Australia: The Land Where Time Began

A biography of the Australian continent 

Sauropod Biology – Digestive Physiology and Feeding

In many herbivore ecosystems of the Mesozoic the sauropod dinosaurs dominated the herbivore niche. It is suggested by Hummel & Clauss that significant symbiotic gut microbial activity can be safely inferred for these large animals, based on evidence from extant herbivores. A hind gut fermentation chamber is believed to be more likely to have been present in sauropods than a foregut digestive system, based on evidence from extant herbivores. There are several herbivore-relevant features, such as the large foraging range, as a result of the length of their long necks, the apparent lack of comminution (such as chewing, grinding up) of these very large animals, which in large extant herbivores, is highly atypical, and the extremely high body mass, which is linked to several key features of the foraging and digestion of herbivores. Hummel & Clauss say the capacity of their gut can safely be assumed, on the basis of regressions of extant herbivores, to have been highly comprehensive in relation to energy requirements. It can be implied by this that they had extremely long food retention times, though this is not necessarily the case. There are some potential implications for the respective harvesters of the various plants available to sauropods, as well as these animal features, and the spectrum of food plants available for sauropods that were in sufficient quantity (such as sphenophytes, e.g. horsetails), gymnosperms, (e.g. conifers, cycads, etc., pteridophytes (ferns and related plants), differed greatly from those available to extant herbivores, the available plants being mostly angiosperms. There is a tendency among gymnosperms to facilitate cropping sizes that are rather large, which are measured in kilograms of dry plant material per bite, which would allow a large intake of food. Experiments concerning the in vitro digestibility of extant representatives of potential sauropod food plants were estimated to be better than expected, being at least comparable to the level of extant browse. According to Hummel & Clauss sauropods must be considered to be one of the greatest success stories in the history of large herbivores, though the sauropods differed greatly form large extant herbivores.

The sauropods are sometimes perceived as huge animals with a huge, though strange, body with an unusual body shape, but Hummel & Clauss say they should be perceived as one of the most successful animal forms that ever existed. They are, however, the largest herbivore ever, as well as the terrestrial vertebrate animal dominating the terrestrial megaherbivore niche on most land masses for 135 Myr from the close of the Triassic to the end of the Cretaceous; they should be regarded as the most successful vertebrate herbivore ever known. The group sauropods is comprised of a large group of diverse herbivores that Hummel & Clauss say should not be regarded as all sauropod species having a uniform digestive physiology, in the same manner as the primates, e.g., which have a large variety of digestive strategies.

It has been repeatedly demonstrated that the differences in the anatomy and dentition of the skulls of different sauropod clades (Calvo, 1994b; Christensen, 2000; Upchurch & Barrett, 2000), e.g., exceed the complexity of those observed between artiodactyls and perissodactyls (even-toed ungulates and odd-toed ungulates (hoofed animals), respectively). Hummel & Clauss suggest it is apparently likely that some of the taxa of sauropods were to some degree specialised to utilise certain groups of plants, as occurs at the present among the modern megaherbivores that can selectively feed on certain plant types such as grass in the case of hippos, white rhino, large bovids), browse (giraffe, black rhino, Sumatran rhino), or both grass and browse (Indian rhino, in which there is some tendency to eat more grass, or in the case of the elephant, the African elephant which as a tendency to include less grass in its diet than the Asian elephant does (Clauss et al., 2008a). It is necessary to keep in mind that when speculating on the digestive physiology of sauropods this speculation is based almost exclusively on extrapolation from the digestive systems of extant animals; i.e., is necessary to extrapolate a long way above the body mass range of the animals on which the research on digestive physiology has been carried out, and it is necessary to use modern analogues from a different clade, mammals, that at present occupy the niche of megaherbivores. It is the soft tissue of the stomach and intestines that is required to study the digestive physiology of these extinct animals, which is obviously the part of the animals that has not been preserved in the fossil record of dinosaurs.

A variety of coprolites (fossilised dung) and gut contents of herbivorous dinosaurs have been fossilised (Stokes, 1964; Chin & Gill, 1996; Hollocher et al., 2001; Ghosh et al., 2003; Prasad et al., 2005), very few of these can be safely assumed to be from sauropod dinosaurs (Sander et al., 2010); therefore this information source is unavailable for research on sauropods at present. Given these circumstances it is necessary to speculate on the digestive physiology of sauropods by the use of educated guesses that are based partly on extrapolation from extant herbivores.

