What Does a Sea Sponge Eat? — Bacteria, Plankton, Dissolved Organics and Sometimes Shrimp
Most sea sponges are filter feeders that eat bacteria, tiny phytoplankton and other particles smaller than a few thousandths of a millimetre, along with dissolved organic matter they absorb straight from seawater. Many also get food from symbiotic microbes, including photosynthetic cyanobacteria. A 1 kg sponge can pump up to about 24,000 litres of water a day. A small group of deep-sea carnivorous sponges abandon filtering and trap tiny crustaceans with hooked filaments.
Water enters through thousands of tiny pores (ostia), flows through canals into chambers lined with collar cells (choanocytes), and leaves through larger openings called oscula. Each choanocyte has a whip-like flagellum that drives the current and a collar of microvilli that traps food particles, which the cell then engulfs. Larger particles are captured by other cells lining the canals. Digestion happens inside individual cells — sponges have no gut.
What exactly do sponges eat?

Mostly food too small for most other animals to use: bacteria, the tiniest plankton, fine detritus and dissolved organic matter.
A sea sponge is an animal without a mouth, a stomach, muscles or nerves, yet it feeds continuously for its whole life. It does so by pumping seawater through its body and removing whatever is edible from that water before sending it back out. The food it targets sits at the very bottom of the marine food web. Most of the particles a sponge eats are smaller than a few thousandths of a millimetre, the size range of free-living bacteria and of picoplankton, the smallest single-celled algae and cyanobacteria. Many sponges also absorb organic molecules dissolved in the water, and for some species that dissolved food is the main part of the diet.
This makes sponges unusual among reef and seabed animals. Corals, clams, tube worms and barnacles also filter the water, but they mostly target larger plankton. Sponges are specialists in the smallest size classes, which are by far the most abundant particles in the sea. They are also among the oldest animal lineages on Earth: sponges are possibly the sister group of all other animals, and their way of feeding is close to what the earliest animals may have done.
| Food | Size | Notes |
|---|---|---|
| Bacteria | below ~0.5–2 µm | Main food for many sponges; caught by collar cells |
| Picoplankton (tiny algae, cyanobacteria) | <2–3 µm | Caught by collar cells and canal cells |
| Dissolved organic matter (DOM) | Molecules | Absorbed by sponge cells and symbionts; dominant in some species |
| Larger particles and detritus | 0.5–50 µm | Trapped in the inlet canals and engulfed there |
| Viruses | Nanometres | Removed from the water in experiments |
| Small crustaceans | Under ~1 mm | Only carnivorous sponges |

How does a sponge sort its food by size?
The water-canal system works like a series of sieves, and different cells deal with different sizes of food. According to the standard account of sponge physiology, particles larger than about 50 micrometres are too big to enter the inlet pores, and the flat skin cells on the outside of the sponge, the pinacocytes, engulf them where they land. Particles between about 0.5 and 50 micrometres pass into the inlet canals, which narrow inwards, and are trapped there; pinacocytes or wandering amoeba-like cells called archaeocytes, which push part of their bodies into the canals, then engulf them.
The smallest particles, bacteria-sized and below about half a micrometre, travel deepest, into chambers lined with collar cells, or choanocytes. Each collar cell has a single beating flagellum surrounded by a ring of tiny finger-like projections. The flagellum drives the current, and the collar acts as a filter: water passes between the projections while particles stick to them and are taken into the cell. Because the smallest particles are so much more abundant than larger ones, collar cells typically capture around 80 percent of a sponge’s food supply.
All digestion happens inside individual cells. A particle is wrapped in a membrane bubble, broken down by enzymes, and the nutrients are packed into vesicles that archaeocytes carry through the jelly-like middle layer, the mesohyl, to cells that do not feed directly. In at least one species, internal fibres act as tracks along which these nutrient-carrying cells travel. The same wandering cells clear mineral grains that threaten to block the inlets and dump them into the outgoing current, or in some species build them into the skeleton.
Why does water slow down inside a sponge?
