Is my goldfish bullying the other one?

Goldfish have no teeth in their jaws. They do not hold territory, do not guard a nest, and have no weaponry beyond their own body weight. As the behaviour guide puts it, they are physically poorly equipped to fight.

So when one goldfish appears to be bullying another, the useful first move is not to separate them. It is to work out what you are actually watching, because it is usually one of four things and each has a different answer.

What you are probably seeing

What it looks like	Most likely	What to do
Relentless pursuit in spring, white pimples on the chaser	Spawning	Nothing, beyond protecting the female
No chasing, but one fish is getting thin	Food competition	Change how you feed
Constant nudging and crowding in a small tank	Too little space	More water, or fewer fish
One fish hangs back and the others push past it	That fish is ill	Test the water, then look at the fish
Torn fins on a fish that is otherwise fine	Decor, or spawning	Check the decor before blaming a tankmate
A non-goldfish tankmate nipping	A species mismatch	Rehome the tankmate

The row people arrive expecting is not on the list, and that is the point. Sustained goldfish-on-goldfish aggression, of the kind seen in cichlids or bettas, is not a normal part of goldfish keeping.

The spawning chase

A common reason anyone reaches this page, and one that needs no intervention.

What it looks like: one or more fish driving another relentlessly for hours, pushing at her flanks, following her into every corner, refusing to let up. It looks like persecution and it is alarming the first time.

How to confirm it: check the chaser’s gill covers and the leading rays of its pectoral fins for small white pimples. Those are breeding tubercles, and they mean a male in condition — sexing goldfish covers telling them from white spot disease, which is the mistake that sends people to the medicine cabinet. Then look at the chased fish, who will often be noticeably full-bodied.

When it happens: spring, or after a large water change, or whenever the temperature climbs a few degrees. Is my goldfish breeding? covers the whole sequence, and breeding goldfish covers doing it deliberately.

What to do: watch the female rather than the chase. Hours of being driven into surfaces costs her scales and fin edges, so the things to check afterwards are missing scales and torn fins — both usually heal, and both are worth keeping an eye on in case they do not. If she is exhausted and it has gone on for days, a few days apart is reasonable.

What not to do about a spawning chase: treat the tank. There is nothing to treat.

Food competition, which is the one that actually hurts

A damaging version of this problem involves no chasing at all, which is why it goes unnoticed for months.

A faster fish does not need to be aggressive to starve a slower one. It simply gets there first, at every meal, for a year. The tankmates guide makes this argument about mixed species and it applies just as forcefully between goldfish: a comet or a ryukin will shoulder a slow fancy off a pellet without anything that looks like conflict, and the result is a thin fish with poor colour and no obvious cause.

How to spot it: watch a whole feed, from the first pellet to the last, and watch one fish rather than the tank. Count what it actually eats. A fish that takes two mouthfuls while another takes twenty is not competing.

What fixes it, in order:

• Feed at two ends of the tank at once. One fish cannot control two places. This alone solves most cases.
• Use sinking food. A surface feed is a race; food on the bottom spreads out and lingers, which is the argument in sinking versus floating food.
• Feed more often, in smaller amounts, so there are more opportunities rather than one scramble — within the limits set out in how often to feed goldfish.
• Hand feed the slow fish, or use a length of pipe to deliver food to the corner it favours.
• Break the sight lines, so a fast fish cannot see and cover the whole tank at once. Aquascaping for goldfish covers doing that without losing swimming room.

And if none of it works, the mismatch is structural. A single-tail and a fancy are different animals with different speeds, and the answer is two tanks or a rehoming rather than a feeding technique — rehoming a goldfish covers where a fast single-tail can actually go, which is usually somebody’s pond.

Crowding, which manufactures the problem

Goldfish are not known to keep a fixed hierarchy, but fish in a group do some nudging and jostling, especially around food. It resolves in a moment, and it needs somewhere to resolve to.

In a tank that is too small there is nowhere. Every encounter is forced, nobody can withdraw, and jostling that would be invisible in a large tank becomes constant contact in a small one. Add to that the water quality consequences of overstocking — higher ammonia and nitrate, lower oxygen — and you have fish that are both crowded and chemically stressed.

This is not a behavioural problem and cannot be fixed behaviourally. How many goldfish can I keep and goldfish tank size set out the working numbers, and the options if you are over them are fewer fish or more water.

Adding fish to dilute it makes it worse, reliably. More fish means less space each, more waste and more encounters — every one of the drivers, turned up.

The fish that is ill, not bullied

Worth its own section because the two look alike and the treatment is opposite.

A goldfish that is unwell stops holding its place. It hangs back at feeding, rests on the bottom, clamps its fins, and the others swim past and through it. To a watching keeper that reads as exclusion, and the instinct is to rescue it from the group.

The tell is direction. Are the other fish pursuing it, or ignoring it? Active, one-way pursuit is social. Indifference while one fish deteriorates is often illness, and sitting on the bottom, clamped fins and loss of appetite are the pages that work through what it might be.

Test the water before you conclude anything social. A common reason a goldfish behaves strangely is not another goldfish; it is ammonia, nitrite or nitrate, and no amount of rearranging the social group addresses that.

When you have just added a fish

A distinct case with its own timetable, and the one most often misread.

A new goldfish going into an established tank can set off a fortnight or so of sorting out. The residents investigate, follow, nudge and occasionally chase; the newcomer keeps to a corner and eats poorly. To someone watching on day two it looks like the group has rejected it.

