Hi, my name is Gilles Cuny, and I am curator of vertebrate palaeontology at the Natural History Museum of Denmark. I would like to tell you about the origins of modern vertebrate faunas. 250 million years ago, the Permo-Triassic mass extinction killed up to 50% of the existing families and up to 96% of the species that lived at that time, inducing therefore a major reorganization of the ecosystems on Earth. Consequently the beginning of the Triassic was a phase of recovery until the ecosystems went back to their previous diversity level, but changing the actors. It took at least 10 million years for the global ecosystem to recover. In the terrestrial realm, the synapsids, or so-called 'mammal-like reptiles', which dominated the ecosystems, were replaced by archosaurs, a clade to which belong the dinosaurs and the crocodiles. The Permian synapsids, here illustrated by Moschops, had a semi-erect posture, meaning that their limbs are set on the side of their body, resulting in a rather slow mode of locomotion. During the Triassic, two lineages of archosaurs, the dinosaurs and the crurotarsans, developed for the first time a more efficient upright posture, but by different means. In crurotarsans, like for example Poposaurus from the Late Triassic of the United States it is the girdles that change orientation so that the femur and humerus are set just below the body. The limbs being now directly under the body, they are much more efficient at supporting the weight of the latter. The locomotion is also enhanced as it is no longer necessary to flex the body sideways when walking, as lizards for example need to do. Dinosaurs achieved an erect posture in a different way. The humerus and femur will develop an inturned head so that the shaft of the bone, and then the rest of the limb, is projected just below the body. Mammals will subsequently evolve an erect posture in exactly the same way, although independently. You can see here the femur of a modern dog. The head of the femur is inturned so that the shaft of the bone is projected directly under the body. Erect posture might have been necessary for endothermy to evolve. Endothermy is the capacity to maintain the body temperature constant independently of external conditions. Indeed, without an erect posture, the flexing sideways of the body during walking expands one lung whereas it compresses the other one. That makes the act of breathing less efficient, and thus reduces the capacity of an animal to produce enough energy to maintain its body temperature stable. The main difference between the Permian and the Triassic worlds was therefore the appearance of faster, more active animals. Mammals and dinosaurs appeared almost at the same time and the most primitive mammal, Adelobasileus, was found in the Carnian of North America and is about 230 million years old. It is known from a single incomplete skull, and it is therefore difficult to know exactly what it looked like, but it was probably quite similar to a modern shrew, its total length not exceeding 15 cm. Dinosaurs and mammals have therefore always been contemporary, but dinosaurs quickly reached supremacy over mammals. Why? One possible explanation is that the arid climate that predominated at the beginning of the Norian, 228 million years ago, gave the dinosaurs an advantage over the mammals. Dinosaurs, like crocodiles and birds, had a metabolism that did not require as much water as that of mammals. Their skin, contrary to that of mammals, was impermeable and they did not sweat. Also, mammals produce liquid urine, causing another source of water loss, whereas dinosaurs produced semi-liquid urine, forming a �paste� like in modern birds. In a Norian world where desert conditions were more common than today, dinosaurs would therefore have a serious evolutionary advantage over mammals. Once the domination of the dinosaurs was established, mammals were not able to rise to a dominant position until another mass extinction gave them such an opportunity. The first stem-mammals were small, mouse-sized animals so their fossil skeletons are quite rare. However, their small teeth, much more prone to fossilization because more mineralized than bones, are often recorded in many areas of the world during the Mesozoic, so that the history of these little furry animals is quite well known, although it is sometimes difficult to know what they looked like. The first representatives of the two major modern groups of mammals, the pouch-bearing marsupials, or Metatheria and the placentals without pouch, or Eutheria first appeared between the Middle Jurassic and the Early Cretaceous, some 65 million years after the appearance of Adelobasileus, the first stem-mammal. Both the first marsupial and the first placental have been found in China. As I said, because of their small size, fossil mammals are more often known from isolated teeth than from complete skeletons, but fortunately exceptions exist that allow us to understand the main steps of their evolution. Here is a reconstruction of the oldest placental known to date. Its name is Juramaia, and the anterior part of its skeleton has been found in the Middle Jurassic of China. It was a small animal, not exceeding 20 cm in total length, but it demonstrates that modern types of mammals appeared some 165 million years ago. However, it would be wrong to think that more primitive mammals, without living relatives, did not evolve during the age of dinosaurs. Again, some more or less complete fossils reveal the existence of surprisingly specialized stem-mammals. Fruitafossor for example is a fossorial stem-mammal, the lineage of which appeared well before the separation between the placental and marsupial lineages. A relatively complete skeleton of this little animal, the weight of which did not exceed 6 grams, was found in the Late Jurassic of North America. Its peg-like teeth devoid of enamel are very similar to those of a modern armadillo, indicating a specialized diet made of colonial insects. It possessed also large front limbs, well adapted for digging, so that it was able to break into termite mounds. Although primitive, Fruitafossor shows therefore a high degree of specialization, both for digging and in terms of diet, challenging the idea that early mammals were restricted to unspecialized niches. Another good example of a specialized stem-mammal is the gliding Volaticotherium from the Middle Jurassic of Asia. Its gliding membrane, or patagium, was supported by the limbs and the tail. It is the first known gliding mammal, although it is only distantly related to modern flying squirrel. The shape of its teeth indicates that it was a specialized insect eater. Castorocauda is a semi-aquatic stem-mammal with a broad tail similar to that of a modern beaver. It was first found in the Middle Jurassic of China and is therefore some 165 million years old. Fossilized impressions of some webbing are present between the toes, reinforcing the idea it was well-adapted to an aquatic mode of life. Its sharp teeth also suggest