According to Hummel & Clauss many authors have written about dinosaur feeding (Coombs, 1975; Bakker, 1978; Krassilov, 1981; Weaver, 1983; Coe et al., 1987; Farlow 1987; Dunham et al., 1989; Weishampel & Norman, 1989; Dodson, 1990; Taggart & Cross, 1997; Tiffney, 1997; Upchurch & Barrett, 2000; Magnol, 2003).

Sauropods are different, showing alternative evolutionary strategies in digestive physiology in vertebrates that would otherwise not have been suspected, which elucidates constraints under which extant herbivorous animals operate, that would not have been noticed as being constraints, instead being taken as a matter of course.

Sauropod feeding and food processing

It appears, based on analyses of their dentition, all sauropods were exclusive herbivores as adults (Upchurch & Barrett, 2000; Weishampel & Jianu, 2000; Barrett & Upchurch, 2005; Stevens & Parrish, 2005; Sander et al., 2009, 2010a), though the hatchlings may have occasionally eaten arthropods or other small animals (Barrett, 2000).


Sauropods had very large bodies and very long necks, which were among their most remarkable features. Among the various forms of sauropods the neck types and body forms varied, different types, such as Brachiosaurus, long neck and long front legs, Diplodocus, long neck and front legs that were rather short, or Dicraeosaurus, which had a rather short neck. The biomechanical function of the neck (Stevens & Parrish, 1999, 2005; Christian & Dzemski, 2007; Christian & Dzemski, this source), and the evolutionary causes leading the evolution of a long neck, such as that that has been hypothesised for the giraffe (Simmons & Scheepers, 1996; e.g. sexual selection, Senter, 2007) have been the subject of considerable attention and discussion (See Sander et al., 2010b for a review). It is only recently that it has been shown explicitly for the giraffe that the long neck is probably a result of feeding competition within the browsing guild (Cameron & Du Toit, 2007), in spite of its apparent obviousness. A clear stratification of the height of feeding between the giraffe and other browsing ruminants in Africa was also described by Du Toit (1990). Hummel & Clauss suggest it appears to be most likely that this interpretation also applies to sauropods, a result which sorts the taxa of sauropods by their height of feeding (Upchurch & Barrett, 2000). The feeding range of most sauropods, given their long necks, must be regarded as being extremely large, a characteristic that is also found in the largest extant herbivore, the elephant, as a result of its trunk (Colbert, 1993).

Skull and teeth

Hummel & Clauss suggest that in diplodocoids and titanosaurs the pencil-shaped teeth that restricted to the front of the snout and in basal sauropods and basal macronarians (a clade of sauropods, Mid Jurassic-end of Cretaceous) the more massive dentition of spoon-shaped teeth that have wear facets can be regarded as extremes in the dentition and skull types of sauropods. The teeth seen in the diplodocoids and titanosaurs suggest a type of cropping that involves raking the leaves off a twig, so leaving the woody shoots that are less digestible, while the teeth of basal sauropods and macronarians suggest there was some degree of biting off and at least potentially some degree of mastication that could have involved puncturing or crushing the material, or damaging the cuticle of the leaf, at least, the cuticle being a major barrier during ingestion to allow the access of the bacteria into the leaf. It has been demonstrated (Fiorillo, 1998) that difference in the microware of teeth and corresponding differences in the food type selected have been proposed (Bakker, 1986; Galton, 1986). An unexpected diversity of dentitions has been uncovered by recent investigations (Barrett & Upchurch, 2005), and a separation of skull types that is more detailed can be applied that resulted in various degrees of oral food processing (Calvo, 1994b; Christensen, 2000; Upchurch and Barrett, 2000; Barrett & Upchurch, 2005). According to Hummel & Clauss the term “oral processing” is too ambiguous and should be avoided, as it is not clear to what extent biting off, forage cropping, or mastication/comminution of the forage that has been cropped is meant. Terms that are more descriptive, such as “biting-off,” “stripping off,” “chewing,” “mastication,” and “reduction of particle size,” and even the components of oral processing, such as swallowing and lubrication, would facilitate better understanding.  Sauropods are exceptional herbivores in that their teeth do not have any adaptation to masticating and grinding food, whatever component of oral process is being referred to in the literature, compared to mammals. It remains vague what a concrete interpretation of what might have comprised the different feeding niches, though it is commonly believed that as well as different feeding heights, dentition differences and the method of cropping food that has been extrapolated, they contributed significantly to separation of niches among sympatric taxa of sauropods.

