plesiosauria@gmail.com

All rights reserved. Last updated May 2007. This site first opened in June 2001 and is continuously underconstruction.

FEEDING AND SENSES

Classifying types of evidence for diet

Pollard (1990) classified evidence for diet into three broad categories: direct, indirect and general. Direct evidence consists of various grades of stomach contents, indirect evidence includes bite marks, and general consists of functional analysis.Martill et al. (1994) identify 8 criteria that can be employed to assess the strength of evidence for trophic links (the links in a food web) (Table 1):

Evidence for diet in plesiosaurs (Superscript suffix numbers refer to criteria in table 1)

Precoprolites, coprolites and bite marks (1, 2)

'Precoprolites' is a term used to include gut/gastric contents, regurgitates and intestinal residues (Pollard, 1990). Exceptional conditions are needed for the preservation of soft stomach contents. Stomach contents are direct evidence, especially if undigested (Pollard, 1990) but are rare in plesiosauromorphs (Sato and Tanabe, 1998). In one known pliosauroid, teuthid cephalopod hooklets are dominant. Rarer are fish teeth and an indeterminate reptile tooth (Martill et al. 1991). The hooklets in a pliosaur from the Lower Oxford Clay of Peterborough closely resemble those of the contemporaneous Belemnoteuthis antiquus. It may be supposed that these have a secondary origin, coming from the stomachs of larger prey but this is unlikely.

A specimen from Northern Japan preserves ammonoid jaw apparatus (possibly two taxa) concentrated in abundance in the stomach region (Sato and Tanabe, 1998), the first and only evidence for a long-standing assumption that ammonoid cephalopods were prey to plesiosaurs. Local and seasonal variation in availability of prey is a factor that should not be overlooked (Massare, 1987; Martill et al. 1994) and accordingly, stomach contents in a few specimens should not be taken as fully representative of diet.

Coprolites are direct evidence for diet, but the diet of what? No plesiosaur coprolites have been positively identified. Coprolites must be used in association with criteria 2, 3 and 5 to determine 'who did it'.

Possible pliosaur (or large crocodile) bite marks are known from small isolated Kimmeridge Clay marine reptile bones including small plesiosauroid limbs (fig 4) (Martill, 1994), ichthyosaur vertebrae (Martill, 1992) and one specimen of Pliosaurus brachyspondylus contains the remains of a thyreophorean dinosaur (Taylor et al. 1993).

Functional morphology of the skull (3)

In vertebrate palaeobiology, a functional analysis of the jaws and dentition (being the primary structure for food input, if not also acquisition and processing) is typicall useful (Pollard, 1990). There is a functional compromise in cranial structure between thin delicate structure enabling food to be caught, at the expense of it becoming weak (Martill et al. 1994). Evidence for twist feeding in some pliosaurs comes from a strong triangular shaped skull, very deeply rooted large teeth and a bulbous symphysis (Martill et al. 1994).

A detailed analysis of the functional morphology of the skull has been made for two reasonably well known pliosaurs, Rhomaleosaurus zetlandicus from the Lower Jurassic (Taylor, 1992) and Pliosaurus brachyspondylus, from the Upper Jurassic (Taylor and Cruickshank, 1993). The former is interpreted as feeding on a wide range of prey and forcibly dismembering large prey, based on evidence of the osteology and musculature. In P. brachyspondylus, the cranium is even more robust and the posterior dentition is unusually recurved (fig 5) to act as a ratchet to pull struggling prey into the mouth (Taylor and Cruickshank, 1993). Large postorbital regions anchor a large M. adductor mandibulae muscle and ensured a powerful bite (Massare, 1988). This has also been identified in plesiosauroids (Cruickshank and Fordice, 2002).

Plesiosaurs occur in a wide range of forms able to deal with pray of all sizes (Martill et al. 1994). In contrast to pliosaurs, the small headed 'plesiosauromorphs' were unable to rip chunks off of carcasses because their skulls were lightly built (Brown and Cruickshank, 1994) and unable to resist torsion (Cruickshank, 1994) (e.g. compressed mandibular rami and weak symphysis) . Plesiosaur teeth were not used for chewing, consequently the size of prey in these forrms was directly limited by the size of their gullet (Massare, 1988), the largest of which is only 15cm or so.

The eyes were flattened (as evidenced by the sclerotic ring) enabling underwater vision (Lambert et al. 2001) and often upwards orientated in plesiosauroids suggesting they ambushed from below rather than from above or as traditionally depicted, with implausible swan-necks darting down at fish from above the air-water interface ("hunting platforms") (Storrs, 1993; Everhart, 2002). Some pliosauroid eyes are positioned laterally (Massare, 1988) (see Storrs and Taylor, 1996 for example) indicating attack on their own level in the water column. It is important to note that the function of the inflexible (Storrs, 1993; Forrest, pers. comm. 2003) long neck is still unresolved (Martill et al. 1994), although most likely is served as a mechanism of approaching prey without being detected (Massare, 1988). It did not have the flexibility to strike prey as do some snakes and pleurodiran turtles (Pough et al. 1996). Stapes are rarely preserved in plesiosaurs but where known, it is evident they were of no use in hearing (Storrs and Taylor, 1996) in air at least (Lambert et al. 2001).