Collar cells can only catch particles if the water passes them slowly, and the architecture of a sponge is built to make that happen. In the common leuconoid body plan, the interior is packed with thousands or millions of tiny chambers. Their combined cross-section is far larger than that of the inlet and outlet canals, so the same volume of water spreads out and slows down in the chambers, then speeds up again as it gathers into the exit canals.
A classic example is the small calcareous sponge Leuconia, about 10 cm tall and 1 cm wide. Water enters each of its more than 80,000 inlet canals at about 6 cm per minute. Inside, it is divided among more than 2 million flagellated chambers and slows to about 3.6 cm per hour, giving the collar cells time to work. It then leaves through a single osculum at about 8.5 cm per second, a jet strong enough to carry waste well away from the sponge so that it does not refilter its own used water.
That design also explains why sponge shape matters. Simple tube-shaped asconoid sponges seldom exceed a millimetre in diameter, because the pumping surface cannot keep up with the volume of tissue to feed. Sponges with pleated walls, the syconoid plan, reach a few centimetres. Only the leuconoid plan, with its network of chambers, allows sponges to grow larger than a metre, and all freshwater sponges and most shallow-water marine sponges use it.
How much water does a sponge filter?
A great deal. Classic measurements, summarised in a major 2007 review of sponge microbiology, suggest that a sponge weighing 1 kg can pump up to about 24,000 litres of water a day. Large reef species are especially impressive. The giant barrel sponge of the Caribbean, which can be big enough for a diver to sit in, pulls water in through its sides and expels it through the osculum at the top, removing bacteria and the tiny cyanobacteria known as prochlorophytes and absorbing dissolved organic compounds directly from the water as it passes.
Sponges are efficient as well as powerful. Glass sponges were once thought to live on dissolved nutrients and to be very sensitive to silt, but a 2007 study found that they extract bacteria and other microbes from the water with about 79 percent efficiency, and can even process suspended sediment grains to get at the microbes attached to them. Because a sponge’s pumping is so vigorous, populations of sponges can clear a large share of the water above a reef or seabed each day, which is why researchers describe them as a key link between the open water and the seabed.
Pumping is not constant. Many sponges can close their inlets and outlets, and some contract their whole bodies. Glass sponges can conduct electrical impulses across their bodies and shut down pumping quickly when sediment or a disturbance reaches them. Others gradually build up waste in their inlets, which leads to one of the stranger recent discoveries about sponge feeding.
Do sponges really sneeze?
They do, in a slow-motion way. In 2022 a team led by Niklas Kornder and Jasper de Goeij at the University of Amsterdam reported in Current Biology that the Caribbean tube sponge Aplysina archeri and an Indo-Pacific sponge of the genus Chelonaplysilla release mucus from their inlet pores. The mucus collects particles the sponge cannot use and accumulates on the surface; from time to time the sponge contracts and pushes the waste-laden mucus off into the water. A single sneeze takes about half an hour.
The researchers suggested that most, if not all, sponges probably do something similar, although only two were studied. The mucus is not wasted. Fish and other reef animals eat it, which means the sponge sneeze is another route by which material a sponge has filtered from the water becomes food for the rest of the reef.
How do sponges feed on dissolved organic matter?
Seawater contains a large pool of dissolved organic matter: sugars, amino acids and other molecules released by algae, corals and microbes. Most animals cannot use it directly, because it is not in particles they can catch. Many sponges can. Research on Caribbean reef sponges, particularly encrusting species that live in the dark cavities inside the reef framework, has shown that dissolved organic carbon can make up the large majority of their diet.
Studies by Jasper de Goeij and colleagues on the cavity sponge Halisarca caerulea in the Dutch Caribbean found that dissolved organic carbon made up about 90 percent of its diet, and that the sponge consumed the equivalent of about half its own body weight each day. Yet it hardly grew. The answer, published in 2009, was that its collar cells divide extraordinarily fast, with a cell cycle of about 5.4 hours, and the sponge sheds old cells as fine detritus instead of getting bigger.