Gastric Mill

The gastric mill of birds is the alternative efficient option for processing food in extant vertebrates. Among vertebrates, apart from birds some herbivorous fish, such as mullet, have been reported to have a functional device for the reduction of food particle size within their guts (Guillaume et al., 1999), though gizzards that have some grinding function have been described in a variety of invertebrates (Morton, 1979; Dall & Moriarty, 1983). The existence of such a device in sauropods has been favoured by a number of authors (Janensch, 1929; Bakker, 1986; Galton, 1986; Farlow, 1987; Weishampel & Norman, 1989; Wing et al., 1992; Christiansen, 1996; Taggart & Cross, 1997; Bonaparte & Mateus, 1999; Upchurch & Barrett, 2000; Saunders et al., 2001). The existence of an avian gastric mill is denied by others, however, who argue that pebble aggregations that have been interpreted as gastric mills are a sedimentological phenomenon (Calvo; 1994a; Lucas, 2000; Wings, 2003,2005) and they have argued that the amounts that have been recovered are far too small to be regarded as functional in such large animals with body sizes of sauropods (Wings & Sander, 2007). The conclusion has been arrived at (Wings & Sander), on the basis of a critical review and evaluation of the fossil record, as well as comparative studies of the gastric mill of the ostrich, that to date no evidence has been found of a gastric mill in sauropods.

Hummel & Clauss suggest the relevance of this assumption cannot be overestimated. The food ingested by herbivores is basically reduced in size in 1 way: mechanical breakdown. There is generally little further breakdown of particles of ingesta in terrestrial herbivores (Pearce, 1967; Poppi et al., 1980; Murphy & Nicoletti, 1984; McLeod & Minsen, 1988; Moore, 19999), after the ingesta has passed the particle reduction site, either oral cavity with it dental apparatus, or gastric mill with gastroliths. If it is supposed that a gastric mill and a masticatory apparatus are not present, this means that it is unlikely there was any other significant means of the ingesta particle size being reduced. Hummel & Clauss suggest that in particular the lack of particle breakdown might have been compensated for by long times of ingesta retention in the gut, and therefore long exposure to microbial fermentation. This would mean the sauropods were the ultimate herbivore nonchewers.

Hummel & Clauss also suggest during initial colonisation rumen fungi increase the rate of degradation of coarse forage by aiding other microbes to gain access to cell walls (Van Soest, 1994; see Bjorndal, 1997 for a discussion of a similar role of nematodes in the digestive trace of herbivorous reptiles). The decreasing effect of large particle size on the rate of degradation would be partially lessened by such action.

Cropping efficiency – animal factors

The head of sauropods that are often said to be small (Russell et al., 1980; Weaver, 1983; Coe et al., 1987; Farlow, 1987; Dodson, 1990; Colbert, 1993), has been the subject of much speculation, the head often being too small to allow sufficient intake for an endotherm-like metabolism. According to Hummel & Clauss there are 2 separate data collections that argue against this scenario (Paul, 1998; Christensen, 1999), but this hypothesis is still sometimes referred to. The skull of sauropods may appear small, this appearance of smallness is mainly the result of its shortness when compared to those of extant herbivores, it is the result of the jaws having no chewing teeth); The mouth opening width in sauropods is within the range expected if it is extrapolated from mammalian herbivores.

It has been stated (Pail, 1998) that the width of the skull of indricotheres and sauropods of the same body mass was not different from mammalian herbivore, and the mammalian herbivore can even be narrower than in the sauropods, and the conclusions of Christiansen (1999) were based on measurements of 11 species of sauropodomorphs and 88 species of mammals, including 27 ungulate species. Hummel & Clauss say it should be added that applying the respective regressions that were set up by Christiansen (1999) for artiodactyls/perissodactyls and sauropodomorphs up to 10,000 kg animals resulted in a value that was considerably higher for the mammalian herbivores compared to the dinosaurian herbivore (29.5 vs 17.7 cm in muzzle width). The predictive value for metabolic rate of any differences there are in the dimensions of the skull should not be overemphasised as there are factors such as plant morphology that may determine the upper limits for an extensive constant increase in the width of the muzzle with body weight.

According Hummel & Clauss the time spent feeding could still be a limiting factor for herbivores the size of sauropods, given the allometric increase in time spent feeding in the activity budget of, e.g., the elephant, that spends up to 80 % of their budget over 24 hours. In models of biomass rate of intake of herbivores there is usually a trade-off between the size of the bite, the amount of food taken in 1 bite, which is a function of availability and structure of the biomass, and the bite rate, the number of bites per unit time, as larger bite sizes usually implies more time elapsed between bites as time is taken in mastication. Therefore, rate of intake increases with the size of the bite, but in an asymptotic function (Spalinger et al., 1988; Shipley & Spalinger, 1992; Spalinger & Hobbs, 1992; Gross et al., 1993a, 1993b; Ginnett & Demment, 1995; Bergman et al., 2000; Illius et al., 2002); this function being usually referred to as a type II functional response curve of a species (sensu Holling, 1959). The assumption that the intake of food is ultimately limited by the amount of oral processing is the basis of all the derived intake models, i.e., mainly mastication (Yearsley et al., 2001).

