December 20 , 2016:

by David Button

Introduction

The sauropods are some of the most iconic prehistoric vertebrates. Their unique body plan — long neck and tail, bulky body and proportionately tiny head — is perhaps the most famous image of ‘a dinosaur’ and the group includes household names such as Brontosaurus, Diplodocus and Brachiosaurus. Sauropod remains have been found on every continent, and they were one of the most important groups of terrestrial giant plant eaters, or megaherbivores, throughout the Jurassic and Cretaceous periods (201 million to 66 million years ago). The single most notable sauropod trait is their gigantic size: the largest sauropods would have measured more than 40 metres from nose to tail, reached 18 metres tall and tipped the scales in the region of 60–80 tonnes, making them the largest land animals known to science. Sauropods show the limits of how big terrestrial life can get, which has led to a considerable amount of research into their evolution, anatomy and ecology.
Relationships, ancestry and evolutionary history

Sauropods belong to the larger group Sauropodomorpha, one of the two main types of saurischian (‘lizard-hipped’) dinosaur. Sauropodomorphs are distinguished by features including an elongated neck and an enlarged claw on the thumb. The other saurischian group, Theropoda, includes animals such as Tyranosaurus, Velociraptor and birds. By contrast, there are no living sauropodomorphs.

Sauropodomorpha includes the sauropods and a series of earlier creatures informally called ‘prosauropods’ (Fig. 1). The first sauropodomorphs are among the earliest dinosaurs known, appearing around 230 million years ago, in the late Triassic period. The group then expanded, with prosauropods becoming the dominant medium-to-large-bodied herbivores in many ecosystems during the late Triassic and early Jurassic. Prosauropods, exemplified by the well-known Plateosaurus, were mostly bipedal, bulky animals with a long neck, tail and grasping hands. They ranged from less than 2 metres to 10 metres long and from around 10 kilograms to well over a tonne. Although common during the Triassic, prosauropod families declined through the early Jurassic before disappearing at the end of the epoch (around 174 million years ago), perhaps due to competition with early sauropods.

Sauropod diversity peaked by 150 million years ago, in the late Jurassic. By this point, very large body size (over 10 tonnes) was the norm and multiple different groups of sauropod had independently developed truly gigantic proportions (lengths of more than 30 metres and masses of over 30 tonnes). Neosauropods were abundant, with animals from the two main neosauropod groups — Diplodocoidea (Fig. 3) and Macronaria (Fig. 4) — dominating sauropod communities. Jurassic Diplodocoids included the relatively small dicraeosaurids and the elongate diplodocids, such as Diplodocus. Macronaria includes sauropods characterized by an enlarged bony nostril and comparatively large forelimbs, such as Camarasaurus and the brachiosaurids.

Sauropod diversity decreased noticeably around the end of the Jurassic period and the start of the Cretaceous, 145 million years ago. They were still important parts of global ecosystems, however, and continued to dominate the Southern Hemisphere. Diplodocids barely limped into the early Cretaceous, with a single genus, Leikupal, and a few more fragmentary remains known from South America. The dicraeosaurids fared better, with a few groups known from the early Cretaceous of South America. The most abundant and widespread Cretaceous diplodocoids were the rebbachisaurids, well known from the early Cretaceous of Europe, northern Africa and South America. Unfortunately, rebbachisaurid fossils are often incomplete, and the group remains mysterious.

Macronarian sauropods were highly diverse during the Cretaceous. The titanosaurs, the largest macronarian group, were particularly successful. They account for about one-third of total known sauropod diversity and were very varied: they are known from every continent and counted among their number both the biggest sauropods (for example, Argentinosaurus, around 70 tonnes) and the smallest (Magyarosaurus, 750 kilograms). Sauropods declined through the late Cretaceous, with non-titanosaur lineages becoming extinct well before the end of the period, potentially due to competition with titanosaurs. Titanosaurs continued to diversify, however, and remained common until the end of the period, when they were wiped out by the extinction event that killed all the dinosaurs.
Anatomy

All sauropods walked slowly on all four legs. The hind feet were broad and the toes short; the interior three bore flattened, banana-shaped claws. Fossil tracks show that the heel of the hind foot rested on a fleshy pad, helping to distribute the weight of the animal. The hands of sauropods were more unusual, with the metacarpal hand bones arranged vertically, forming a column-like structure which was semi-circular in cross-section. The hand lacked any kind of rear fleshy pad, and so left horseshoe-shaped prints. The fingers of most sauropods were highly reduced and only the thumb bore a claw. This may have been used as a weapon and for gripping slippery surfaces, tree trunks or partners during mating. The hands of later titanosaurs lost the digits entirely, with the animal instead walking on the stump-like ends of the metacarpal bones.

