by Peter Ward
5. Late Cretaceous time. Dinosaur diversity skyrocketed. Most of this diversification came through large numbers of new ornithischians: ceratopsians, hadrosaurs, and ankylosaurs among others. Only a small number of sauropods were present.
No evolutionary history can ever be pinned on one factor. Dinosaur morphology changed from predator-prey interactions, competition among themselves and others of their world, perhaps even climate change driven in large part by the incredible rises and falls of sea level during the Jurassic and Cretaceous—at one point a sea level rise so large that North America became two separate minicontinents, separated by a large if shallow north-south-running sea. Nevertheless, oxygen levels must have played a part.
The time of the first dinosaur grouping, the late Triassic assemblage, was a time of low oxygen levels, and this coupled with very high carbon dioxide levels—not asteroid impact—was the major cause of the Triassic-Jurassic mass extinction. The combination of low oxygen and high global temperatures was the killing mechanism. Yet studies of the number of land vertebrate taxa before and after the T-J mass extinction clearly show that saurischian dinosaurs survived this mass extinction event better than any group of vertebrates, and one important reason may have been because of their superior respiration system, because of their air-sac lungs, which gave them competitive superiority over other terrestrial animals with different lungs.
Ornithischian dinosaurs, on the other hand, did not possess as effective a respiratory system as did saurischians. However, they were competitively superior to herbivorous saurischians with regard to food acquisition, larger heads, stronger jaws, and better teeth. With the rise of oxygen to near present-day levels in the Cretaceous, ornithischians became the principal herbivores because of this superiority, leading to the extinction of many saurischian herbivores through competitive exclusion.
While the Jurassic to Cretaceous interval marked a relatively rapid and significant rise in atmospheric oxygen, other events were taking place. One of these was the breakup of the once-global continent of Pangaea into smaller continents. Another, and perhaps more significant for the distribution and taxonomic makeup of the later Mesozoic dinosaur faunas, was the radical change in flora. Dinosaurs evolved in a gymnosperm-dominated world—with conifers, but ferns, cycads, and ginkgos as well. But in the early part of the Cretaceous a new kind of plant appeared, a flowering plant.
With this new kind of reproduction and other adaptations, these plants, the angiosperms, underwent a rapid adaptive radiation. They outcompeted the earlier flora nearly everywhere on Earth, to the extent that by the end of the Cretaceous, some 65 million years ago, the angiosperms made up as much as 90 percent of vegetation. This transition in available food types would have affected the herbivores, and the kind of herbivores available as food would have directly affected carnivore body plans. Killing a late Jurassic sauropod would have been very different from killing a late Cretaceous hadrosaur.
Herbivory is dependent on the correct kind of teeth for the available plants. The sauropods may have lived on pine needles, their huge barrel bodies being, essentially, giant fermenting tanks for digestion of a relatively indigestible food source. The appearance of broad-leafed plants, the angiosperms, would have required different teeth and biting surfaces than those optimal for slicing pine needles off trees. Thus the transition from the sauropod-dominated faunas of the Jurassic to the ornithischian-dominated faunas of the Cretaceous was surely related in some part to the change in plant life. But respiration may have played a part as well, and perhaps if oxygen had not risen above 15 percent, the ornithischian takeover would not have taken place.
JURASSIC-TRIASSIC DINOSAUR LUNGS AND THE EVOLUTION OF BIRDS
Here it is proposed that the first dinosaurs were of a kind of animal never before seen or alive today: through upright posture and an evolving air-sac system they developed respiratory efficiency (the amount of oxygen extracted from air per unit time, or per unit energy expended in breathing) superior to any other then-extant animal. But these early forms may have lost endothermy, replacing it with a more passive homeothermy. That was their trick, using homeothermy to reduce oxygen consumption while at rest, and a superior lung system to allow extended movement without going into rapid anaerobic (and thus poisonous) states when active. We know that birds, a group of dinosaurs first appearing in the Jurassic, eventually had both endothermy and a very different kind of lung than in any extant reptile.
