Hammer Coral Losing Heads or Dying? Causes and What to Do

A hammer coral losing heads is almost always an energy problem, not an infection. The tissue at the base of each head carries almost no symbiotic algae, so it is the first thing the colony stops paying for when its energy budget goes negative — which is why you see bare skeleton creeping up from the bottom while the tentacles above still look normal. Whether you can save the rest depends almost entirely on one thing most people never check: whether you own a branching hammer or a wall hammer.

Wait — do you actually have “heads”?

This sounds pedantic. It is the single most useful question in this article. Branching hammers (phaceloid growth) put each polyp on its own stalk. The heads are physically separate structures joined only at the base. Wall hammers (meandroid growth) are one continuous ridge of tissue over one continuous skeleton, with mouths strung along a valley. A wall hammer does not have heads. It has segments of a wall. That distinction is not cosmetic. Swain and colleagues published a skeletal integration index in 2018 (Marine Ecology Progress Series 586) ranking coral growth forms by how physically connected their polyps are. Meandroid forms sit at the integrated end; phaceloid forms sit at the isolated end. When a branching hammer loses a head, one semi-independent module has died. When a wall hammer “loses a head,” a shared sheet of tissue is receding across a shared skeleton, and everything downstream of that edge is connected to the problem. So: if you have a branching hammer and one head is gone but the others are open and fat, that is often survivable and sometimes not even progressive. If you have a wall hammer and bare skeleton is appearing, the clock is running on the whole colony. One naming note, because it affects what you find when you search. Branching hammers and frogspawn are Fimbriaphyllia, not Euphyllia — the genus was resurrected on molecular grounds and a widely-shared 2017 hobby article got the assignment backwards. Plenty of care sheets still repeat it. If a page tells you branching hammers are Euphyllia paraancora, it has not been updated in nearly a decade, which tells you something about the rest of its content.

Why does the base go white while the tentacles still look healthy?

Because the base was never photosynthetic in the first place. The only tissue-level transcriptome of Fimbriaphyllia ancora — Shinzato et al., Communications Biology, 2024 — separated tentacle, mouth and body-wall tissue and found Symbiodiniaceae abundant in the tentacles and oral region and nearly absent from the body wall. The band of tissue wrapping the base of a hammer head is effectively aposymbiotic. It does not make its own food. It is subsidized by the photosynthetic tissue above it. Once you know that, the classic presentation stops being mysterious. A colony under any sustained energy deficit — low light after a fixture change, months without feeding, alkalinity swinging, a stubborn pest — sheds the tissue that costs the most and returns the least. That is the base. So you get white skeleton climbing up from the bottom while the tentacles above stay inflated for days or weeks, and every forum reply tells you the coral “looks fine, must be something else.” It is not fine. You are watching triage.

The practical version: bare skeleton at the base with good tentacle extension is an early warning, not a late one. It is the best window you will get.

Is it brown jelly, or is that just what everyone calls it?

“Brown jelly” is a description of an appearance, not a diagnosis. It is what you see when ciliates and bacteria bloom on tissue that is already dying and dissolving. The jelly is the consequence. Something else opened the door. This matters because treating brown jelly as the disease sends people down a path of dips, antibiotics and turkey-baster blasting while the actual driver — a pest, a burn, a chemical swing, an unnoticed injury — keeps running. Blasting the jelly off and getting a week of apparent improvement, then watching it return, is the signature of treating a symptom. It is also worth flagging that the paper most often cited as the brown jelly reference (Sweet et al., 2013) was working with a white-syndrome presentation on a Euphyllid, not the brown mucoid mass hobbyists mean. The citation gets copied around without anyone reading it. There is remarkably little rigorous work on this specific condition in captive Euphyllids — which is the honest state of things, and worth saying rather than papering over. What we can say from practice: if the jelly is present, remove the coral to a container of tank water, blast the affected tissue off gently with a baster, and get it away from the rest of your Euphyllia. Then go looking for the cause. If you cannot find one, look harder at the two things below.

What actually eats hammer coral tissue?

Two culprits are worth real attention, and one of them is very new. The Euphyllia-eating flatworm. Small, flat, almost perfectly color-matched to the coral, and it sits on the base and the underside of heads — not on the tentacles where you would look. It is still an undescribed species; there is no formal literature on it. Dip a suspect colony in a coral dip and hold the dip water up to a light. If you see them, you have your answer, and the treatment is repeated dips plus manual removal, not water chemistry. Bristle worms inside the skeleton. Julian Sprung raised this in April 2026: worms living inside the hollow branching-hammer skeleton and consuming polyp tissue from below, producing head loss with no visible surface cause. This is a small number of observations from a credible observer and it has not been replicated or published. Treat it as a hypothesis worth checking — not as established fact. But if you have a branching hammer losing heads with clean water, no visible pests and no jelly, looking into the skeleton is now a reasonable step. If you have not dipped this colony, dip it. Our coral dipping guide covers doing it without adding a second injury on top of the first.