In order to support the animals’ bulk, the limb bones of sauropods were massively built, and held in a columnar posture. Titanosaurs held their limbs outwards slightly in a ‘wide-gauge’ stance that may have made them more stable. Sauropods would not have been able to run; estimates of speed from fossil tracks and computer models indicate that they would have moved at a ponderous 5–7 kilometres per hour, similar to human walking speed. Although they would have moved on all fours, sauropods had relatively long hind limbs and robust hips so some may have been capable of rearing up, balancing on the back legs and tail. In diplodocids, the animal’s centre of mass was placed far back, close to the hips. This, coupled with their relatively long hind legs, suggests that they may have been particularly good at rearing. By contrast, sauropods with larger forelimbs, such as Brachiosaurus, were less suited to rearing and it seems unlikely that they spent much, if any, time up on their back legs.

In most sauropods, the body was deep but relatively narrow, but others, such as Camarasaurus, had wider ribcages. Titanosaurs such as Opisthocoelicaudia were particularly fat, with very broad hips. Although no direct evidence of the internal organs is preserved, the deep body cavity of sauropods provided plenty of room for capacious guts, important for large herbivores.

Sauropods’ vertebrae were lightweight, with hollows and air cavities. In early sauropods, these features are limited to depressions in the vertebrae of the neck and thorax. Through evolution, they became increasingly elaborate and spread farther down the spinal column, reaching as far as the base of the tail in diplodocoids and titanosaurs, and even the pelvis in the rebbachisaurid Tataouinea. Pneumatic features like these are also seen in the bones of theropods, including modern birds, where they result from invasion of the skeleton by air sacs. These air sacs are part of the respiratory system, where they act as bellows pushing air in and out of the lungs, which extract oxygen during both inhalation and expiration. The presence of similar pneumatic features in sauropods indicate that they too would have sported air sacs and similar lungs. The large volume of the air sacs would have been important in temperature regulation and oxygen extraction, making breathing easier than it would be down a long, narrow windpipe with a mammalian-style lung.

This elaborate respiratory system was therefore essential in the development of the most famous sauropod feature: the long neck. Extremely long necks (more than 8 metres) evolved multiple times independently. The neck of the diplodocoid Supersaurus is estimated to have reached 15 metres. The 12-metre neck of Mamenchisaurus was as long as the animal’s body and tail combined, and the macronarian Erketu may have had a neck more than twice the length of its body. Different sauropods evolved very long necks in different ways, either by elongating individual neck bones, ‘creating’ new vertebrae, co-opting vertebrae from the back, or a combination of these methods.

Not all sauropods had long necks — the necks of dicraeosaurids and some titanosaurs were relatively short. The neck of the dicraeosaurid Brachytrachelopan was less than 2 metres, granting the animal the dubious honour of being the shortest-necked sauropod. The necks of dicraeosaurids were also remarkable for very tall, forked spines on the vertebrae — in Amargasaurus, these were elongated to form a double row of long spikes that may have been used for display and possibly gave the neck some protection.

The posture of sauropod necks is controversial. Establishing neck flexibility is complicated in that important soft tissues such as cartilage are not preserved, and various lines of evidence have been used to argue that sauropods held their necks either vertically, at an angle or horizontally. The great variability in neck architecture reflects the variety of postures that would have been adopted by different sauropods, and a single species would have used different postures for different behaviours. Some, such as Brachiosaurus and Euhelopus, have necks that seem to be well adapted to being held vertically, but were stiff. By contrast, features of the necks of diplodocoids suggest that they were held closer to the horizontal for much of the time, but had greater lateral (side to side) flexibility. The necks of other sauropods were held in a range of inclinations between these extremes.

The neck was counterbalanced by the tail, which was relatively long in all sauropods, particularly in diplodocids, where it could account for around half of the total length of the animal. Tracks show that the tail was held clear of the ground. The heavy tails of most sauropods would probably have served as competent weapons; some early sauropods such as Shunosaurus bore a small club at the tip to increase its offensive presence. The tails of dicraeosaurids and diplodocids bear a distinctive ‘whiplash’ tip formed of many small, rod-like bones. It has been suggested that they used these tails for defence, cracking them like a bullwhip. However, the small terminal bones would have been too fragile to sustain impacts and it seems more likely that the animals would have used the heavier base of the tail or their thumb claws to repel predators.

Perched on the end of the neck was the relatively small head, less than 2% of the total length of the body in most sauropods. The skulls of sauropods are generally lightly built, with large openings. They had large bony nostrils set well back on the snout: the fleshy nostrils would have been much smaller, and close to the end of the snout. The large internal airways would have been important in heat transfer, keeping the brain cool. Sauropod brains were very small compared to their titanic bodies — the brain of a 2.5-tonne Ampelosaurus was about the size of a walnut. The well-developed olfactory tracts seen in many sauropod brains suggest a good sense of smell.