That shed material is the basis of the “sponge loop”, set out by de Goeij and colleagues in Science in 2013. Corals and algae release dissolved organic matter; sponges take it up; sponges shed detrital particles; and snails, worms, brittle stars and other detritus feeders eat that detritus and pass the energy on up the food web. Coral mucus is an important source, since most of it, an estimated 56 to 80 percent, dissolves in the water. The sponge loop helps explain a long-standing puzzle known as Darwin’s paradox: how coral reefs maintain such high productivity in clear, nutrient-poor tropical water.

Do sponges get food from symbiotic microbes?

Yes, and in many species the microbes are as important as the sponge’s own cells. Sponges have no functional mucus barrier, so their tissue is colonised by microbial symbionts, which can make up a large fraction of the animal: estimates range up to about 35 percent of biomass, and up to 40 to 50 percent of wet mass in some species. These microbes belong to more than 50 bacterial and archaeal groups, including ones found almost nowhere else, and they recycle nitrogen, fix carbon and produce chemicals that may defend the sponge.
Photosynthetic partners are the clearest case of microbes feeding their host. Many shallow-water marine sponges host cyanobacteria, and some host dinoflagellates; freshwater sponges often carry green algae inside their cells. Symbiotic cyanobacteria can make up about a third of the living tissue in some sponges, and some species get 48 to 80 percent of their energy from these partners. Such sponges are most common in clear, food-poor water and often grow in flat, leafy shapes that catch more light. A 2008 study even reported that silica spicules can conduct light into the sponge’s interior, where the symbionts live.
Not every symbiosis is a meal ticket, however. The giant barrel sponge’s cyanobacteria give it its colour and the sponge goes through periodic, harmless bleaching episodes, but it does not appear to depend on them for food. In many other sponges, by contrast, the microbes do consume dissolved compounds that the sponge’s own cells could not process, so “what a sponge eats” is really what the sponge and its microbial partners eat together.
Do sponges eat viruses?
They remove them from the water, at least. Viruses are the most abundant biological particles in the sea, and in 2020 Dutch researchers led by Jennifer Welsh tested whether non-host animals could take them out of seawater. In laboratory experiments published in Scientific Reports, sponges reduced the number of viruses by up to 94 percent within three hours, the best result among the animals tested; Japanese oysters, by comparison, removed about 12 percent. It is not yet known how much nutrition sponges actually gain from viruses, but the study suggests that sponge beds influence how many viruses are present in coastal water.
Are there carnivorous sponges?

Yes. A small group of sponges, mostly in deep, food-poor water, has given up filtering and catches small animals.
Carnivory in sponges was confirmed in 1995, when Jean Vacelet and Nicole Boury-Esnault described in Nature a sponge living in Mediterranean caves at depths of about 17 to 23 metres. Instead of pumping water, it catches crustaceans less than a millimetre long by entangling them with fine filaments, slowly envelops them with more filaments over several days, digests them, and returns to its normal shape. There is no evidence that carnivorous sponges use venom.
About 137 carnivorous sponge species are now known. Most belong to the family Cladorhizidae, with a few in the Guitarridae and Esperiopsidae. Most live in deep water, and one has been found at about 8,840 metres. Many have lost the water canals and collar cells entirely and rely on hooked spicules or sticky threads to snag prey. The genus Chondrocladia is an exception: it keeps a modified water system and uses it to inflate balloon-like capture structures. The best known example is the harp sponge, Chondrocladia lyra, described in 2012 by researchers from the Monterey Bay Aquarium Research Institute from depths of about 3,300 to 3,500 metres off California, whose candelabra-like branches strain passing prey from the current.
At least one carnivorous sponge living near hydrothermal vents has taken another route, hosting methane-consuming bacteria and digesting some of them. Deep-sea exploration keeps adding new species, and researchers expect many more carnivorous sponges to be found.
Who benefits from what sponges eat?