An olfactory system (Cruickshank et al. 1991) (fig 8) has been suggested as a common adaptation in the Plesiosauria (Brown and Cruickshank, 1994). The anteriorly placed internal nostrils have palatal grooves to channel water into them, the flow of which would be maintained by hydrodynamic pressure over the posteriorly placed external nares during locomotion. During its passage through the nasal ducts, the water would have been 'tasted' by olfactory epithelia. However, a recent study by Buchy et al. 2006 questions the position of the internal nares, proposing that they are actually located at the rear of the palate (the posterior interpterygoid vacuities).

Post-cranial Functional Morphology: Locomotion (3)

Swimming capabilities have implication for diet (Robinson, 1975; Massare, 1988). Plesiosauroids and pliosauroids had different locomotoay repertoires. The latter 'sprinted' in short bursts possibly using large flippers to take very fast moving prey "by stealth rather than pursuit" (Martill, 1992). The plesiosauroids were typically endurance swimmers with lower flipper aspect ratios and drag-inducing long necks (Robinson, 1975). Massare (1988) comes to similar concusions based on hydrodynamic properties of Mesozioc reptiles calculating speed of plesiosauroids at 2.3 m/sec (= slow ambush predator) and pliosauroids slightly faster (= pursuit predator). (For more information on locomotion in plesiosaurs, visit the locomotion page)

Gastroliths (3, 5)

The intentional swallowing of stones has been known in plesiosaurs and living and extinct crocodiles for a long time (see Williston, 1893, 1894, 1904). The significance of gastroliths in prehistoric reptiles has been investigated by Kobayashi et al. (1999) who observe that in herbivorous birds gastroliths and gizzards are common, but they are absent in carnivorous birds. The presence of gastroliths is common in plesiosaurs, Pollard (1990) notes that plesiosauroid stomach contents "usually contain gastroliths", (fig 9) (especially elasmosaurids see Everhart (2000) and Cicimurri and Everhart (2001) for example) yet are rarer in pliosaurs (Storrs, 1993; Sato and Tanabe, 1998) However, gastroliths are reported in one specimen cf. Liopleurodon sp. Martill, (1992), along with a sand fraction), and recent discoveries in Arizona (see my news page), preserve gastroliths with large pliosaurs (Schmeisser, pers. comm). It is likely that the bias is a result of lack of study in pliosauroids, which are large and difficult to work on (Forrest, pers. comm. 2002). Alternatively, the differecne may be genuine, and represents different functional regimes between pliosaurs and plesiosaurs(Storrs, 1993).

Presuming for now, this gastrolithic discrepancy between long and short-necked forms is genuine, and that as Kobayashi et al. (1999) suggest, gastroliths point to herbivory, then what are we to make of their presence in plesiosauroids? A role in buoyancy control may be possible (Chatterjee and Small, 1989), a less physiologically expensive way of attaining negative buoyancy than pachyostosis (Martill and Naish, 2000), presumably short-necked forms are adept without such stones. Recent work by Don Henderson (2006) suggests that gatroliths have a role in stability within the water cloumn, rather than buoyancy control.

Where present, gastroliths are usually found in small concentrations although >100 are known for some elasmosaur specimens (Everhart, 2002). However, even here the relative weight would be insignificant in such large animals (Ciccimurri and Everhart 2001, Everhart, 2002), although it only takes a few grams to tip a balance. Despite conflicting with so much of the other evidence, omnivory in plesiosauroids is worth consideration. Is a long neck advantageous for browsing in the aquatic environment as it is for terrestrial sauropod dinosaurs and modern giraffes? Indeed, if a long neck is useful for predation why is it so rare? It is a feature lacking in all other large marine carnivores past and present.

Gastroliths may alternatively have been used for grinding food in the plesiosaurs guts. Gastroliths may have had a dual or even multi-purpose (Storrs, 1993).

Analogy (5, 8)

Plesiosaurs were air breathing reptiles and must have surfaced frequently and could not have dived for food for prolonged periods of time. However determining the details of such behaviour is problematic. Analogy can be used to infer diet in marine reptiles. Martill et al. (1994) assumed a number of extant marine organisms including cetaceans, penguins and pinnipeds to be analogous to plesiosaurs in many aspects, and Massare (1987) presents a number of similarities between the teeth of marine reptiles and modern large marine carnivores (fig 10).

