There is a question that has plagued palaeontologists interested in dinosauria for as long as the animals have been studied; were dinosaurs endotherms (warm-blooded) or ectotherms (cold-blooded)?
Today, endothermy is seen in mammals and birds but not in reptiles and is actually little to do with the temperature of the blood. An endotherm generates heat, climate-controlled internally to produce perfect conditions for the biochemistry inside the animal. This is because most biochemical reactions are catalysed, usually by enzymes, and those enzymes operate most efficiently in a narrow temperature range. Unfortunately, this level of control uses an enormous amount of energy so endotherms need to eat a great deal more than ectotherms.
Terrestrial ectotherms use the warmth from the sun to heat their bodies to optimum temperature and use shade, water or other means to lose heat should that become necessary. Consequently, ectotherms are not 'all-weather' animals and are only active in a fairly narrow range of conditions. The huge advantage of ectothermy is that it requires vastly less energy and therefore vastly less food.
This entry is not intended to answer the question, merely to describe the evidence for both sides.
Overview of the Question
Dinosaurs clearly evolved from reptiles - there is little debate about that. In addition, it is known that mammals and birds also evolved, directly or indirectly, from the same group. It is believed, in fact, that the mammals evolved from a group known as Therapsids and the birds and dinosaurs evolved from a group called Thecodontids. In either event it is known that warm-blood evolved in reptiles at least twice. The question is whether it could have done so a third time. Moreover, birds are essentially defined by their feathers and those, all agree, evolved as insulation, something only required in a warm-blooded animal. If we are therefore looking for a warm-blooded animal that could have developed feathers and evolved into birds, Theropod1 dinosaurs are a major candidate. In this instance we are back to two appearances of warm-blood, which suits Occam's razor (the principle that the simplest explanation is usually the correct explanation), but only if there is evidence for this line of descent.
If you had asked this question a century ago there would have been no doubt at all. At that time dinosaurs were believed to be cold-blooded. The reason for this stems from an historical viewpoint more than from evidence. When the first dinosaurs were classified, they were described as giant extinct reptiles, due to their bone structure. Dinosaur skeletons are very reptilian in form, having tails not unlike a lizard, for example. The word 'dinosaur' means 'terrible lizard' and there seems little doubt that when Owen2 coined the term he believed that that is exactly what they were. Socially, indeed, it was eminently suitable to categorise them as such. Darwin had published his opus The Origin of the Species in 1859 and Owen had been vehemently opposed to its precepts. Owen is, somewhat unfairly, remembered more for his fruitless struggle against the theory of evolution than for his outstanding work in the field of palaeontology. Clearly he saw no social gain in any perception of the dinosaurs as lizards; some of his contemporaries, however, did.
Owen, bolstered by the Church, encouraged the thought that the dinosaurs represented the pinnacle of reptilian evolution. This could demonstrate that evolution did not work since modern reptiles showed degeneration, rather than progress. In fact evolution had no inherent requirement of progress, merely of change, and change had certainly occurred. Some of the proponents of evolution made increasingly wild cases for dinosaurs being less impressive than the fossil evidence seemed to show. They could be seen as lumbering, archaic giants, slow witted and slow moving. Fortunately for science, the evolution supporters won the arguments and Darwin's work became an important part of biology. Sadly, this did no favours for the dinosaurs.
Now that evolution was accepted, it was seen as 'inevitable' that dinosaurs were primitive and poor animals. Attacks came from all quarters; an effort was made to redesign dinosaurs so that they sprawled like lizards. Sauropods were described as being too heavy to even support their own weight; the braincase of the Stegosaurus was picked on as an archetype of stupidity. A century ago this was clearly the perception. Not only did this support the more ignorant among the evolutionists, it also found eventual favour with the Church, adept at changing sides mid-argument. Possibly their very weakness was what caused a kind God to put them out of their misery and permit their extinction, the argument ran. Throughout this entire debate, however, the actual evidence of the fossils themselves had been largely neglected.
Evidence for Ectothermy
Approaching the subject from a new position, unaffected by the beliefs of those earlier palaeontologists, we must look for evidence for cold-blood.
Saurian Skeleton Structure - Dinosaurs had fairly lizard-like skeletons; lizards are cold-blooded.
Descended from Reptiles - Reptiles are cold-blooded, so dinosaurs are likely to be cold-blooded as well.