Most sauropods had a blunt snout, with spoon-shaped teeth that interlocked when closed (Fig. 5a). Brachiosaurids bore a distinctive muzzle, and a crest formed out of the arched nasal bones. Diplodocoids and titanosaurs, although only distantly related, had similar skulls, with a long snout, very strongly retracted bony nostrils and pencil-shaped teeth (Fig. 5b–c). The teeth of rebbachisaurids, dicraeosaurids and titanosaurs met in a shearing bite, whereas those of diplodocids no longer touched — instead, they gripped food and detached it from the plant with tugging or raking motions. Even odder was the extremely lightweight head of the rebbachisaurid Nigersaurus, which was unlike that seen in any other vertebrate (Fig. 5d).

Sauropod skin impressions are rare. Those that exist show that the hide was composed of small, non-overlapping scales. Diplodocids had a row of dermal spines down their back. The hides of some titanosaurs were lightly studded with bony plates called osteoderms, potentially providing some defence against predators. However, in many cases there seem to have been only a few, sparse osteoderms, which were delicate, making for a poor suit of armour. Instead, the osteoderms may have served as mineral stores.
Ecology

All sauropods were herbivores. They would have processed the food very little, if at all, in their mouths, and they lacked cheeks, allowing a wider gape. Rather than chewing their food thoroughly, they gulped down as much plant matter as they could, as quickly as possible.

Lacking the heavy batteries of chewing teeth seen in many other herbivores allowed the sauropod head to remain light enough to be perched on top of an elongate neck. The necks of sauropods were more like that of a goose than that of a giraffe, and they would have let the animal reach lots of food in both the vertical and the horizontal directions without having to move its heavy body. In this way, the sauropod head and neck would have allowed the animal to feed very quickly, providing the prodigious intake required to support their titanic bulk. Sauropods’ high tooth-replacement rates were required because the teeth would have been worn away by this vast diet; the exceptional rates in diplodocoids and titanosaurs may have been a response to increased grit in the diet from browsing on vegetation low to the ground.

Contrary to earlier belief, sauropods do not seem to have ground up their food using gastroliths (‘stomach stones’). The polished stones occasionally found in sauropod ribcages are not like the abraded stones known from ostrich gizzards, and there are very few of them; they were probably just swallowed by accident. Rather than any kind of mechanical processing, sauropods processed the food by fermenting it in their capacious digestive tracts.

The variability of the sauropod feeding apparatus reflects various feeding behaviours (Fig. 5), and differences in neck length and rearing ability meant that sauropod species would have exploited different feeding heights, ranging from high-canopy browsing to low ‘grazing’. Analyses of biomechanics and tooth wear have demonstrated different diets and foraging strategies in species that lived in the same time and place, helping to explain how many sauropod communities could consist of a lot of different species.

Sauropods are known from every continent and a wide range of habitats, from open fern prairies to riverside woodland. Most are associated with coastal plain deposits, whereas titanosaurs seem to have preferred more inland environments. Cretaceous sauropods tend to be recovered from desert or semi-desert settings near the equator; at higher latitudes, the most common herbivores were from another group called ornithischian dinosaurs. Sauropods are unknown from latitudes of over 66◦ in either hemisphere, conditions this far north and south seem to have been too cold for them to thrive. The large size of sauropods would have given them advantages such as greater efficiency of travel and increased resistance to fasting and drought, potentially making them better at dealing with seasonal dryness than other dinosaurs. There is some indication that Cretaceous rebbachisaurids were more tolerant of extremely dry environments than macronarians living at the same time.

Fossil tracks indicate that sauropods lived and travelled in herds. Their voracious appetite means that they would have rapidly denuded an area of vegetation, and so would have been continually on the move. Oxygen-isotope signatures from Camarasaurus teeth demonstrate that these sauropods migrated, presumably in search of food or water.
Sauropod life history

The largest sauropod eggs were roughly the size of a football; quite small for such large animals. Mothers laid multiple clutches of 5–50 eggs in a bowl- or kidney-shaped nest dug by the clawed hind feet. The eggs were then buried in sediment or covered in mounds of rotting vegetation. A nesting site from Patagonia, preserving thousands of eggs, demonstrates that some sauropods nested in large colonies and returned to the same site year after year. Some species seem to have favoured geothermal regions for nesting, exploiting this warmth to incubate their eggs.