it was piscivorous, eating mainly fish. It is the very first mammal adapted to a semi-aquatic mode of life in the fossil record, and such a specialization will not reappear in mammals before the Eocene, 56 million years ago. The forelimbs of Castorocauda were robust and adapted for digging as well. They are quite similar to those of the modern Platypus, which both digs and swims. However, this is again a convergence as Castorocauda is a stem-mammal only distantly related to Platypus. Another interesting feature about this critter is its size. It could grow longer than 40 cm and it is actually the largest Jurassic mammal known to date. Repenomamus from the Early Cretaceous of China is about 125 million years old. It is thus younger than Juramaia, Fruitafossor, Volaticotherium or Castorocauda. It could exceed 1 m in length and its overall appearance is quite reminiscent to that of a modern Tasmanian devil. One fossil specimen displayed some remains of a juvenile of the small ornithischian dinosaur Psittacosaurus at the level of its stomach. Repenomamus was therefore an active predator, which may also have scavenged dinosaur carcasses. It is the largest known Mesozoic mammal, and it was indeed larger than some of the predatory theropod dinosaurs that could be encountered at the same time in China. So, the very first mammals evolved in the Late Triassic, at the same time as the dinosaurs, from one of the synapsid lineages that survived the end-Permian mass extinction. However, the rather arid climate at the time was more beneficial to dinosaurs than to mammals and the former quickly outcompeted mammals and reached dominating position in the Late Triassic ecosystems. We shall see, however, in the next video that it was not so easy for the dinosaurs. That does not mean, however, that mammals did not evolve during the age of dinosaurs. It is often said that Mesozoic mammals were all small nocturnal mouse-like animals. This is clearly an oversimplification. Even if mammals did not reach dominant positions in Mesozoic ecosystems before the demise of dinosaurs at the end of the Cretaceous, they nevertheless diversified during the Mesozoic, becoming gliding, digging or aquatic. Most of the actors of this first diversification event are not directly related to modern mammals though. The first representatives of the two main modern lineages, the marsupials and the placentals, appeared anyway as early as the Middle Jurassic, 165 million years ago. The Late Triassic was also the time when many other lineages that characterize our modern vertebrate faunas appeared. Among them, we found the lizards. The first representative of the lizard and snake lineages, the Lepidosauromorpha appeared in the Carnian, 225 million years ago, and belongs to the Sphenodontia, a lineage represented nowadays by a single survivor, the Tuatara, or Sphenodon, from New Zealand that you can see on this picture. The first sphenodontians were small animals, between 15 and 35 cm long. Their teeth suggest that some were insectivore, others herbivores. It took, however, nearly 100 million years before the most derived members of the Lepidosauromorpha, the snakes, appeared in the Early Cretaceous. Another modern clade to appear in the Late Triassic is the turtles, although as we shall see their origin is still quite mysterious. Turtles have the most unusual body plans of all amniotes, meaning the tetrapods that lay eggs on the ground. As we can see on this fossil from our collection, the body of a turtle is protected by a ventral plastron and a dorsal shell which is fused to the ribs. That means that the scapula, the bone making your girdle that you can feel here, which normally is outside the rib cage, is in the turtle inside the rib cage and this is really a major modification of the general organization of a vertebrate. These highly unusual tetrapods appeared for the first time in the Norian (215 million years ago). One of the oldest turtles, the emblematic Proganochelys, is known in Europe, North America and Thailand, indicating that these animals were very quickly successful in many parts of the world. Proganochelys was very similar to modern turtles, except that it could not retract its head under the carapace, and instead the neck was protected by pointed plates. It also possessed teeth, as well as dermal plates covering the extremity of its tail, making it a defensive weapon. It was a rather large animal, measuring about one metre in length. Later turtles will develop the capacity to retract their heads under the carapace when they are threatened by danger. The way they do it defines two groups today. The pleurodires do it by making a sideways bend in the neck, whereas the cryptodires make a vertical bend. It is always problematic to try to decipher the phylogenetic affinities of a group as specialized as the turtles. The characteristic that actually makes a turtle, that means the carapace, is unlikely to be present in the putative ancestor. The problem is then to understand how the carapace appears to identify the potential sister group of the turtles, and then to understand how they originate. Recent discoveries have revived the debate. A fossil found in the Norian of New Mexico, Chinlechelys shows an animal with a very thin dorsal shell not yet fused to the ribs. This animal could therefore be potentially very important to help us solve the relationships of the turtles with the other amniotes. It would nevertheless suggest that the dorsal part of the carapace was the first to appear and resulted from the fusion of small dermal bones covering the body of the putative ancestor as we can see on this picture. The covering of the neck and tail will subsequently disappear. Another fossil found in the Carnian of Guizhou Province (South China) is Odontochelys, and it offers a rather different story. The dorsal carapace is hardly developed while the ventral plastron is fully developed, which is exactly the contrary of the condition observed in Chinlechelys. Moreover Odontochelys is aquatic whereas Chinlechelys is fully terrestrial. So was the earliest turtle aquatic or terrestrial? Odontochelys is now widely accepted as the sister-group of turtles, and an aquatic phase in the early evolution of turtles appears therefore more likely. However, there are still quite a lot of discussions in the scientific community as where to nest the turtles: at the base of the diapsids lineage, a clade grouping the lizards, crocodiles and birds, or among the anapsids, close to the more primitive pareiasaurs like this Bradysaurus from the Permian? In the current state of our knowledge, it is still difficult to answer this question with certainty. Many questions remained to be solved before we can fully understand the evolution of life on our planet. We have looked in this first video at how some of the major groups of modern tetrapods: mammals, lizards and turtles appeared in the Late Triassic. Next time we'll have a look at the origin of the crocodiles as well as modern bony fishes and modern sharks.