Gigantothermy - The huge size of the animals is explained as being necessary in order to maintain constant activity by reducing heat loss. The volume of an animal increases much more than its surface area when it is made larger so heat loss would be reduced.
Evidence for Endothermy
Size - In exact opposition to gigantothermy it can be demonstrated that size is unhelpful for cold-blooded animals. In order to move a massive body for a prolonged period of time great strength is required. This can be seen in many modern terrestrial mammals but not in any cold-blooded animals. It is possible to demonstrate that cold-blood is only possible up to a certain size in purely terrestrial animals, something the size of the Diplodocus3 is clearly past that limit.
Bone Structure - It is possible to note significant differences in microscopic bone structure between warm-blooded and cold-blooded animals. Dinosaur bone clearly shows a warm-blooded structure. This appears much stronger evidence than the mere overall design of the bone and effectively refutes the structure argument for ectothermy.
Predator/Prey Ratios - If animals are warm-blooded then there is a large amount of prey biomass required per kilogram of predator biomass in order to support a stable population. If they are cold-blooded then the amount required is far less. Fossil discoveries point to a ratio of about 50 tonnes of prey biomass per tonne of predator biomass. This is what we would expect of warm-blooded animals.
Dynamic Body Structure
Stance - Dinosaurs stood erect, not sprawled like lizards. This stance is seen in warm-blooded animals only.
Form - Some dinosaurs have forms extremely reminiscent of fast ground birds, in fact Dromiceiomimus4 was built to a faster design than an ostrich. If we make the reasonable assumption that the form fulfilled the same function in both animals then Dromiceiomimus must have been capable of speeds approaching 60 miles per hour. Replace the warm-blooded animal with a cold-blooded one and those long, powerful legs would have propelled the animal at a less impressive four miles per hour. Moreover it would have had to sit down to rest every hundred yards or so! Obviously the animal must have been warm-blooded.
Insulation - It is now clear that at least some dinosaurs, and some pterosaurs, were insulated. Insulation is a hazard in cold-blooded animals but a benefit in warm-blooded animals, mammals use fur or blubber, birds use feathers. Clearly some dinosaurs were warm-blooded. Moreover, this cancels out another cold-blooded argument. If some dinosaurs had feathers then they could have been ancestors of the birds. It is not possible to argue that dinosaurs were cold-blooded because they were descended from reptiles since it could be equally true to say that dinosaurs were warm-blooded because they were the ancestors of the birds.
Growth Rate - We have excellent ranges of skeletons for some species, showing the development of the animals, and the growth rate appears to be extraordinary; from hatchling to adult Hadrosaurus5 in five years. This rate is confirmed by the examination of growth rings in bone. Such speed is only possible for warm-blooded animals.
Infant Differentiation - In cold-blooded animals infants are essentially identical to adults, merely smaller. In warm-blooded animals there is significant differentiation in many cases, this is also observed in dinosaurs.
Chest Capacity - Warm-blooded animals require huge lungs and powerful hearts. Cold-blooded animals do not and so have much smaller chest capacities relative to the size of the animal. Dinosaurs had clearly large chests for their mass.
Geographical Range - Dinosaurs lived in some arctic conditions where sunlight was hard to come by and the air was freezing. Among terrestrial animals only those with internal heating could have adapted to such an environment.
Species with Nocturnal Adaptation - Some dinosaurs have features that appear to assist with nocturnal activity. Such animals must be warm-blooded as the sun is required for cold-blooded animals to remain active.
Success over Thecodontids - A reason must be found for the startling success that the dinosaurs achieved over the Thecodontids at the end of the Triassic period. They were clearly the more adaptable group. While this does not posit the evolution of warm-blood, it is certainly a possible advantage they may have had.
Eating Behaviour - The Sauropods had enormous guts, multiple stomachs and used gastroliths6 to assist in the digestion of food. Such systems are needed if food is to be processed quickly and in bulk, a prerequisite of warm-blood.
Due to this inequity between the evidence, almost all contemporary palaeontologists support the warm-blooded position, outlined first to a popular readership by Adrian J Desmond in The Hotblooded Dinosaurs and based principally on the sea of change caused by the discovery of Deinonychus (a fast and aggressive Dromaeosaurid dinosaur)7 by John Ostrom.
On the Internet
Dinosauria Online is filled with information and excellent articles concerning dinosaur research.