Should I frag off the healthy heads?

Sometimes — and the reasoning is better than the usual “cut ahead of the infection” framing. Work by Gavish, Kramarsky-Winter and Kushmaro on Euphyllid tissue loss found that lesions consistently began in the coenosarc — the connecting tissue immediately surrounding a polyp — rather than in the polyp itself. The tearing you see at the base is plausibly the colony attempting to isolate the affected polyp from the rest. In that reading, fragging is not overriding the animal. It is finishing a quarantine the animal has already started, faster and more cleanly than it can manage alone. The counter-evidence deserves equal billing. The only rigorously monitored firebreak trial on wild corals — Neely et al., 2021, cutting trenches ahead of advancing stony coral tissue loss lesions with an angle grinder and sealing with chlorinated epoxy — measured under 20% effectiveness at three months. Different disease, different scale, but it is the best-controlled test of the underlying idea that exists, and it did not work well. Cutting is a bet with a real failure rate, not a rescue. If it is branching: cut low, well below the affected head, with bone cutters. Take the heads that still look good, put them in a separate system or a container away from your other Euphyllia, and do not dip freshly cut tissue. If it is a wall: do not simply saw through live tissue. The trick that works is a ligature — run a rubber band tightly along the line you intend to cut and leave it. Over roughly two weeks the polyps along that line withdraw and the skeleton beneath goes bare. Then you cut through bare skeleton, and you have made an incision without opening a wound across living tissue.

Is my flow wrong?

Probably not in the direction you have been told. The hobby consensus is “hammers want low flow,” and Veron’s habitat descriptions do not support it as a blanket rule. F. ancora, the wall hammer, is recorded from shallow environments exposed to moderate wave action. F. paraancora, the branching hammer, is the one described from water protected from wave action. The generalization is backwards for half the hammers in the hobby. What is real is that a hammer in flow strong enough to keep tentacles pinned flat cannot feed and cannot fully inflate, and a hammer with essentially no flow sits in its own waste. Aim for indirect, variable, moderate movement — tentacles drifting, never streaming. If you are running a wall hammer, do not be afraid to give it more than you would a branching one.

What about light, alkalinity and the rest?

These matter as inputs to the energy budget, which is the whole story of this article.

• Light. Hammers do well in the moderate band, roughly 75–150 PAR. The failure mode is not usually a wrong number, it is a fast change — a new fixture, a raised light, a cleaned lens. If you changed anything about lighting in the six weeks before the head loss started, that is your first suspect. A rented PAR meter answers this in one evening; we rent them.
• Alkalinity. Stability beats the target. A colony held at 7.8 dKB indefinitely does better than one bounced between 7.5 and 9.5 twice a week. Swings burn energy that the base tissue is the first to lose access to.
• Feeding. Because the body wall does not photosynthesize, a hammer that is never fed is running its base entirely on surplus from above. Small meaty foods once or twice a week meaningfully change what that colony can afford to keep.
• Neighbors. Hammers have long sweeper tentacles and get hit by them too. A recurring burn at one edge of a colony looks a lot like disease. Check what is within reach at night.

How long until I know if it stopped?

Photograph the colony from the same angle, in the same light, once a week. Not daily — daily comparison is noise and it will make you intervene when you should not. A receding edge that is in the same place across two weekly photos has stopped. A colony that has stopped receding and is inflating normally is done; it will not regrow tissue over bare skeleton quickly, and often never fully, but the skeleton will get colonized by coralline and the colony will keep growing from the living edge. If it is still moving at three weeks despite good parameters, no visible pests and a dip, the realistic call is to save what you can rather than keep the colony whole.

The short version

• Identify branching versus wall first. It changes the diagnosis and the fix.
• Bare base with healthy tentacles is an energy-budget failure showing up in the tissue that has no symbionts — it is early, act now.
• Brown jelly is a symptom. Find what opened the door.
• Dip and check for the undescribed Euphyllia flatworm before touching your dosing.
• Fragging assists a quarantine the coral already started, but the best-controlled test of firebreaks came in under 20% effective. It is a bet.
• Wall hammers: ligature first, cut bare skeleton second.
• Weekly photos, same angle. Not daily.

There is no peer-reviewed literature specifically on captive hammer coral head loss. Everything above is either drawn from adjacent published work on Fimbriaphyllia physiology and coral lesion progression, or from what we see in our own systems. Anyone telling you there is a settled answer is not reading the same sources.

Related. A hammer losing heads shortly after purchase is often carrying injuries from the trip and the vat before it. Why the cheapest coral costs the most covers what happens upstream and how to judge a seller before you buy.

The Underwater Gardener grows aquacultured corals in Key West, Florida (FL Aquaculture Certificate #AQ0254072). Our corals are raised under stable reef lighting in our own systems, so they arrive already adapted to a well-run tank. Corals ship UPS Next Day Air every Tuesday and Wednesday.