Taking a Bite out of Anomalocaris  

Anomalocaridids are hypothesized to have consumed trilobites.  Other than their large size, their midgut glands, and malformed trilobites, there is little direct evidence that they did so.  New taphonomic, compositional, and modelling evidence suggests that anomalocaridid mouths were soft, could not close completely or chew, had biting kinematics incompatible with many trilobite malformations, and were well suited to manipulate or suck soft prey. 

Anomalocaridid mouth plates and their tips are never broken, nor are tips worn.  If plates were hard, and were used to manipulate, puncture, crush, or masticate biomineralized prey, they would be expected to show evidence of abrasion or breakage.  Absence of this evidence is striking given the frequency (0.01-1%) of healed malformations in extant marine arthropods, most of which are due to prey manipulation or feeding.  Moreover, anomalocaridid plates and their biting tips are commonly wrinkled, exhibit preburial shearing and tearing, and mantle or are deformed by biomineralized fossils such as brachiopods, trilobites, and Scenella.  Plates are preserved as organic carbon and exhibit fracture patterns typical of desiccating arthropod cuticle.  Thus anomalocaridid plates, including their tips, were unmineralized and pliable in life. 

Computer aided design modelling of the kinematics of mouth opening and closure, together with comparison with muscle movements used by modern circular-mouthed organisms, suggests several plausible models for anomalocaridid mouth movement.  These include sphincter-like constricting closure of the circlet of plates, and full- or half-eversion or inversion of the circlet; the latter two movements generate sufficient subambient pressure for suction feeding.  In all closure modes, laterally-adjacent opposing plates intersect one another when the mouth closes, which prevents the circlet from closing more than half-way. Orientations of plate tips are consistent with a partial mouth closure model; if full closure was possible, opposing plate tips would not articulate or interlock with one another, as is expected from teeth optimized to masticate or puncture, or teeth which intersect at tooth tips to crush, puncture, or break. 

Although bilaterally-oriented trilobite malformations can plausibly be explained by a closure of a circular mouth, most trilobite malformations are arc- or U-shaped.  Suction-, eversion-, and sphincter-movement of anomalocaridid jaws cannot produce U-shaped bite marks; these are better explained by predators who had opposable jaws or claws.  Finite Element Analysis, and modelling of anomalocaridid plates using cuticle yield strengths from modern shrimp (Pandalus) and lobster (Homarus), illustrate that plates could withstand maximum forces of up to 6.2 and 13.0 N, should they have bitten into the thoracic segments of a trilobite.  The most commonly malformed Cambrian trilobites had maximum skeletal yield strengths ranging from 3.7–37.1 N, suggesting that only weakly mineralized taxa, such as Elrathia kingii, could have been broken by an anomalocaridid bite. 

Anomalocaridids may have bitten some soft trilobites, but it is more likely that they were suctorial feeders, perhaps using their preoral appendages to comb soft-bodied invertebrates from the benthos.  This feeding strategy makes sense given the recent discovery of multiple rows of inwardly pointing serrated plates inside some anomalocaridids’ oral cavity; these may have prevented prey from exiting the mouth, or may have been part of a buccal cavity or eversible grasping organ.