THE AVIAN DINOSAURS—AND DINOSAURIAN AVIANS
With perhaps the exception of the always-fascinating tyrannosaurids, no group of dinosaurs has received more attention in recent times than the basal birds. Vigorous debate centers on their body covering, but most important, on when flight first evolved, and why.
The first birds appeared about 150 million years ago, and the famous first bird remains Archaeopteryx. That is just before the start of the Jurassic. Oxygen had been rising for 50 million years at that time. Gigantism in dinosaurs was common. The immediate ancestors of the birds were fast, ground-running dinosaurs that may have used their forelimbs for a type of predation, a motion that was preadapted for a wing stroke in a flier, according to Berkeley paleontologist Kevin Padian. The fossil record suggests that the ancestors of the first bird were the bipedal carnivorous saurischians known as troodontids or perhaps the dromaeosaurids, forms that appear to have been already feathered.
Could Archaeopteryx fly? Most specialists now think so. But there is debate about when true flight took place. Could the late Jurassic “birds” really fly, at a time when their competition in the air would have been the diverse and successful pterodactyls? The fossil record does show that by the lower Cretaceous there is a bird fossil (Eoluolavis) that had evolved a “thumb wing,” an adaptation that allows flight at slower speeds with greater maneuverability. Thus, within a few million years after Archaeopteryx, fairly advanced flight was present. New discoveries from China have revealed an unexpectedly high diversity of birds already in place by the early part of the Cretaceous. Flight was an adaptation that stimulated a rapid evolution of new forms. We will return to the evolutionary history of birds in a subsequent chapter, as much of their story happened after the Jurassic.
Flight in birds is highly energetic. They use a great deal of energy to fly, and that, added to their relatively small size and endothermy, makes them great users of oxygen. So their air-sac system serves them well.
DINOSAUR REPRODUCTION AND OXYGEN LEVELS
One of the great discoveries of twentieth-century paleontology was the finding of dinosaur eggs.6 In the latter half of that century, complex patterns found in associated fossil eggs hinted at behavioral complexity in dinosaur reproduction, or at least the egg-laying parts of it. In this century, ready access to new machines—small desktop CT scanners—has led to a third revolution in understanding dinosaur reproduction. Now eggs can be examined without mechanical damage to reveal delicate embryos within, and there is increasing understanding not only of the growth of these embryos, but of the construction of the eggs themselves—the how and why of a dinosaur egg.
Birds show little variation in at least one aspect of their reproduction. Extant birds, our best window to the dinosaurs, all lay eggs with a porous, calcareous shell. There is no live birth in birds, in contrast to extant reptiles, which have many lineages using live birth. There is also great variation in egg morphology between birds and some reptiles. While the eggshell in birds and reptiles consists of two layers, an inner organic membrane overlain by an outer crystalline layer, that amount of crystalline material varies greatly, from a thick calcium carbonate layer like that in birds to almost no crystalline material at all, so that the outer layer is a leathery and flexible membrane. Even the mineralogy of the crystalline layer varies, from calcite in birds, crocodiles, and lizards, to aragonite (a different crystal form of calcium carbonate) in turtles. Eggs are thus divided into two main types: hard or crystalline, and soft or parchment eggs. Some workers further subdivide the parchment eggs into flexib
le (used by some turtles and lizards) and soft (used by most snakes and lizards). Not surprisingly, the fossilization potential of these different hardness categories of eggs differs markedly. There are numerous fossil hard eggs known (many from dinosaurs), a few flexible eggs, and no undisputed soft eggs preserved.
Because of the great interest in dinosaurs there has been much speculation about their reproductive habits (the thought of two gigantic Seismosaurus mating rather boggles the imagination), and there are still many mysteries. One of the seminal discoveries about dinosaurs was that they laid large, calcareous eggs with calcite crystals making up the mineral layer, a find from the first expedition to the Gobi desert by the American Museum of Natural History expedition of the 1920s. Since then thousands of Cretaceous dinosaur eggs have been found, and the nesting patterns discovered and publicized by Jack Horner in Montana have opened a window into dinosaur reproduction. But are these Cretaceous finds characteristic of dinosaurs as a whole? This question remains unresolved and controversial. While most workers assume that all dinosaurs laid hard-shelled eggs, this is far from proven, and as we shall see below, there is indirect evidence that some early dinosaurs may have utilized parchment eggs or even live birth.