Because sponges convert food that most animals cannot use into tissue, mucus and detritus, they support a surprising number of other species. Shrimps of the genus Synalpheus live inside sponges and feed on particles that collect in them; as many as 16,000 individuals have been counted in a single loggerhead sponge. On coral reefs, angelfish, some parrotfish, hawksbill turtles and nudibranchs eat sponges directly, and the detritus and mucus sponges release feed snails, worms and brittle stars. Large sponge grounds in cold and deep water, such as the glass sponge reefs off British Columbia, form habitat for fish and invertebrates in the same way.
Sponges are also sensitive to what is in the water. Heavy sediment clogs their pores, which is why most sponges live in quiet, clear water and why dredging, trawling and run-off can kill sponge beds. Their filtering also concentrates pollutants, so sponges are sometimes used to monitor water quality.
How do you feed sponges in an aquarium?
Indirectly. Sponges need fine particulate and dissolved food, which a mature reef tank supplies through fish waste, bacteria and phytoplankton.
Sponges tend to starve in very “clean” tanks with heavy protein skimming and fine mechanical filtration, because those remove exactly the bacteria and dissolved organics that sponges eat. Hobbyists who keep them successfully usually run mature systems with a healthy population of bacteria and add live phytoplankton or very fine particulate foods. Placing sponges in shaded areas with gentle, steady flow helps, as does avoiding nutrient levels so low that the tank becomes a “marine desert”.
Never lift a sponge out of the water. Air trapped in the canals cannot easily be expelled and the tissue around it dies, which often kills the whole sponge or causes it to decay and foul the tank. Photosynthetic sponges need good light; most others prefer shade, so read up on the species before buying. See sponges in a reef tank and phytoplankton for reef tanks.

What do sea sponges eat?
Bacteria, the tiniest plankton, fine detritus and dissolved organic matter filtered or absorbed from the water. A few carnivorous sponges eat small crustaceans.
How do sponges eat without a mouth?
They pump water through pores and canals; collar cells and other cells trap particles and digest them inside individual cells.
How much water does a sponge filter?
Up to about 24,000 litres a day for a 1 kg sponge, according to classic measurements.
Do sponges eat algae?
They eat microscopic algae and cyanobacteria, not seaweed. Some host photosynthetic microbes that supply them with food.
What is the sponge loop?
Sponges absorb dissolved organic matter and shed cells as detritus that other reef animals eat, recycling nutrients on reefs.
Are there meat-eating sponges?
Yes. About 137 carnivorous species are known, mostly in deep water; they trap small crustaceans with filaments or hooked spicules.
Do sponges sneeze?
Yes. Some sponges expel waste-laden mucus from their inlets by slowly contracting; one sneeze takes about half an hour.
- Taylor, M. W. et al. (2007). Sponge-associated microorganisms. Microbiol. Mol. Biol. Rev. 71: 295–347 (pumping rates, after Vogel 1977).
- Vacelet, J. & Boury-Esnault, N. (1995). Carnivorous sponges. Nature 373: 333–335.
- de Goeij, J. M. et al. (2009). Cell kinetics of the marine sponge Halisarca caerulea reveal rapid cell turnover and shedding. J. Exp. Biol. 212: 3892–3900.
- de Goeij, J. M. et al. (2013). Surviving in a marine desert: the sponge loop retains resources within coral reefs. Science 342: 108–110.
- Kornder, N. A. et al. (2022). Sponges sneeze mucus to shed particulate waste from their seawater inlet pores. Current Biology.
- Welsh, J. E. et al. (2020). Marine virus predation by non-host organisms. Scientific Reports 10: 5221.
- Lee, W. L. et al. (2012). An extraordinary new carnivorous sponge, Chondrocladia lyra. Zootaxa 3608: 301–323.
- Wikipedia: Sponge; Giant barrel sponge; Hexactinellid (anatomy, feeding and symbiont figures).
Blane Perun has been exploring Earth’s oceans and marine life for over 25 years. Founder of TheSea.Org in 1999, underwater photographer, coral aquaculture pioneer, and explorer of 80+ countries.
About Blane Perun → More on Sponges →