Sauropods were not good parents. Adults do not appear to have remained at the nest to guard the eggs or chicks. Although fossil tracks indicate that sauropods habitually moved in herds, these generally seem to have been age-segregated. The large size range that sauropods grew through, and changes to the skulls and teeth seen in some species, indicate that different sauropod life stages would have occupied different ecological niches. Indeed, there is some evidence that juvenile and adult diplodocoids lived in different geographical areas. Although associations of differently aged sauropod individuals do exist, these are comparatively rare and there is no reason to assume that the adults showed parental behaviours.

Sauropod babies were instead capable of fending for themselves straight after hatching, proceeding to grow fast and (mostly) die young. Death rates in juvenile sauropods were high, with only a small number making it to adulthood. Studying the texture and growth rings of sauropod bones indicates that juveniles might have put on 500–2,000 kilograms per year during the fastest growth phase. A 5-kilogram sauropod hatchling would have reached sexual maturity in its teens and full body size in its late twenties. This elevated growth rate is closer to those of mammals and birds than reptiles, providing evidence of a warm-blooded metabolism for young sauropods. This would have been problematic at very large sizes: the surface-area-to-volume ratio decreases with increasing size, so large terrestrial animals struggle to keep cool. Growth slowed as sauropods reached maximum body size and it has been argued that this accompanied a shift to lower metabolic rates. As adults, even with cold-blooded metabolisms, the colossal sauropods would never have cooled and may have struggled to shed excess heat. Their large nostrils and expansive respiratory systems would have helped them to remain cool, and the long neck and tail would have provided large surfaces from which to radiate heat.

The high death rate of juveniles means that adult sauropods were relatively rare. Sauropods reached sexual maturity long before attaining full size, and at any one time a population would feature a large number of juveniles suffering rapid turnover, with a few adults. For the lucky few that reached adulthood, the large size of most species would have made them virtually invincible to predators. Establishing the maximum lifespan of a dinosaur is all but impossible, but remains from individuals of at least 38 years old are known, and it is likely that sauropods could have lived for considerably longer than that.
Sauropod body-size evolution

The earliest sauropods were already very large, weighing in at around 5 tonnes. Sauropods would continue to push the envelope of body size for land animals for the remainder of their evolutionary history. The shift towards large body size, which occurred relatively early in sauropodomorph history, probably began around the same time as they began to eat plants, feeding on high-fibre parts such as leaves and stems. Large body size has several advantages for herbivores, including greater gut capacity and possibly digestive efficiency, greater defence against predators, greater resistance to drought and famine, and more efficient movement over long distances. This diet interacted with the unique combination of morphological and life history characters discussed above in evolutionary feedback loops to allow the evolution of gigantic body sizes (Fig. 6).

Sauropods are iconic creatures of the fossil record for good reason. They expand markedly upon the picture of body forms and, particularly, size, that we would get from considering living animals alone. They highlight the importance of the fossil record in providing a complete picture of the limits of life on Earth.
Further reading

Barrett, P. M. Paleobiology of herbivorous dinosaurs. Annual Review of Earth and Planetary Sciences 42, 207–230 (2014). DOI: 10.1146/annurev-earth-042711-105515

Bates, K. T. et al. Temporal and phylogenetic evolution of the sauropod body plan. Royal Society Open Science 3, 150636 (2016). DOI: 10.1098/rsos.150636

Benson, R. B. J. et al. Rates of dinosaur body mass evolution indicate 170 million years of sustained ecological innovation on the avian stem lineage. PLoS One, 12, e1001853(2014). DOI: 10.1371/journal.pbio.1001853

Button, D. J., Rayfield, E. J. & Barrett, P. M. Cranial biomechanics underpins high sauropod diversity in resource-poor environments. Proceedings of the Royal Society B: Biological Sciences 281, 20142114 (2014). DOI: 10.1098/rspb.2014.2114

Peyer, K., Allain, R. A reconstruction of Tazoudasaurus naimi (Dinosauria, Sauropoda) from the Early Jurassic of Morocco. Historical Biology 22, 134-141 (2010). DOI: 10.1080/08912960903562317

Sander, P. M. et al. Biology of the sauropod dinosaurs: the evolution of gigantism. Biological Reviews 86, 117–155 (2011). DOI: 10.1111/j.1469-185X.2010.00137.x

Sander, P. M. An evolutionary cascade model for sauropod gigantism — overview, update and tests. PLoS One 8, e78573 (2013). DOI: 10.1371/journal.pone.0078573

Sereno, P. C. et al. Structural extremes in a Cretaceous dinosaur. PLoS One 2, e1230 (2007). DOI: 10.1371/journal.pone.0001230

Whitlock, J. A. Inferences of Diplodocoid (Sauropoda: Dinosauria) feeding behaviour from snout shape and microwear analyses. PLoS One 6, e18304 (2011). DOI: 10.1371/journal.pone.0018304