Almost all dinosaur eggs come from the Cretaceous. There is great variability in the nature of their crystal form and size, number, and the pattern of pores in them. But it is not the variety that is the most interesting scientific question. There are far fewer Jurassic dinosaur eggs, and almost none known from the Triassic. Why would that be? Has it to do with different kinds of preservation characteristics of the Cretaceous world and how things on land fossilized or did not? Or was it because of the far lower oxygen levels of the Triassic and Jurassic compared to the Cretaceous (and especially the late Cretaceous, where the vast majority of dinosaur eggs are recovered)?
There are several possibilities for this: perhaps there is indeed some preservation bias, with pre-Cretaceous eggs as common as those of the Cretaceous, but the lesser extent of Triassic and Jurassic dinosaur beds compared to the vast expanse of Cretaceous-aged beds has caused this difference. Thus the difference simply comes from sample size. Another (and different) possibility is that pre-Cretaceous eggs fossilized much less readily than those from the Cretaceous. This would certainly be the case if pre-Cretaceous eggs were leathery like those of extant reptiles, rather than calcified like birds. And if, like the marine ichthyosaurs, some dinosaurs utilized live birth rather than egg laying, there would certainly be fewer eggs to find. As in so many other aspects of the history of life, the level of atmospheric oxygen may have played a major role in dictating mode of reproduction.
Fossil eggs from Cretaceous deposits attributed to dinosaurs have a calcium carbonate covering like a chicken egg (but thicker), but unlike chicken eggs, which are smooth, the dinosaur eggs were usually ornamented with either longitudinal ridges or nodular ornamentation. Presumably ornamentation allowed the eggs to be buried after emerging from the female, with the ornament allowing airflow between the eggs and the burial material. The ability to bury eggs may have aided their fossil preservation potential, and perhaps helps explain why there are so many Cretaceous eggs, and so few other kinds. The heavy calcification would also help the eggs withstand the overpressure of burial in soil or sand.
Also now known is the complex behavior in nest making and orienting the eggs in burial mounds in the late Cretaceous, but not before. The Late Cretaceous bipedal dinosaur Troödon arranged its eggs tightly together in pairs, or vertically, while Jack Horner has shown complex burial patterns in late Cretaceous hadrosaurs from Montana.
The advantages of calcareous eggs is that they are strong, harder for predators to break into, and also aid in development: as the embryo grows inside the egg, some of the calcium carbonate is dissolved from the eggshell itself to be used in bone growth. They also might shield the egg from bacterial infections. But this comes at a price. Calcium carbonate, even an eggshell-thin layer of it, will not allow the passage of air or water into or out of the shell. But developing embryos need both water and oxygen. All calcareous eggs thus have pores so that oxygen-laden air can enter, but not so many pores that water quickly leaves by desiccation. To ensure sufficient water, the interior of the egg has a large amount of a compound known as albumin (familiar to us as the “white” of a chicken egg), which provides water to the embryo. This kind of egg is found in all birds and crocodiles.
The second kind of reptilian egg, the parchment egg, is found in turtles and most lizards. This kind of egg can take up water and actually expand in size with water uptake. But water permeability is a two-way street: parchment eggs can easily lose water as well. Burying these kinds of eggs in nests, the habit of many turtles and alligator-crocodiles, reduces water loss as well as hiding them from predators.
Burying an egg presents some danger: all developing embryos require oxygen, and thus the embryo requires an egg that can allow the passage of oxygen from the atmosphere into the egg. If the egg is buried too deeply or in impermeable sediment, the embryo will suffocate. And if the egg is laid at high altitude, it runs the risk of the same fate, even being smothered by parental care. So far, biologists have concentrated on temperature as the major variable affecting development rates in reptiles and birds. But the clues given by high-altitude lizards suggest that oxygen levels certainly play a part as well.
