Fortey and Owens (1999) conducted
a review of trilobite feeding habits in which several patterns were highlighted.
These are discussed below, and many of the figures are adaptations of those
accompanying their article:
Olenoides sp. (Corynexochida) This formidable looking trilobite was probably an active predator of benthic invertebrates such as worms.
We know about the limbs of Olenoides and other trilobites thanks to the remarkable preservation of the Burgess Shale specimens, and those of other Konservat-Lagerstatten (very well conserved fossil deposits)
The majority of early trilobites are thought to have been predators of benthic invertebrates, such as worms, and Cambrian trilobites such as Olenoides (far left) often bore expanded and spiny gnathobases (labeled "Gn" in the examples at left). Fossilized trilobite trails sometimes stop when they intersect worm burrows (suggesting that the trilobite was hunting for worms, and stopped to eat when it found one in its burrow). Presumably the worm was extracted, subdued and crushed or torn apart with the leg spines and strong gnathobases, then passed forward between the legs to the anterior mouth, where last processing was done against the hypostome platform before ingestion. In crustaceans and insects, all of these functions are served by specialized anterior mouthparts on the head of the animal for processing food before ingestion. However, in trilobites, most of the processing occurred in the longitudinal medial groove between the limbs, with their repeated pairs of gnathobases, meaning that the "mouthparts" of a trilobite occupied the length of its underbody, rather than being primarily anterior. Another piece of inference on the predatory nature of early trilobites can be gained from looking at the relatives of trilobites. The sister-taxa of trilobites, such a Naraoids, also included predators (see the fang-bearing gnathobase of Naraoia compacta above, for example). Predatory trilobites, argued Fortey and Owens, would need to have conterminant hypostomes (firmly attached to the frontal doublure), essentially stabilizing the hypostome against the cephalic exoskeleton for aid in processing prey. It is interesting to note that there is a considerable variation in the size and shape of conterminant hypostoma, suggesting there might have been a great deal of trophic specialization. Below are a few examples of this variation:
©2000 by S. M. Gon III created using Macromedia Freehand
They also suggested that species with inflated glabellae (such as Phacopina, Proetoidea, etc.) might be considered predatory, with the large glabella housing a sizable digestive chamber for initial processing of large chunks of prey (vs bits of detritus). As a final note, they suggested that larger species of trilobites were very likely predatory, including many species of Asaphids, Phacopids, and Redlichiids.
Finally, since there are few depictions of trilobites eating anything, here is my reconstruction of a large Olenoides serratus, subduing a small Ottoia (a priapulid worm) that it has just pulled from its shallow burrow. Its spiny limbs pin the hapless worm to its ventral midline, where its large gnathobases stab and tear at the worm's tough outer epidermis. Once subdued, the gnathobases will tear the worm apart and feed chunks into the mouth.
Although it is argued that trilobites with conterminant hypostomes were probably primarily predatory, a large proportion of trilobites (such as Modocia at right) have natant hypostomes, which Fortey and Owens suggest indicate a shift away from predation and into particle feeding, which includes scavenging for bits of benthic detritus (as the group of olenids below might be doing), or perhaps grazing on beds of algae. Trilobites are not typically depicted doing much of anything except lounging about alone or in groups!
The majority of Ptychopariid trilobites would fall into the category of particle feeders, suggesting that this was a prominent trophic niche for trilobites in the Cambrian and Ordovician. There were many species that assumed the so-called "generalized ptychoparioid form" in the Cambrian and lower Ordovician, attesting to its success. The consistency of hypostome shape, size and form among natant trilobites suggests that particle feeding was a rather generalized habit that did not require much specialization of mouthparts. Indeed, through a long history, natant hypostomes tended to maintain a much more consistent form than conterminant species. Below are some examples of natant hypostomes:
Deposits that bear extremely large numbers of individuals of ptychopariids such as Elrathia also argue that these trilobites were near the base of the food chain (e.g., herbivores), supporting much larger population numbers than less-common predatory species.
Modocia (Ptychopariida :
the successful generalized
Wujiajiania sutherlandi (Ptychopariida : Olenina) lived in the dark, anerobic benthos and may have derived much of its nutrients from symbiotic sulphur-eating bacteria that were housed in thoracic gill filaments.
Among the Ptychopariida, the Olenina include a number of species that were known to occupy deep benthic, nearly anoxic substrates, high in sulphur compounds. Modern crustaceans in those situations (as at mid oceanic ridges and thermal vents) often live in a symbiotic relationship with sulphur-eating bacteria that are housed in the gill structures. Fortey suggested that this relationship may have originated with olenid trilobites in the Palaeozic. He cites the very wide thoraxic pleurae of many olenimorphs, increased numbers of thoracic segments (both traits providing ample gill surface for symbionts), and a reduced hypostome, so small that it suggests that much of the nutrient requirements for these trilobites were not being processed and ingested, but absorbed through the gills from symbiotic bacterial colonies there. Olenids are known for their well-developed gill structures, as in this image of Triarthrus.
