The Ceriths, and the Shell Shape Three Families Invented Separately

The shape three unrelated families arrived at

Pick up a small shell far taller than it is wide, with many whorls winding to a point, and you have narrowed things down much less than it feels like you have. That silhouette belongs to several groups of sea snails that are not closely related and do not live alike.

Three of them cause most of the trouble. Ceriths, family Cerithiidae, crawl around scraping film off surfaces. Turret shells, family Turritellidae, mostly stay put and strain food out of the water. Augers, family Terebridae, are carnivores, and many of them hunt with venom.

Here is the part that catches people: two of those three are close relatives, and it is not the pair you would guess. The auger, the shell most often mistaken for a cerith, belongs to a different order of snails entirely. This page is about that gap, and about what a cerith actually is once you stop defining it by the things it resembles.

What actually makes a shell a cerith

The family is Cerithiidae J. Fleming, 1822, AphiaID 128, an accepted name in the World Register of Marine Species and in MolluscaBase, its mollusk component. Fleming erected it in The Philosophy of Zoology in 1822, spelled "Cerithiadae" at the time, and the record was last reviewed in MolluscaBase by Philippe Bouchet in 2025.

Above it sits the superfamily Cerithioidea, AphiaID 14769, within the subclass Caenogastropoda. Between those two ranks there is a deliberate gap. WoRMS assigns ceriths to no named order, and the entry reads "Caenogastropoda incertae sedis," Latin for "of uncertain position." That is an honest placeholder, not an oversight, and many caenogastropod families sit in the same holding pattern. The family is flagged both marine and brackish, so it straddles full seawater and estuarine water, and WoRMS lists 57 direct children under it.

For scale, Strong and Bouchet in 2018 called Cerithiidae one of 19 then-accepted families of Cerithioidea, with 219 species considered valid. Carry the date: the same lead author's 2011 work used 17 cerithioidean families, so counts here are not stable enough to quote bare. The family runs worldwide from tropical to temperate waters, most often shallow, a few species reaching bathyal depths.

The hard parts: four characters, a door, and a first shell

The most useful published description comes from Richard Houbrick's 1992 monograph on the genus Cerithium, a 211-page volume of Smithsonian Contributions to Zoology. His diagnosis: morphologically highly variable species characterized by many-whorled, elongate shells having ovate apertures, slightly reflected anterior canals and sculpture.

• Many-whorled and elongate. The high spire itself, and the source of every problem on this page.
• An ovate aperture. Oval, not round.
• A slightly reflected anterior canal. A short spout at the front, where the siphon sat: present, short, turned back on itself. This one carries the most information.
• Sculpture. Beads, spiral cords, axial ribs.

One caveat travels with all of it. Houbrick was diagnosing the genus Cerithium, not the whole family, and texture does not survive as a family rule: Strong and Bouchet's 2018 description of the cerithiid Limatium calls its shells smooth and polished. Varices, the thickened ridges marking where an outer lip used to sit, are widely called a cerith trait, and one genus is named Varicopeza for them, but we could not confirm a family-level statement in a primary source. Probable, not checked.

Two more hard parts. The operculum, the door the animal shuts behind itself, is paucispiral where it has been examined: few coils, with a large nucleus set off center. The protoconch, the first shell formed before the larva settles, is multispiral in the genus where it was measured, which means a feeding larva that drifted in the plankton. Both come from genus diagnoses, not a family diagnosis.

The animal inside is a grazer, and there are a lot of them

What holds this family together is not the shell. It is a job.

Strong and Bouchet put it in one sentence in 2018: as far as is known, most species are microherbivorous grazers, and they usually occur in large populations.

A microherbivorous grazer eats the film: not seaweed and not other animals, but the thin skin of algae, diatoms, bacteria and organic debris coating every submerged surface and every seagrass blade. The tool is the radula, a ribbon of teeth drawn over a surface like a rasp, and the published description of the central tooth in Limatium gives it a cutting edge with three sharply pointed cusps. A three-point scraper.

The radula is not uniform across the family, and the same authors say so: of that genus they write that its radula does not readily suggest an affinity to any of the cerithiid subfamilies.

"Usually occur in large populations" is the underrated half. The same paper flags one cerithiid as remarkable precisely because it is rare, which the authors call unusual for the family. These are abundance animals.

Published work on the grazing itself is thinner than you would hope: Van Montfrans, Orth and Vay studied grazing by the grass cerith on eelgrass periphyton in Aquatic Botany in 1982. We could read the citation and not the paper, so we attach no numbers to it, and it is eelgrass work rather than Gulf work.

Two of the three lookalikes are cousins. The third is from another order.

Here is the taxonomy of the three, checked in the World Register of Marine Species.

Cerithiidae and Turritellidae sit in the same superfamily. Terebridae is not merely in a different superfamily, it is in a different order, and its superfamily Conoidea also holds the cone snails and the turrids.