Modern lizards living at altitude often show viviparity, or live birth. They also hold the eggs in the birth canal for long periods of time. In both cases the explanation has been that they do this to maintain relatively high temperatures in an environment where there can be very cold temperatures that could slow development. But both of these adaptations would lessen or completely remove the time that the embryo is enclosed in a capsule that itself reduces the rate of oxygen acquisition. Calcareous eggs cannot be held in the mother because they do not allow oxygen to enter the egg until they emerge from the mother.
So we have a mystery. Reptiles show four different kinds of reproduction: live birth, parchment eggs that are held within the mother for extended periods of time, parchment eggs that are laid soon after formation within the mother, and calcareous eggs. And following birth there is also a series of possibilities: the eggs are buried or not, and when not buried, the eggs can be cared for by the parent or not. The advantages of each of these and the time they first appeared are still unknown.
And we have a second mystery. Most known dinosaur eggs are of Cretaceous, and as noted above, mainly late Cretaceous eggs, and are calcified; also, the appearance of burial behavior in dinosaurs is also characteristic of the Late Cretaceous. But what of the pre-Cretaceous dinosaurs? While there are eggs from sauropods and bipedal saurischians from the Late Jurassic—most spectacularly from deposits in Portugal, where the eggs contain the bones of embryos—earlier rocks are nearly barren of dinosaur eggs and/or nests. Only a few confirmed eggs are known from the Triassic.
Just when these various kinds of eggs first evolved thus remains a mystery. In 2005 it was proposed that calcareous eggs first appeared at the end of the Permian as an adaptation to avoid desiccation in the increasingly dry late Permian through Triassic global climates. Unfortunately, there is no fossil evidence to support this: there are no accepted Permian eggs in spite of the existence at that time of anapsids (the group that would give rise to turtles), diapsids (the group that would give rise to crocodiles and dinosaurs), and synapsids (the group that would give rise to us). Furthermore, only a small number of late Triassic eggs are known that may be from dinosaurs. But this poses a great dilemma: dinosaur eggs preserved commonly in Cretaceous sediments are not found in the same kinds of depositional environments of Permian and Triassic-aged strata. In all likelihood, if archosaurs had used hard eggs in the Permian or Triassic—if any group of reptiles had used hard eggs then—we would have already found them.
The absence of evidence is always a dangerous tool, but eventually numbers must be accepted. All evidence suggests that hard eggs were not
commonly produced by pre-Cretaceous terrestrial egg-laying organisms. Even the 2012 discovery of hard dinosaur eggs in South Africa became but a single exception to the rule. It is hard to see how future collecting, no matter how intense, can now overcome this trend.
Only two shapes of dinosaur eggs are known: rounded and elongate. But seven different patterns of crystal arrangements making up the eggshells are now recognized. This diversity of egg wall morphology would be surprising if all dinosaurs had evolved from a single egg-laying ancestor. But it would be what we would predict if hard-shelled egg laying evolved numerous times by separate lineages of dinosaurs. If we add the additional (and different) eggshell morphologies found in extant reptiles and birds, there are a combined twelve separate eggshell microstructures that have evolved during the now-long history of reptile, nonavian dinosaurs, and the real avian dinosaurs: birds.
Perhaps each of these is an adaptation to a different kind of stress an egg belonging to a particular group or species is normally subjected to: a turtle egg in a deep burrow, for instance, has a very different series of challenges facing it than does a robin egg in a nest high in a tree. But another possibility is that the various calcareous eggs are evidence of an independent evolutionary history, in which hard eggs separately evolved in multiple lineages, including dinosaur lineages.
THE “IDEAL” OXYGEN LEVEL
One of the most interesting of new discoveries about the evolution of many of the modern stocks of land animals is that so many came from a fairly narrow time interval—during a time in the late Paleozoic when oxygen was higher than now. This is true for many groups of extant vertebrates, including the first members of families that would go on to be lizards, turtles, crocodiles, and mammals. But it is not just land vertebrates that suggest this trend: many of the terrestrial invertebrates, including the basal stocks of many insects, arachnids, and land snails, also began during the Carboniferous period of more than 300 million years ago. New experiments of the last five years are indicating that there is a “magic” level of oxygen for the fastest rate of embryonic development within both land vertebrate eggs and insect eggs, and that number is 27 percent.
