Many serious hobbyists learn this skill; a cover-slipped gill filament or skin scraping in a drop of tank water takes under two minutes to prepare. What you are looking for: ciliates (ich, chilodonella) move with characteristic ciliary beats; monogenean flukes are flat, elongated, and anchor with visible hooks; velvet trophonts are smaller with internal pigment. The absence of any motile organism, combined with intense flashing, redirects attention toward chemical irritants or bacterial gill disease (BGD), which requires culture or histology to confirm.

Reach for a vet or diagnostic laboratory when a fish is clearly sick and the cause is not visible, when an appropriate OTC treatment has not worked after a full course, when several fish are dying in rapid succession, or when the problem appears internal or systemic. An aquatic veterinarian can prescribe prescription-only antibiotics (including some that are fish-specific and less disruptive to nitrifiers than the OTC alternatives), perform anaesthesia-assisted physical examination, and submit tissue for histopathology or culture — the only route to a reliable diagnosis for many internal conditions.

Shotgunning OTC medications — trying one after another without a diagnosis — stresses the fish, degrades water quality, and selects for resistant strains. It is an expensive way to reach the same poor outcome faster.

• Microscope scrape (40–100×): confirms flukes, ich, velvet, chilodonella — distinguishes them by morphology
• Absence of organisms on scrape + intense flashing → consider chemical irritant or bacterial gill disease
• Vet/lab indicated: OTC failure, mass mortality, suspected internal or systemic disease
• Prescription antibiotics exist that are more targeted and less filter-damaging than OTC options
• A diagnosis beats shotgunning medications — which breeds resistance and delays recovery

Isolating a sick fish — and when not to

A dedicated hospital tank — a small, bare, gently filtered vessel kept ready — lets you treat one fish without dosing your whole display, and without medications harming plants, shrimp, or the main filter. Moving a fish there is the right call for an individual problem: an injury, a bite wound, a single ulcer.

There is a catch, though. For a contagious disease that has already been in the main tank for days — ich is the classic example — removing the one symptomatic fish treats the symptom, not the outbreak. The infective stages are already in the shared water; the tank itself is the patient. Match the response to the problem: individual ailment, individual care; contagious outbreak, treat the system.

Honest prognosis: some conditions are rarely winnable. Dropsy in an adult fish, once the pineconing is well established, represents end-stage organ failure — the kidneys or liver are typically beyond recovery. Palliative isolation is appropriate; aggressive medication rarely changes the outcome and may extend suffering. Systemic bacterial septicaemia presenting with haemorrhaging across the body surface also carries a poor prognosis once it is advanced. Knowing when a fish is past saving is part of the skill, and euthanasia (clove oil, properly dosed by a vet or using established fishkeeping guidance) is the responsible choice when recovery is not realistic.

Quarantine: the habit that prevents most of this

Almost every disease that reaches an established aquarium arrives on something new — a fish, sometimes a plant, occasionally shared water or equipment. Quarantine means holding new arrivals separately and observing them for a few weeks before they join your established fish. It lets a hidden infection reveal itself where it can be dealt with cheaply and in isolation, and it gives a stressed, freshly-shipped fish time to recover away from competition.

Four weeks is the practical minimum for a serious quarantine — long enough to catch ich (which can cycle at room temperature in under two weeks), flukes, and most bacterial presentations. Some experienced keepers run six weeks, particularly for wild-caught fish or species known to carry asymptomatic internal parasites (many wild cichlids, most wild plecos). Equipment and nets shared between tanks bypass quarantine entirely; dedicate them per tank or disinfect between uses.

Prophylactic treatment during quarantine — a round of praziquantel for flukes and tapeworms, and possibly metronidazole for internal flagellates, on wild-caught fish — is standard practice in fishrooms running serious breeding programmes. Done in the quarantine tank, it clears subclinical loads before the fish joins a clean population. This is a case where treatment without visible signs is justified because the epidemiology strongly predicts the parasite.

It is the least glamorous and most effective practice in fishkeeping. Aquatic-health authorities make the same point for systems large and small: quarantine and steady husbandry prevent far more loss than treatment ever recovers. The best disease guide is the one you rarely need.