The teeth of the piscivorous gavial (Gavialis gangeticus) show similarities with plesiosauromorphs (fig 10. A and C) and the teeth of killer whales (Orcinus orcus), which can eat large mammals, show many similarities with pliosauromorphs (fig 10. B and D) (Massare, 1987). Plesiosaurid teeth also interlock, another adaptation of piscivores (Benton, 1990). A 'rushing upwards' style of attack is inferred for large pliosauromorphs, an analogy drawn from the modern great white shark (Charcharodon) (Martill and Naish, 2000). As preumed ectotherms (Chatterjee and Small, 1989), plesiosaurs could presumably go for considerable periods of time without food.

Incomplete carcasses (2, 5)

Isolated bones and fragmentary yet articulated skeletons are found in many deposits contemporaneous with large pliosaurs, this is further evidence suggesting twist feeding and shake feeding (Martill et al. 1994).The bones were probably dropped in the water column, although they may have dropped from floating carcasses.

ECOLOGICAL GUILDS

Because plesiosaurs are so diverse it is sensible to split the group into ecological guilds, independent of artificial taxonomy. Massare (1987) performed a division for all of the large aquatic reptilian groups based on tooth form and presented the guilds identified, in a triangular diagram (fig 11). Of seven guilds, the Plesiosauria transcends three: pierce, cut and general, although subsequent guilds have since been proposed. The properties of any tooth (table 2) can be presented in a diagram using a variety of absolute (ratios) and semi-quantitative measurements (fig 12).

Predators?

Carnivory does not automatically invoke feeding behaviour and this must be treated separately. Evidence for scavenging of floating corpses comes from stomach contents including dinosaurs (Taylor et al. 1993) (see above) and pterosaurs (Massare, 1987). This is also evidence supporting Martill's (1992) premise that large pliosaurs were opportunistic.

Predation consists of phases: (1) search, (2) capture, (3) penetration and (4) ingestion (= subjugation), (5) digestion and (6) defecation (Brett, 1990). Phases 1 to 3 especially, will invoke repercussions on the biology of and predatory organism (table 3).

Predation phase evidence in Plesiosauria
1. Search Large eyes, olfaction system, locomotor adaptation.
2. Capture Tooth and functional morphology, Isolated bones and articulated but incomplete skeletons of possible prey.
3. Penetration Tooth and functional morphology: i.e. fast snapping muscular jaws and supporting pterygoid flanges.

Table 3. Phases of predation and evidence for each in the Plesiosauria: clearly, plesiosaurs were predators.

'Pre-ingestive breakage' can be direct ichnological evidence but diagnostic bite marks represent predation (rather than scavenging), only if subsequent regrowth of the damaged object can be observed. Here would be evidence of an unsuccessful attack. Fossils of carnivores in the act of predation are rare, and none are known for the Plesiosauria. An interesting but very indirect source is anti-predation adaptations in presumed prey (Brett, 1990), for example Jurassic hybodont sharks contemporaneous with plesiosaurs have spines (Martill, 1991).

Unusual plesiosaurs

The suite of characters possessed by the heavily set pliosaur Pachycostasaurus (fig 13) from Peterborough, UK (Dawn, 1997) does not fit neatly into any of the categories proposed by Massare (1987) and the authors adopt the genus as a "generalistic feeder". Its pachyostic ribs allowed it to traverse the seafloor, perhaps searching for benthos. Another case arises for the Cimoliasaurids (sensu O'Keefe, 2001) with their hundreds of long slim and delicate teeth (e.g. Morturneria (fig 14)), an additional guild has been proposed to accommodate such oddities: a 'trap guild' (Chatterjee and Small, 1989). This may be analogous with the extant Crabeater seal, which has sieve-like teeth for capturing krill (Martill et al. 1994). Cryptoclidus also has such teeth and shares its environment with the very common decapod crustacean (Mecochirus), a relationship fulfilling criteria 2 (table 1). However, it is unlikely than any plesiosaurs were strict suspension feeders in the way baleen whalws feed today (Colin and Janis, 1997).

Conclusions

All plesiosaurs were predatory carnivores belonging to one or more of four ecological feeding guilds. Pliosaurs were top carnivores in their respective foodwebs (Martill, 1992, Martill et al, 1994; Sato and Tanabe, 1998), perhaps only exceeded in ferociousness by large mosasaurs in Cretaceous ecosystems. They were both pursuit predators of various sized prey (up to other marine reptile size) and opportunistic feeders belonging to the general and cut guilds of Massare (1987). Twist feeding was employed for larger meals (Taylor, 1992).

The plesiosauroids such as Plesiosaurus were less generalised feeders (Maisch and Rucklin, 2000) but still belonged to two guilds, the pierce 1 and general guilds. The teeth were used to pierce small soft-bodied prey, especially fish (Massare, 1987). Some taxa have made excursions from these genearal trends and perhaps employed filter feeding (trap guild). Hard and soft-bodied cephalopods also formed part of the diet of all plesiosaurs. Plesiosaurs hunted visually and(or) with the use of a directional sense of olfaction.

References Page