Strong and colleagues published a phylogeny of Cerithioidea in the Zoological Journal of the Linnean Society in 2011, combining 151 morphological characters with 16S and 28S ribosomal RNA data across 47 cerithioideans from 17 families plus 9 outgroups, and Turritellidae is analyzed as a cerithioidean throughout. Their result deserves quoting carefully: Batillariidae, Cerithiidae, Pachychilidae, Pleuroceridae, Semisulcospiridae, Thiaridae and Turritellidae came out monophyletic in most but not all topologies. "Most but not all" is doing real work there, and we come back to it below.

So the shell most reliably called a cerith by mistake belongs to an animal from another order, and the animal that genuinely is a cerith's relative is the one hardly anybody connects to it. Resemblance and relationship point in opposite directions.

The turret shell filters, and has nearly given up its rasp

The turret shell is the strange one: the cousin, and the one that changed most.

Warren Allmon's 2011 status report on turritelline gastropods in Malacologia describes the living animals as mainly sessile, semi-infaunal suspension feeders, though he notes they can occasionally be more active, crawling on the surface. He also calls them among the most abundant and widespread marine gastropod clades of the past 130 million years.

Alastair Graham worked out the mechanism in 1938, in the Proceedings of the Zoological Society of London, on the European Turritella communis. Water enters on the left of the mantle cavity and leaves on the right, the intake guarded by a complex array of tentacles preventing the ingress of large particles. Particles are collected on the gill filaments, moved into a deep groove along the floor of the mantle cavity, built there into a ropy string with mucus, and passed slowly forwards to the mouth, into which the string is raked.

Now the detail that makes this page. In that animal's buccal cavity, Graham found a reduced odontophore: the apparatus that carries the radula and works it against a surface. The turret shell has all but put the rasp down.

Two snails, nearly one outline. One is a mobile scraper with a three-cusped tooth. The other is a half-buried filter feeder with a shrunken version of the same machinery. And those two are the close relatives.

The auger hunts, and many but not all of them carry venom

Now the outsider. Terebridae sits in Conoidea alongside the cone snails, and the family business there is predation.

The venom is the famous part and the part most often overstated. Holford and colleagues, in PLoS ONE in 2009, found that some terebrid groups have secondarily lost their venom apparatus: species in the Acus clade do not have one, while those in the Terebra clade do. Modica and colleagues went further in Systematic Biology in 2019: terebrids have lost it several times, the losses sit in phylogenetically clustered terminal taxa, and reversal is extremely unlikely.

Gorson and colleagues then ran the first molecular diet study of the family, in Toxins in 2021, metabarcoding gut contents from 71 specimens across 17 species. Presence or absence of a venom gland proved strongly correlated with dietary breadth, and the direction is the surprise: the species without a venom gland showed the greater dietary diversity.

So, the safe sentence. Augers are carnivores, many carry a venom gland and hypodermic radular teeth, and several lineages have lost the venom apparatus entirely. Anyone who says flatly that all augers are venomous is repeating something the literature has ruled out.

Here augers are a contrast, and the contrast is the point: the shell nearest a cerith in appearance belongs to the animal furthest from it in behavior. The page on the cone snails covers how a conoidean delivers venom.

Why the same shape keeps getting reinvented

So why does it keep happening?

First, what we will not do. It would be easy to hand each animal a tidy functional reason for its high spire. We could not source one, so we are not writing one. Plausible mechanics dressed up as findings is how bad shell writing gets made.

What can honestly be said is more useful. Shell shape is generated by a very small number of coiling parameters, and the page on how a shell is actually built covers David Raup's 1966 demonstration that three or four of them reproduce essentially every shell shape that exists. A tall, narrow, many-whorled cone sits in a small, easily reached corner of that space, and shapes that are cheap to reach get reached repeatedly.

The lesson is about limits rather than tricks. Outline is the least informative thing about a shell in this group, precisely because outline is what converged. The information sits in the details that did not: the aperture, the anterior canal, how the operculum is built, how the sculpture is organized. And much of the real story is not in the shell at all: a radula, an odontophore, a venom gland, none of which you will ever meet in an empty shell. The Florida identification guide covers what a shell can and cannot tell you.

The fossil record puts 165 million years between them

The convergence also happened at wildly different times, which is the clearest evidence that it is convergence and not shared inheritance. Queried against the Paleobiology Database, which aggregates first appearances from published occurrence records and taxonomic opinions, they line up like this.

Ceriths are a Triassic family. Augers do not show up until after the dinosaurs are gone. Roughly 165 million years separate the two, which rules out a common ancestor handing the shape down.

Carry the caveat, because it is real. These dates rest partly on fragmentary and sometimes misassigned material, and the deepest of them, the Carboniferous superfamily date, is the softest. What survives is the relative order, and that is no close call: not a few million years apart, but most of the age of the dinosaurs.

Every one of these families answers to somebody else's name

The common names in this corner of the shelf really are working against you, and that is documented rather than imagined.