A number of trilobites are thought to have been able to swim freely in the water column, and had wide distributions globally. Several of these had streamlined bodies and large eyes (such as the bathyuroids Carolinites, Telephina, and Opipeuterella), which would be appropriate for either active visual hunters of zooplankton, and as a means of predator detection and avoidance for a small nectonic animal in the water column. Below is a reconstruction of the pelagic trilobite Carolinites genacinaca (Proetida : Telephinidae), depicted swimming in a venter-up orientation.
Cryptolithus tesselatus (Asaphida : Trinucleioidea)
There are a number of trilobite species with large cephalic chambers, of sagitally convex form to house a filtering area, and bearing long genal spines or prolongations. The thorax and pygidium had a much less convex profile, and if extended straight back from the cephalon would be suspended significantly above the substrate. In a few of these species, the hypostome is also elevated, leaving an even larger space below for a filter chamber enclosed by the margins of the cephalon. Some good examples of this morphotype include harpetids (Ptychopariida : Harpina), trinucleioids (Asaphida : Trinucleioidea, such as the Cryptolithus at left), and perhaps some bathyurids (Proetida : Bathyuroidea, such as Uromystrum) and brachymetopids (Proetida : Aulacopleuroidea), such as Cordania. These are shown below:
©2000 by S. M. Gon III created using Macromedia Freehand, after Fortey and Owens 1999
In trinucleioids, the cephalic filter chamber is also often marked with rows of pits that are actually fenestrae, extending all the way through the exoskeleton from dorsal to ventral. This strongly suggests that they were used to allow outflow from the filter chamber, leaving edible particles behind. Fortey suggested a posture for Cryptolithus tesselatus in which the gill filaments on the thoracic and pygidial limbs were used to sweep sediments into the cephalic filter chamber from behind. The anterior legs were used to sort material in suspension, and processed water would exit from the pits in the cephalic fringe. The figure below illustrates this filter-feeding posture:
©2000 by S. M. Gon III created using Macromedia Freehand, after Fortey and Owens
UPDATE: In April 2001, at the Oxford Trilobite Conference, Fortey provided a brief update indicating that experimental testing of the model of water current circulation shown above yielded equivocal results, and did not support the idea of the trinucleid fringe as having an exhalant function, and that further tests and refinements are needed before a revised model can be developed.
The simple form and small size of agnostid trilobites has led to much speculation over the nature of their biology. The majority were either small-eyed or eyeless, so they were not visual hunters. One trilobite worker suggested they might be parasitic, but this is hard to reconcile with the large numbers of individuals sometimes found (they can become rock-forming material, as in the example shown below). This suggests to some that they may have been planktonic feeders, such as many ostracod crustaceans, hovering in haphazard swarms above the benthos, or moving upward and downward in the water column according to a day-night cycle, as many small planktonic organisms do today. If so, they may have fed on phytoplankton as one of the lowest of the primary consumers. Some may have perched on strands of algae, or perhaps fed there. At least one fossil specimen of agnostids preserved in a line suggests they were perched on a strand of seaweed when they were suddenly buried alive.
On the other hand, most pelagic free-swimming species of trilobites have large eyes, while Agnostida are typically blind. Adult forms are often found extended and in the company of benthic trilobites (for example, the association of Elrathia with Peronopsis). The global distribution of Agnostida suggests the majority were deep water benthic organisms, rather than wide-ranging planktonic species. It may be that some larval agnostida were planktonic, while adults settle out into a benthic lifestyle.
A first major division is to distinguish four species that appear to be pelagic, that is, swimming in the water column rather than moving over the bottom (benthic). These four can be further divided into two categories according to whether they appeared to be fast-swimming, streamlined taxa (Microparia and Degamella), or less-streamlined and presumably more sluggish, slower-swimming species (Pricyclopyge and Cyclopyge). Size is another diversifying attribute: while three of the four are relatively small species (ranging from 5 to 30 mm), one of them, Degamella evansi, reaches 50 - 60 mm in length, a quite large species for a free-swimming trilobite.
The remaining majority (seven taxa) of the Pontyfenni trilobite are benthic, and occupy three major trophic guilds: predatory-scavenging, filter feeding, and particle feeding guilds. These also were further subdivided according to size, with one rather tiny shumardiid particle feeder (Shumardia), and a range of small to large predator/scavengers (Dindymene, Colpocoryphe, Ormathops, and Illaenopsis). Two presumed filter-chamber feeders (Bergamia and Ampyx) complete the set of trophic specialists of the Pontyfenni trilobite fauna.
Trophic Partitioning in the Pontyfenni
(after Fortey and Owens 1999)
12 - 25 mm
50 - 60 mm
8 - 15 mm
12 - 30 mm
Pelagic species occupying the water column
Benthic species occupying reefs, slopes, and flats
5 - 10 mm
10 - 30 mm
15 - 35 mm
52 - 60 mm
8 - 22 mm
5 - 10 mm