Hummel & Clauss suggest there should not be a trade-off in animals that do not chew their food, such as extant birds, though also stegosaurs or sauropods, and for any given bite size any gain in the size of the bite should translate more directly into a gain in foraging rate. Oral processing should consist of only cropping and swallowing in such animals, both process that can be considered to be process that are much less dependent on the size of the bite in terms of the time they require. The increase in rate of intake with increasing size of the bite might therefore be either linear (type I functional response) or should contain a linear component that is longer before the limiting effect cropping/swallowing sets in. Then the maximum intake would not be dependent of the process of mastication but by the maximum amount that could be cropped and/or swallowed by such animals,

In herbivorous birds type I functional response curves have been found (Rowcliffe et al., 1999), thorough type II curves that that have an assumed limitation that is due to the increasing effort of swallowing with increasing size of the bite have also been described (Durant et al., 2003). It has also been suggested that in birds the oral handling of cropped food may be a time-consuming process similar to chewing that is present in mammals (Van Gils et al., 2997). There are no studies that have been undertaken on the functional response in reptiles.

Cropping efficiency – Plant factors

The morphology of plants that are actually cropped is an additional important factor that is only rarely considered when the cropping efficiency or efficiency of intake is being discussed. Theoretically it is expected that food plants favoured by animals should have characteristics that allow the grasping of a large amount of biomass per bite, which would make for more efficient cropping, over plants that have a structural form that allows only small amounts of biomass to be cropped per bite. When plant distribution is expressed as a measure of density in terms of biomass per area it correlates with intake by mammalian herbivores (Stobbs, 1973; Trudell and White, 1981; Wickstrom et al., 1984; Spalinger et al., 1988; Wallis, de Vries & Daleboudt, 1994; Hartley et al., 1997; Shipley et al., 1998; Illius et al., 2002). It was demonstrated by these studies (Shipley et al., 1998) that for moose up to 20 times the mass of twigs and leaves are made available by conifers as opposed to dicotyledonous trees, though these studies mainly dealt with the effect of moving between feeding places. According to Shipley et al. this effect was due particularly to fundamental differences in architecture of the plants between the 2 groups, though the study was carried out in winter, a time when deciduous trees would have been leafless anyhow. It has been demonstrated in feeding trials with captive deer, Odocotleus virginianus, that plant morphology, that for the terminal 20 cm of a twig the leaf mass was measured, which correlated positively with efficiency of foraging in terms in the rate of intake (Koerth & Stuth, 1991. There the assumption that a high biomass of twigs will result in higher intakes is supported empirically.

Hummel & Clauss suggests that perceptions of sauropod feeding can be influenced by such considerations. According to Hummel & Clauss they wanted to test their subjective experience that comparably sized twigs of conifer are considerably heavier than twigs of deciduous broad-leafed trees, which indicates that the conifer twigs provides more biomass per bite for herbivores. The biomass per sauropod bite was investigated for 20 individual shoots from different individuals of 14 plant species that included different conifers, Gingko, and Metasequoia. A Diplodocus skull was used to estimate the depth of 30 cm to which a twig could be cropped. The first 30 cm of a shoot was clipped, and this piece was divided into 2 15 cm parts, which represented the first and second to be encountered by a potential herbivore.

There were significant difference between plant groups for all parameters, leaf weight and twig weight at 15 and 30 cm, respectively, and total weight at 30 cm.

Sources & Further reading

  1. Jürgen Hummel & Clauss, Marcus, Sauropod Biology and the Evolution of Gigantism: What do We Know? In Nicole Klein, Kristian Remes, Carole T. Gee & P. Martin Sander (Eds.), 2011, Biology of Sauropod Dinosaurs, Understanding the Life of Giants, Indiana University Press, Bloomington & Indianapolis  


Author: M. H. Monroe
Last updated  05/05/2016
Journey Back Through Time
Experience Australia
Aboriginal Australia
National Parks
Photo Galleries
Site Map
                                                                                           Author: M.H.Monroe  Email:     Sources & Further reading