WoRMS lists "auger shells" as an English vernacular for two different families, Turritellidae and Terebridae. A 1981 entry in the Journal of Molluscan Studies is titled, in part, "TURRITELLIDAE ... TURRITELLA COMMUNIS Risso, 1826, auger or screw shell." Cerithiidae has the same problem from the other side. Its listed English vernaculars include horn shells and hornsnails, which are also the standard English names for Potamididae. WoRMS carries them on Cerithiidae for a historical reason: Potamidinae used to sit inside Cerithiidae as a subfamily, and that record is still there, marked superseded in rank and pointing out to Potamididae.

Which brings us to three names that look like ceriths and are not.

• Cerithideopsis scalariformis (Say, 1825), AphiaID 758592, is in Potamididae. Older guides list it as Cerithidea scalariformis and file it with the ceriths. Much of the western Atlantic's snail-and-trematode literature is on this genus, and it is not cerith literature.
• The "West Indian false cerith" is Lampanella minima (Gmelin, 1791), AphiaID 446438, in Batillariidae.
• Cerithiopsis, AphiaID 137764, is in Cerithiopsidae, superfamily Triphoroidea.

The response is not to memorize traps. It is to lead with the binomial. A common name tells you what somebody once believed the animal was related to; the AphiaID tells you what the current authority holds.

The ceriths recorded in Southwest Florida

The list below came from occurrence data, not a field guide. An OBIS checklist query for Cerithiidae inside a Southwest Florida box returns exactly these six and no others, each then checked accepted in WoRMS.

Record counts measure collecting effort as much as abundance, so "most recorded" is not "most common."

Two name notes. Bittium varium is a superseded combination; the accepted name for the grass cerith is Bittiolum varium, out of Houbrick's 1993 generic review of the Bittiinae in Malacologia. But the genus Bittium Leach, 1847 is still valid and still holds species: a species moved genus, the genus was not sunk, and reversing that is the easiest mistake here. And Cerithium variabile C. B. Adams, 1845, used in older sources for Cerithium lutosum, is invalid because it was preoccupied, not Deshayes, 1833: a junior homonym, never available to Adams at all.

One word on live animals, since ceriths are small, abundant and often still occupied. Our rule aboard the boat is our own rule and it is short: empty shells only. An occupied cerith is easy to overlook in shell hash, so look into the aperture first. The page on what is legal to take covers the law under it.

A disputed name, and a family that cannot diagnose its own halves

Ask which cerith is "the dark cerith" and you get two confident, incompatible answers. WoRMS attaches both "Florida cerith" and "dark cerith" to Cerithium atratum, and calls Cerithium lutosum the "dwarf Atlantic cerith." United States field guides mostly do the opposite. We found no authority reconciling the two, so name the binomial and the disagreement stops mattering.

The deeper version sits inside the family. Cerithiidae has two main subfamilies, Cerithiinae and Bittiinae, both accepted in WoRMS. Here is Strong and Bouchet in 2018 on telling them apart: there is presently no known diagnostic feature of shell morphology or internal anatomy that allows unambiguous placement in Cerithiinae versus Bittiinae, with internal features of the midgut the only possible exception, and that requiring further study to confirm.

Read that again. No known character, in shell or anatomy, reliably separates the family's own two halves. The same paper kills a shortcut people reach for: the notion that bittiines can be distinguished by small adult size is, in their words, a frequent misperception.

Two further gaps stay open. How many subfamilies the family has is unsettled: WoRMS and MolluscaBase recognize the two above, while Strong and Bouchet in 2018 list three, adding Argyropezinae, for which WoRMS holds no record. And the 2011 phylogeny recovered Cerithiidae as monophyletic in most but not all topologies, so the accurate statement is that the family usually comes out as one natural group, not that it is one.

The most famous cerith ever found was not a cerith

One last story, a two-century-old version of the mistake this page has been about. In 1804 Lamarck described an enormous fossil gastropod from the Eocene of the Paris Basin and named it Cerithium giganteum, the giant cerith. It became one of the best-known fossil snails in Europe.

It is not a cerith. It is Campanile giganteum, in a different family, Campanilidae, carried in the Paleobiology Database as an extinct taxon of roughly 56 to 37.7 million years ago. We are deliberately not giving you a maximum shell length: figures circulate widely, and we could not trace one to a scientific publication. The genus did not vanish with it. WoRMS holds exactly one accepted living species of Campanile: Campanile symbolicum Iredale, 1917, AphiaID 527560, from Western Australia, which Houbrick described in 1984 as an Australian relict marine snail.

The most famous cerith ever found was never a cerith, and it took most of two centuries to sort out. If a shell that large could be filed under the wrong family on the strength of its shape, the small turreted thing in your palm has earned the same humility. Keep the binomial and the authority you took it from, and let the outline be the last thing you trust. The rest of the Shell Monographs take the families one at a time.