Most of that settles by itself. What you are seeing is the group settling around an extra animal, not a rejection, and the useful judgement is made at two weeks rather than at two days.

What makes it go well:

• Quarantine first, always, for reasons that have nothing to do with behaviour — quarantining a new goldfish covers them, and a fish that arrives already settled and eating is a fish better placed to hold its own.
• Rearrange the decor on the day, as many keepers do, so the residents are re-learning the tank at the same time as they are meeting the newcomer. It is common practice rather than something that has been tested.
• Add at a feeding time, with food already in, so the first encounter happens while everyone is busy.
• Add more than one fish at once if you are going to add more than one eventually. Attention divided is attention diluted, and the arithmetic in how many goldfish you can keep still has to hold afterwards.
• Turn the lights down for the first evening.

When to step in: if the newcomer is still not eating after several days, if it is accumulating fin damage rather than healing, or if one resident has singled it out and does nothing else. Choosing and bringing home a goldfish covers the rest of the arrival.

A large fish and a small one

Size differences worry people more than they should, and matter in a way people do not expect.

Goldfish rarely attack smaller goldfish. Predatory behaviour towards another fish of the same species at anything like a comparable size is not typical.

What size does decide is food, and it decides it completely. A 20 cm fish eats faster, reaches further and displaces a 6 cm one without intending anything, so a mixed-size group needs the feeding tactics above rather than supervision.

The genuine limit is mouth size. Goldfish will eat anything that fits, and that includes very young fish of their own kind. Fry and small juveniles do not go in with adults — which is why the grow-out tank in breeding goldfish is a separate tank rather than a corner of the display.

And growth compounds the problem. A small fish that loses every meal grows slowly, which widens the gap, which loses it more meals. Catching this early matters more than the size difference itself.

Chasing in a pond

The same behaviour, harder to see and harder to interrupt.

Ponds show more of it, because ponds spawn. Outdoor fish get the seasonal cue properly, so spring in a pond is weeks of driving rather than a morning of it, and a pond through the year covers what else is happening at the same time.

Intervening is usually impractical, and usually unnecessary. Catching a pond fish is a considerable operation with its own costs.

What helps instead is structure. Dense marginal planting, a deeper zone and somewhere a chased fish can break away all reduce the pressure without anyone being netted — goldfish ponds covers building that in.

What to watch for is a fish that stays hidden for days, one that is visibly thinner at autumn feeding, and damage that is not healing as the water warms.

When the tankmate is not a goldfish

A separate problem with a cleaner answer.

Goldfish are poorly equipped to harass other fish and quite well equipped to be harassed by them. Nippy species take fancy fins as a target of opportunity, and long-finned slow-moving fish are close to ideal prey for anything that nips.

It also runs the other way: goldfish will nip at the gills of a slow-moving tankmate, and will attempt to eat anything small enough to fit in a mouth.

Goldfish tankmates works through which combinations are workable and which are not. Where the answer is a mismatch, rehoming the other species is the fix, and no amount of tank rearrangement substitutes for it.

If you do separate them

Sometimes the right call, particularly for an exhausted female or a fish that is visibly losing condition.

Move the aggressor, not the victim. This is the counterintuitive part and keepers find it the part that works. Removing the chased fish leaves the tank exactly as it was and returns it to the same situation later. Removing the chaser for about a week, then putting it back, means it re-enters as a newcomer to an established group and the group settles around it. The hospital tank page covers the setup, and it is the same bare tank with a sponge filter that every other problem eventually needs.

Keep the separation short. A week is usually enough. Long separations turn a temporary problem into two permanent tanks.

Use a real tank, not a breeder box. A plastic box hung inside the tank is a few hours’ measure at best; a fish confined in one for days is worse off than it was.

Catch them carefully. A chased fish that is then chased around the tank with a net is in worse condition than when you started — handling and netting covers doing it with two nets and no drama.

Fix the cause during the week. Separation buys time; it does not change anything. If the tank was crowded when the fish came out, it will be crowded when it goes back in.

Bringing a fish back

Reintroduction is straightforward if the week was used properly.

Return it at a feed, when everyone is occupied and looking at the food rather than at each other.

Rearrange the decor first, even slightly, as many keepers do. A tank that looks different gives everyone the same unfamiliar start, which is the outcome you want.

Expect a day or two of sorting out, and judge by whether it settles rather than by whether it happens.

And watch the returning fish, not just the one that was being chased. A week alone in a bare tank is its own stress, and a fish that has been out of the group comes back to water it is no longer adapted to if the parameters have drifted apart.

What this looks like when it is working

A goldfish tank in good order is not a tank with no jostling. It is a tank where jostling happens and nothing comes of it: fish that nudge at feeding time and then disperse, a loose pattern around food that shifts week to week, and every fish in the same body condition as every other.

That is the standard worth holding to, and it is mostly bought with water rather than with intervention. Goldfish welfare makes the general version of the argument — most of what looks like a behavioural problem in a goldfish tank is an environmental one wearing a disguise.

The short version

Work out which of the four it is before doing anything. In spring, with white pimples on the chaser, it is spawning and it needs no fix. If one fish is getting thin with no visible conflict, it is food and you change how you feed. If the tank is tight, it is space and nothing else will do. And if the others are ignoring rather than pursuing the fish you are worried about, test the water, because that fish is probably ill.

If you separate them, take out the one doing the chasing, keep it brief, and use the week to change something.